Citation
Psychophysiological and reaction time responses to laterally presented emotional stimuli

Material Information

Title:
Psychophysiological and reaction time responses to laterally presented emotional stimuli
Creator:
Cimino, Cynthia Rodrigues, 1958-
Publication Date:
Language:
English

Subjects

Subjects / Keywords:
Acceleration ( jstor )
Asymmetry ( jstor )
Emotional expression ( jstor )
Emotional states ( jstor )
Facial expressions ( jstor )
Heart rate ( jstor )
Hemispheres ( jstor )
Mental stimulation ( jstor )
Visual fields ( jstor )
Warnings ( jstor )

Record Information

Source Institution:
University of Florida
Holding Location:
University of Florida
Rights Management:
Copyright Cynthia Rodrigues Cimino. Permission granted to the University of Florida to digitize, archive and distribute this item for non-profit research and educational purposes. Any reuse of this item in excess of fair use or other copyright exemptions requires permission of the copyright holder.
Resource Identifier:
23090726 ( ALEPH )
22495798 ( OCLC )

Downloads

This item has the following downloads:


Full Text














PSYCHOPHYSIOLOGICAL AND REACTION TIME RESPONSES
TO LATERALLY PRESENTED EMOTIONAL STIMULI












BY

CYNTHIA RODRIGUES CIMINO





















A DISSERTATION PRESENTED TO THE GRADUATE SCHOOL OF THE UNIVERSITY OF FLORIDA IN PARTIAL FULFILLMENT
OF THE REQUIREMENTS FOR THE DEGREE OF
DOCTOR OF PHILOSOPHY UNIVERSITY OF FLORIDA


1988


















ACKNOWLEDGMENTS

I would like to extend my gratitude to those who provided the encouragement and support that enabled the completion of this project. First, I am grateful to my dissertation committee. I would like to thank my chair, Dr. Dawn Bowers, for her time, hard work, and patience over the years; for her skill at making me really think; and especially for her belief in my ability. I am also grateful to Dr. Kenneth Heilman for providing material and moral support and for sharing his knowledge and never-ending enthusiasm, creativity, and wonderment. I thank Dr. Rus Bauer for his advice on psychophysiological technique and statistical method and for his infrequent but apt clinical interpretations. I would also like to tnank Dr. Eileen Fennell for ner discerning comments on methodology, her ground-rooted advice on professional development, and her kindness. I am also grateful to Dr. Ed Valenstein for his support in my defense and for his thought-provoking question at the end of my dissertation copy (humble as always). Finally, I am grateful to Dr. Hugh Davis for so many things, but especially for his guidance in development of my clinical abilities, his warmth and playfulness, and his love of language and verbal tapestries.

I owe a debt of gratitude to Cindy Zimmerman for her skill in organizing the typing and completion of the manuscript. My appreciation also goes to Dr. Roger Blashfield for his unconditional




ii












support and friendship and for his honest and direct means of pushing me to grow, personally and professionally. I thank Mieke Verfaellie and Karen Froming for their solid support and friendship over the years. With love and deepest appreciation, I also thank my husband, Pat, for his continued encouragement and enduring love and support.

















































11



















TABLE OF CONTENTS


Page

ACKNOWLEDGMENTS..................................................... ii

ABSTRACT............................................................ vi

INTRODUCTION ..................................................... I

Right Hemisphere Superiority for Recognizing Emotional
Aspects of Stimuli...............................................3
Emotional Prosody Studies.................................. 3
Affective Faces Studies...................................8
Related Research.........................................13
Right Hemisphere Dominance in Regulation of Mood and Affect....16
Emotional Activation Studies .............................16
Mood and Affect Studies..................................20
Left Hemisphere Superiority for Positive Affect; Right
Hemisphere Superiority for Negative Affect ...................25
Mechanisms of Emotional Processing: The Role of Arousal.......31 Neuropsychological Models of Emotional Processing.............. 36
The Model of Fox and Davidson ............................37
Kinsbourne's Model.......................................38
Tucker's Model...........................................40
Heilman's Model............................................42
Critical Issues................................................ 45
Hypotheses and Predictions.....................................49

METHOD..............................................................52

Subjects..........................................................52
Experiment I: Reaction Time Task ..............................52
Stimuli..................................................52
Apparatus.. ..............................................53
Procedure ................................................54
Experiment II: Psychophysiological Responses to Laterally
Presented Emotional Material.................................55
Stimuli...................................................55
Apparatus ................................................55
Procedure................................................57







iv












RESULTS ............................... ............................ 58

Experiment I: Reaction Time Task..............................58
Reaction Time Responses.................................58
Percent Correct Responses ...............................67
Experiment II................................................ 72
Heart Rate Data Reduction ...............................72
Skin Conductance Data Reduction........................... 84

DISCUSSION ..................................................... 88

Critical Issues..............................................95
Conclusions.................................................105

REFERENCES....................................................... 109

BIOGRAPHICAL SKETCH.............................................. 127








































V

















Abstract of Dissertation Presented to the Graduate School of the University of Florida in Partial Fulfillment of the
Requirements for the Degree of Doctor of Philosophy

PSYCHOPHYSIOLOGICAL AND REACTION TIME RESPONSES
TO LATERALLY PRESENTED EMOTIONAL STIMULI BY

CYNTHIA RODRIGUES CIMINO

December, 1988

Chair: Dawn Bowers
Major Department: Clinical and Health Psychology

Neuropsychological investigations of brain injured and

neurologically intact subjects have suggested that the two hemispheres differ in terms of their contribution to emotional processing. Several different models have been proposed to account for these observed hemispheric asymmetries. One model, the right hemisphere emotion model, suggests that the right hemisphere (RH) is globally involved in all aspects of emotional processing. A second model, the hemispheric valence model, suggests that the left hemisphere (LH) is dominant for processing positive emotions, whereas the RH is dominant for processing negative emotions. A third model, the preparatory model, suggests the RH is dominant for mediating arousal/activation and, as such, is more intrinscally involved in processing emotional stimuli that have greater "preparatory" significance for survival (i.e., fight-flight emotions such as anger and fear). In contrast, the LH is more involved in mediating nonpreparatory emotions (i.e., happiness, sadness, disgust) that place less immediate or "phasic"


vi











demands on the organism for survival. Alternatively, it is possible that the RH may be dominant for mediating arousal/activation responses to stimuli, regardless of their emotional/nonemotional content.

The focus of the present study was to further examine these different conceptualizations of the hemispheric processing of emotional stimuli in neurologically intact male and female subjects. This was accomplished in two separate experiments: (a) a choice reaction time task was used to investigate subjects' activation responses to a centrally presented, neutral stimuli when it was preceded by neutral or emotional warning stimuli and (b) heart rate

(HR) and skin conductance (SC) measures were used to investigate subjects' responses to laterally presented neutral and emotional stimuli. Findings from Experiment I, the reaction time experiment, failed to support any of the laterality models of emotion. Findings from Experiment II, using HR and SC measures, were more congruent with models of hemispheric differences in processing of emotional stimuli. Males showed lateralized effects of HR, arousal responses which support greater RH involvement in production of arousal responses. However, this effect was present regardless of the emotional/nonemotional content of the stimulus. Skin conductance responses in male subjects did provide some support for hemispheric specific emotional valence effects but this did not reach statistical significance. In contrast, to the hemispheric asymmetries in arousal responses observed in male subjects, female subjects did not show significant differences in arousal responses across left and right visual fields. However, females were differentially impacted by the



vii












emotional valence of the stimuli. Female subjects showed autonomic patterning effects across emotional categories, a finding which may be accounted for by imagery differences across male and female subjects. Taken together, these findings point to the relative importance of considering both sex and imagery ability of subjects in future investigations of emotional processing and autonomic responding.














































Vii


















INTRODUCTION



Neuropsychological approaches to the investigation of emotional processing have evolved, in large part, from early clinical observation of brain injured patients and subsequent systematic investigation of their performance on a variety of emotional tasks. One of the earliest reports was provided by Babinski (1914) in which he noted that patients with right hemisphere damage appeared indifferent or euphoric. Denny-Brown, Meyer, and Horenstein (1952) also reported evidence of such "indifference" reactions after right hemisphere lesion and noted its co-occurrence with unilateral neglect syndrome in whicn patients failed to orient, report, or respond to the left side of their body. In 1952, Goldstein published his observation that "catastrophic" emotional responses often accompanied left hemisphere damage. These reports were later corroborated by Hecaen (1962), who also noted that catastrophic reactions most often followed left hemisphere insult whereas indifference reactions were more frequent following right hemisphere damage.

In 1972, Gainotti reported a large scale study of 160 patients who had sustained left or right sided lesions. Based on systematic observation of the frequency and type of symptomatology, Gainotti reported a consistent relationship between behaviors indicative of a





1











depressive-catastrophic reaction and left hemisphere damage (LHD). Behaviors of emotional indifference and minimization of deficits were associated with right hemisphere damage (RHD). Similarly, Terzian (1964) and Rossi and Rosadini (1967) reported findings from unilateral carotid injection of sodium amytal and found depressive-catastrophic reaction following left sided injection and inappropriate euphoria following right sided injection. Milner (cited in Rossi & Rosadini, 1967) attempted to replicate these findings without success. In 1982, Sackeim and colleagues reported 119 cases of pathological laughing and crying in response to unilateral lesions. Results were congruent with Gainotti's findings with laughing outbursts more frequent following RHD and crying more often following LHD.

Goldstein (1948) as well as Gainotti (1972) have interpreted such depressive-catastrophic reactions following LHD as a "normal" response to a significant loss of physical as well as psychological function. In contrast, emotional changes seen with RHD were interpreted by Gainotti (1972) as an abnormal response associated with anosognosia or denial of illness.

More recent interpretations of such findings, however, have suggested that the indifference reaction may be due to the RHD patient's inability to accurately comprehend and/or express affect. This led to the hypothesis that the right hemisphere, in the intact state, is superior for the perception and/or expression of emotional material.

In the years which followed, more systematic investigation was applied to the understanding of the right hemisphere's role in the











processing of emotional stimuli. Two major areas of research have addressed this specific question: those which examine the processing of the prosodic elements of speech and those which examine the processing of affective faces. Related areas of investigation will also be discussed.



Right Hemisphere Superiority for Recognizing Emotional Aspects of Stimuli

Emotional Prosody Studies

It is well known that in right handers, the left hemisphere is more adept than the right hemisphere in decoding the linguistic content (semantic and phonemic elements) of speech (Benson & Geschwind, 1971). However, speech may carry at least two levels of information content: the linguistic content which conveys what is said and the prosodic content which conveys the way in which it is said. Prosodic elements which are defined as pitch, tempo, and rhythm, carry information about the emotional as well as nonemotional content of prosodic speech (Paul, 1909). Nonemotional prosody is important for conveying whether a sentence is a question, a statement, or a command. Emotional prosody is critical for conveying affective information.

In 1975, Heilman, Scholes, and Watson studied the ability of

right temporo-parietal and left temporo-parietal damaged patients to identify affective prosody. Patients were presented with semantically neutral sentences which were read in one of four emotional tones-happy, sad, angry, or indifferent. In this study, the subject's task











was to identify the emotional tone of the speaker. Results demonstrated that the RHD performed significantly worse on this task than the LHD group.

In a subsequent study, Tucker, Watson, and Heilman (1977b) investigated patient's ability to discriminate between different affectively toned sentences. In this task, subjects were not required to identify the affective tone but to discriminate between sentences presented with the same or different affective tone. RHD patients, again, performed more poorly than LHD patients, providing further evidence for the right hemisphere's greater role in the processing of emotional aspects of speech.

Weintraub, Mesulam, and Kramer (1981) have suggested that RHD

patients have difficulty in both emotional and nonemotional aspects of prosodic speech. In their study, RHD patients had significantly greater difficulty in distinguishing whether filtered sentences were questions, commands, or statements. Based on their findings, Weintraub et al. suggested that the poor performance of RHD patients on emotional prosody tasks may be accounted for by a more general deficit in the processing of prosodic information. No LHD group was reported in this investigation.

In a subsequent study which addressed this question, Heitman, Bowers, Speedie, and Coslett (1984) investigated the ability of RHD and LHD patients to comprehend filtered sentences which contained either emotional (happy, sad, angry) or nonemotional (declarative, imperative, or interrogative) prosody. Results demonstrated that both RHD and LHD patients were impaired on the nonemotional prosody task











compared to control subjects. In contrast, RHD patients performed significantly worse on the emotional prosody task relative to LHD patients, suggesting a greater role for the right hemisphere in the comprehension of emotional prosody.

In addition to these reports of deficits in the comprehension and discrimination of affective prosody, Tucker et al. (1977b) also found that RHD patients had difficulty in producing effectively intoned speech. Patients were asked to say a semantically neutral sentence using either a happy, sad, angry, or indifferent tone. RHD patients performed significantly worse than LHD patients suggesting that their deficits include not only the comprehension and discrimination of affectively intoned speech but also the expression of affectively intoned speech.

This finding was later supported by Ross and Mesulam (1979) who reported two patients who could not express affectively intoned speech but could comprehend affective speech. In addition, Ross (1981) has also reported patients who could not comprehend affective intonations but could repeat affectively intoned speech. Ross has suggested that the right hemisphere may mediate the comprehension, repetition, and production of affective speech much in the same way as the left hemisphere does for propositional speech with anterior lesions producing primarily production defects and posterior lesions producing primarily comprehension defects.

With the advent of experimental procedures such as tachistoscopic presentation and dichotic listening in which stimulus processing is initially restricted to the left or right-hemisphere, investigations












of emotional processing in normal, neurologically intact subjects have also examined the right hemisphere's role in the processing of emotional prosody. Using a dichotic listening (DL) procedure, Haggard and Parkinson (1971) paired speech babble with short sentences spoken in one of four emotional tones. They found that accuracy in identifying the emotional tone was significantly better on left ear trials, suggesting a right hemisphere superiority.

Similarly, Safer and Leventhal (1977) used monoaural presentation of sentences with positive, negative, and neutral content spoken in a positive, negative, or neutral tone. Results demonstrated that subjects who listened to sentences in their left ear tended to use the intonation in making their judgements whereas subjects who listened to sentences in their right ear tended to use the content of the sentence in making judgement. Interpretation of these findings, however, has been questioned because of the use of a between groups comparison for each ear presentation. The findings are suggestive, nevertheless.

Ley and Bryden (1982), using a DL procedure, had subjects report on the emotional tone and content of a sentence arriving at either the left or right ear (specified on each trial). Subjects were more accurate in judging the emotional tone of the sentence when monitoring the left ear and more accurate in judging the content of the sentences when monitoring the right ear.

In a subsequent experiment, Bryden, Ley, and Sugarman (1982)

investigated hemispheric differences in ability to judge the emotional tone of musical stimuli. They did this by taking advantage of the fact that in Western culture music written in a major key is generally







7



described as happy while music written in a minor is more often described as sad (Davies, 1978). In a DL procedure, subjects were required to identify the emotional tone of a short seven-note passage while monitoring either the left or right ear. Subjects were more accurate when identifying the emotional tone of passages presented to the left ear relative to those presented to the right ear, again supporting the notion of the right hemisphere dominance in processing of emotional stimuli.

Dichotic listening procedures in normal subjects have also

demonstrated left ear advantages in recognition of other nonverbal, emotional aspects of human speech such as laughing and crying (Carmon & Nachson, 1973; King & Kimura, 1972). In a study which used a variant of the monoaural paradigm in whicn subjects heard spoken captions and laughter in either the left or right ear, cartoons were judged as funnier when the laughter was heard by the left ear relative to the right ear (DeWitt, 1978).

Recently, Mahoney and Sainsbury (1987) investigated hemispheric asymmetries in perception of human, nonspeech emotional sounds. During conditions of divided attention, a left ear advantage emerged during the second block of trials. Under conditions of selective attention, however, this left ear advantage was seen on the first olock of trials. In addition to providing support for a right hemisphere superiority in processing of emotional nonspeech sounds, these findings also suggest that effects of attention influenced the rate and development of observed laterality effects but not the direction of these effects.







8



In summary, investigations in both normal and brain impaired subjects have provided evidence that the right hemisphere is preferentially involved in the comprehension and expression of emotional prosodic elements of speech and other nonverbal vocalizations. Similarly, a large body of literature has also investigated the role of the right hemisphere in the processing of affective faces.

Affective Faces Studies

In 1980, DeKosky, Heilman, Bowers, and Valenstein reported a

study which investigated RHD and LHD patients' ability to make neutral facial discriminations as well as affective facial discriminations. They found that RHD patients performed more poorly than LHD patients on both facial affect judgements as well as neutral facial discrimination. In fact, when the two groups were statistically equated for performance on the neutral discrimination task, differences between RHD and LHD patients on the affective facial discrimination task disappeared. These findings suggested that RHD patients' poor performance on facial affect judgements can be solely accounted for by their poor performance in facial discrimination ability. This has led to the question of whether processing the emotionality of a face, in fact, involves a "stimulus-content dimension" in its own right or whether such processing merely involves an increase in the configurational complexity of the stimulus and consequently increases the demand on right-hemisphere mediated visuospatial skills.












More recent investigations have challenged this latter

interpretation. Support for such a dissociation has recently been provided in a study by Kolb and Taylor (1981). Their findings revealed that patients with parietal excision were impaired on both matching of facial affect and matching of facial identity. However, these authors also report that patients with damage restricted to right temporal and right frontal regions were more impaired on processing of facial affect relative to processing of facial identity. Similarly, a study by Freid, Mateu, Ojemann, Wohns, and Fedio (1982) reported that neutral facial matching was disrupted by stimulation of right parietal-occipital regions, whereas naming the affect depicted on pictures of faces was disrupted by stimulation of right posterior middle temporal gyrus.

In contrast, Cicone, Wapner, and Gardner (1980) found no relationship between RHD patients' performance on an emotional perception task and a facial identity task, suggesting that deficits associated with RHD cannot be accounted for by facial recognition deficits alone.

More recently, Bowers and Heilman (1984) reported a case which demonstrated a dissociation between processing of affective and nonaffective faces. Although this RHD patient was able to perform well on neutral facial tasks and on same-different emotional faces tasks, he was impaired on naming and comprehension of verbal labels for facial expressions.

In a subsequent investigation, Bowers, Bauer, Coslett, and

Heilman (1985) again addressed the question of whether defects shown







10



by RHD patients on facial affect tasks are dissociable from defects in visuoperceptual processing. They cite two criticisms of the DeKosky study which directly address this issue. First, they noted that a small subset of RHD patients in the DeKosky study performed normally on the neutral visuoperceptual task, yet, were impaired on the facial affect tasks. This suggested that visuoperceptual deficits alone cannot account for impaired processing of affective faces in all RHD patients.

Secondly, they suggested that the use of same actors on affective faces trials may have allowed subjects to rely on a pure template matching strategy in which judgements about emotionality could have been made on the basis of whether the two faces had the same physiognomic configuration. A defect in this type of perceptual process could then potentially affect performance on both facial identity as well as affective facial tasks. As an alternative, Bowers et al. required subjects to make affective facial judgements across different actors so that such judgements would take place in an "associative" context with less reliance on potential defective perceptual matching.

Results of this study revealed that when patient groups were statistically equated on visuoperceptual ability (facial identity task), RHD patients still performed worse than LHD patients and control subjects on (a) emotional discrimination of different actors,

(b) naming the emotion of a single face, and (c) picking the named emotion from four pictures of the same actor. These findings provide strong evidence that differences in LHD and RHD patients' abilities to












make affective judgements cannot be accounted for solely by differences in the visuoperceptual processes underlying facial identity discrimination.

Recent support for the relative dissociation of facial identity judgements from facial affect judgements has also been provided by Tranel, Damasio, and Damaslo (1988). Tranel et al. described four patients with bilateral lesions of occipitotemporal or temporal regions whose performance on facial affect tasks were significantly better than their performance on facial identity tasks.

Research in normal subjects has also investigated the role of the right hemisphere in the processing of affective faces. Ley and Bryden (1979) tachistoscopically presented faces to the left visual field (LVF) and right visual field (RVF), and subjects made either facial identity judgements or facial affective judgements. They found a LVF superiority for both tasks. However, when performance on the facial affect task was reanalyzed using performance on the facias identity task as a covariate, tne LVF superiority for making affective facial judgements remained. These findings, which are similar to those of Bowers et al. (1985), suggest that the right hemisphere superiority for processing facial affect exists above and beyond the superiority for processing facial identity.

In 1977, Suoeri and McKeever reported a reaction time (RT) task in which subjects responded to previously memorized emotional or nonemotional faces presented in LVF or RVF. Subjects who memorized emotional faces showed significantly faster reaction times to LVF targets than subjects who memorized nonemotional faces. Based on







12



these findings, Suberi and McKeever argued that the LVF effect for processing nonemotional faces is significantly enhanced by presentation of emotional faces.

McKeever and Dixon (1981) used emotional imagery and neutral faces to investigate right hemisphere effects in processing of affective material. They instructed subjects to imagine that something very sad happened to a number of predetermined target faces. In a subsequent target/nontarget discrimination task with lateralized presentations, they report that the use of emotional imagery significantly enhanced LVF (right hemisphere) performance. This effect, however, was demonstrated in female subjects only.

Safer (1981) reported a study in which subjects memorized faces by either empathizing with their emotional expressions or by labeling the emotional expressions. Results demonstrated that subjects who used empathy recognized more faces presented to the LVF than RVF. No laterality effect was demonstrated for those who labeled faces. This laterality effect for the empathy condition, however, was found for male subjects only. Similarly, Buchtel, Campari, DeRisio, and Rota (1978) reported faster responding to both positive and negative stimuli presented in LVF relative to neutral targets. Hansch and Pirozzolo (1980) and Strauss and Moskovitch (1981) also reported a LVF effect for neutral and emotional faces.

In summary, investigations in both normal and brain impaired subjects have supported the notion that the right hemisphere is preferentially involved in the processing of affective faces. In addition, several studies have also suggested that right hemisphere







13



advantage for processing of facial affect exists above and beyond the right hemisphere's advantage for processing of facial identity. Evidence about possible differences due to sex of subjects, however, remains equivocal.

Related Research

Several studies have implicated the right hemisphere in memory for emotionally charged materials. Weschler (1973) reported one of the few studies of emotional memory in brain impaired subjects. Right hemisphere and LHD patients were presented with two types of stories-one emotional and the other nonemotional. When asked for subsequent recall, RHD subjects made significantly more errors in recalling emotional stories relative to LHD patients.

Cimino, Verfaeliie, Bowers, and Heilman (1988) investigated whether RHD patients have difficulty remembering past affective episodes by asking them to recall prior emotional and neutral experiences. Findings revealed that RHD patients produced significantly less emotional reports than control subjects as judged by independent raters. However, their own emotionality ratings were no different from those of control subjects suggesting some discordance between their actual production of emotional memories versus their own perceived emotionality of such memories. Unfortunately, most patients with LHD are aphasic and could not be used in this study. Therefore, this report cannot conclude that this defect is specific to RHD.

An investigation in normal subjects (Gage & Safer, 1979) looked at hemispheric differences in mood-state dependent effects for







14



recognition of emotional faces. The mood-state dependent effect states that stimuli will be better recalled when subject's mood at encoding and retrieval are congruent than when they are disparate (Bower, 1981). Using a mood induction procedure, Gage and Safer demonstrated that recognition of faces initially presented in a discrepant mood was significantly worse when presented to the LVF (right hemisphere) than when presented to the RVF (left hemisphere). Based on these observations, the authors suggest that the right hemisphere stores the subject's mood as an integral part of the memory representation to a greater extent than the left hemisphere.

Although it is well known that the left hemisphere is specialized for mediating speech and linguistic stimuli in the vast majority of right handed adults, the right hemisphere also appears to have some reduced language capacity (Geschwind, 1969; Papanicolaou, Moore, Deutsch, Levin, & Eisenberg, 1988). Several recent studies have suggested that the right hemisphere may better process emotional versus nonemotional verbal stimuli, at least at the single word level. For example, Graves and coworkers (Graves, Landis, & Goodglass, 1980) found that aphasic patients with alexia due to left hemisphere lesions could read emotional words better than nonemotional words. In a subsequent study, these same investigators (Graves, Landis, & Goodglass, 1981) found that neurologically intact males subjects better recognized emotional words than nonemotional words when these stimuli were presented to the LVF (right hemisphere).

Other researchers (Brody, Goodman, Holm, Krinzman, & Sebrechts, 1987) have looked at the effects of lateralized affective priming







15



stimuli (faces/words) on subsequent judgements of the affective value of laterally presented emotional and nonemotional verbal target stimuli. They found that affective primes presented to the RVF (left hemisphere) resulted in decreased accuracy judgements of the target stimuli that were also presented to that hemisphere. In contrast, affective primes presented to the LVF (right hemisphere) resulted in increased accuracy judgements regarding the affective value of verbal stimuli presented to the right hemisphere.

Taken together, findings with both aphasic and neurologically

intact individuals suggest a right hemisphere advantage in processing emotional verbal stimuli. However, this view is not entirely clearcut in that other studies have failed to replicate the "right hemisphere" laterality effect for identifying emotional versus nonemotional verbal stimuli (Strauss, 1983).

Another area of investigation concerning the role of the right hemisphere in processing of emotional stimuli is that of humor appreciation. Brownell et al. (1984) have recently reported that RHD patients have significant difficulty in understanding narrative humor as portrayed in short story jokes. Similarly, BinrLe et al. (1986) have reported that RHD patients performed significantly worse than LHD patients on a nonverbal cartoon completion task.

In summary, a large body of literature suggests that the right

hemisphere plays a greater role in the comprehension and expression of emotional information. These findings have been consistently demonstrated in both neurologic as well as normal populations across a wide range of tasks including affective facial judgements and







16



production, emotional prosody judgements and production, humor appreciation, and emotional memory.



Right Hemisphere Dominance in Regulation of Mood and Affect

In addition to studies which provide evidence that the right hemisphere is superior for recognizing emotional aspects of information, recent investigations have also suggested that the right hemisphere is dominant in regulating mood and affect. These studies fall into two major categories. The first category includes those studies which suggest that the right hemisphere is preferentially activated during episodes of felt emotion, primarily in normal subjects. The second category includes those studies which correlate psychiatric disorders of mood and affect with decrement in right hemisphere functions.

Emotional Activation Studies

Investigations which have looked at the right hemisphere's

activation during period of felt emotion have used several different indices of cerebral activation. These have included such measures as electrocortical activity (usually in terms of decreased alpha power), measures of lateral eye movements, and asymmetries of facial expression.

Davidson and Schwartz (1976) reported that subjects showed

greater right than left hemisphere EEG activity when recalling past events associated with anger or relaxation and during self-reported emotional reactions to visual material (Davidson, Schwartz, Saron,







17



Bennett, & Golemena, 1979). Right hemisphere activation has also been reported during hypnotically induced depression (Tucker, Stenslie, Roth, & Shearer, 1981), during generation of emotional imagery and during painful stimulation (Karlin, Weinapple, Rochford, & Goldstein, 1979).

In addition to comparisons of left versus right sided activation, several authors also emphasize the importance of relative differences in level of activation in anterior versus posterior regions. Tucker (1981) reported frontal activation but not posterior activation in depressed mood. Likewise, Davidson et al. (1979) reported that mood valence varied with right versus left activation in frontal regions, but that posterior regions showed right hemisphere activation irrespective of valence.

Lateral eye movements (LEM) as indices of hemispheric activation have also been used to assess the role of the right hemisphere in regulation of mood and affect. Prior investigations have revealed a tendency toward right LEM (left hemisphere activation) with verbal processing and left LEM (right hemisphere activation) with visuospatial processing (Kinsbourne, 1972). Schwartz, Davidson, and Maier (1975) have reported a greater frequency of left LEM in subjects performing emotional versus neutral mental tasks.

Similarly, Borod, Vingiano, and Cytryn (1988b) measured LEM while subjects were asked to generate emotional images of positive and negative valence in auditory, visual, and tactile modalities. Overall, subjects looked significantly more to the left than to the







18



right in response to emotional instructions. These findings suggest a greater role of the right hemisphere in generating emotional imagery.

In addition, Tucker, Roth, Arneson, and Buckingham (1977a) have reported more left LEM in anxious than nonanxious subjects. Woods (1977) has also suggested that habitual left eye movers are higher in intensity and frequency of emotional reactions than right eye movers.

While findings from LEM studies may appear to be conceptually

apparent, interpretations of findings from such investigations must be cautioned in terms of the questionable reliability and validity of LEM as indicators of hemispheric activation. Berg and Harris (1980) were unable to replicate previous findings in LEM studies and concluded that the validity of the LEM procedure as a measure of hemispheric activation has yet to be established. Ehrlichman and Weinberger (1978), in a detailed review of the LEM literature, similarly concluded that the use of LEM in investigations of hemispheric functioning was questionable at best.

Recently, lacrimal flow has also been utilized as an index of hemispheric involvement in production of mood states. Delp and Sackeim (1987) looked at lacrimal flow following sadness and happiness mood manipulation in male and female subjects. For female subjects, the sadness manipulation resulted in greater relative left eye lacrimal flow, whereas the happiness manipulation resulted in a shift toward greater relative reduction in left eye flow. Although these findings may be interpreted as support for greater right hemisphere involvement in lacrimal flow, this interpretation must be observed with some caution as the assumed lateralization of specific







19



neuroanatomical pathways regulating lacrimal flow have not been clearly established.

Measures of facial asymmetries observed during emotional

expression have also provided support for the hypothesis of right hemisphere dominance in regulating affect and mood states. Musculature of the lower part of the face is contralaterally innervated and asymmetries observed with respect to facial expression may be used to infer relative hemispheric involvement in production of emotional expression.

Studies with normal subjects have revealed that the left hemiface moves more extensively during posed facial expression (Borod & Caron, 1980; Borod, Caron, & Koff, 1981; 3orod, Kent, Koff, Martin, & Alpert, 1988a; Borod, Koff, & White, 1983; Moskovitch & Olds, 1982). One criticism of such studies, however, is that they have used posed facial expressions which may not actually reflect the underlying affect or mood of the subject. In response to this criticism, several authors have regarded spontaneous facial expression as a more valid index of the subject's affective state. Ekman, Hager, and Friesen (1981) failed to find such asymmetries of emotional expression during spontaneous facial expressions. In contrast, other investigators (Borod et al., 1983; Moskovitch & Olds, 1982) have observed greater left sided (right hemisphere) involvement for both spontaneous and posed facial expressions.

Several studies have also used composite photographs in which the mirror image of the left or right half of the face is combined with the original ipsilateral image. This process results in a complete







20



facial photo composed entirely of the left or right half of the face. These studies report that left sided composites of posed facial expressions are judged as more intense than right sided composites (Heller & Levy, 1981; Rubin & Rubin, 1980; Sackeim, Gur, & Saucy, 1978). This effect has also been reported for spontaneous expressions as well (Dopson, Beckwith, TucKer, & Bullard-Bates, 1984).

Clinical reports have indicated that spontaneous emotional facial expressions are less likely to occur in RHD patients compared to LHD patients (Borod, Koff, Perliman, Lorch, & Nicholas, 1986; Buck & Duffy, 1980; Ross & Mesulam, 1979). Kolb and Milner (1981), however, were unable to find differences between RHD and LHD patients in spontaneous facial expressions and movements (only some of which were emotional). They did find, however, differences with respect to the anterior versus posterior distribution of the lesion, witn anterior lesions resulting in less spontaneous movements than more posterior lesions. Borod et al. (1986) recently reported the only systematic investigation, to date, of posed and spontaneous facial expressions in brain impaired patients. They found that RHD patients produced fewer posed as well as spontaneous emotional facial expressions than did LHD patients.

Mood and Affect Studies

A second group of studies has also examined disorders of mood and affect associated with right hemisphere function. These investigations have studied mood and affective changes in patients with known hemispheric pathology. Patient groups have included







21



epileptic patients with unilateral foci, head injury patients, and patients with unilateral subcortical infarction or surgical removal.

In a factor analytic study which assessed interictal personality changes associated with right temporal lobe epilepsy, Bear and Fedio (1977) reported that right foci were more associated with affective changes while left foci were more associated with cognitive changes. Similarly, Taylor (1972) described a predominance of right sided foci in a sample of epileptics with associated symptoms of depression, anxiety, and phobias. In 1969, Flor-Henry reported that of patients with unilateral epileptic foci, manic-depressive psychosis was found twice as many times in individuals with right sided foci than in those with left sided foci.

Lishman (1968), in his review of 144 cases of head injury, found that affective disturbances were more common following right hemisphere injury while cognitive/intellectual changes were more often seen following left sided injury. These findings parallel those of Bear and Fedio (1977) which looked at changes associated with left and right epileptic foci. Recently, mania has also been reported following right thalamectomy (Whitlock, 1982) and right thalamic infarct (Cummings & Mendez, 1984).

Investigators have also looked at signs of hemispheric

dysfunction in patients with primary mood disturbance as a means of investigating lateralized hemispheric functioning in relation to disorders of mood and affect. Several investigators, utilizing EEG as a measure of hemispheric activity, have reported low left versus right hemisphere activity in depressed patients with differences occurring







22



primarily in frontal and central regions (d'Elia & Perris, 1973; 1974; Perris, 1975). These studies have also found that increased right sided activity correlated significantly with the severity of depression as well as performance on a verbal learning task. These findings have been replicated by other investigators as well (Rockford, Swartzburg, Chaudberg, & Goldstein, 1976). Perris (1974) has also reported lower left to right amplitudes of visual evoked responses in depressed patients relative to schizophrenics and normal controls. Taken together, these findings suggest greater right hemisphere relative to left hemisphere activity in depressed mood. Two interpretations of these findings have been suggested. One maintains that such hemispheric differences represent a relative left hemisphere underresponsiveness (Rockford et al., 1976). A second possibility is that such differences represent a right hemisphere overresponsiveness (d'Elia & Perris, 1974).

Recently, subtle left sided neurological signs have been reported in depressed patients, suggesting right hemisphere involvement. The first report was presented by Brumback and Staton (1981). They described two depressed children who presented with pronator drift of the left arm, hyperreactive left deep tendon reflexes, and left extensor plantar responses; these symptoms resolved following treatment with tricyclic antidepressants. Similarly, Freeman, Galaburda, Cabal, and Geschwind (1985) reported a case of a 62-yearold depressed female who had left sided facial weakness, a gaze preference to the right, and limited use of her left arm. Again, these symptoms resolved following treatment with ECT.







23



Findings from investigations of galvanic skin responses (GSR) have provided only partial support for the presence of lateralized dysfunction in depressed patients. Schneider (1983) reported lower right handed GSRs in a depressed sample. In addition, Myslobodsky and Horesch (1978) have noted higher left handed GSRs in depressed subjects. Toone, Cooke, and Lader (1981), however, were unable to replicate these findings. Such discrepancies may be accounted for by conflicting reports which suggest that GSR responses may be controlled ipsilaterally, contralaterally, or bilaterally (Holloway & Parsons, 1969; Lacroix & Comper, 1979; Myslobodsky & Rattok, 1977).

The performance of depressed patients on tests likely to require greater right hemisphere processing has also been investigated. Several studies sugggest that depressed subjects perform more poorly on visuospatial tasks than verbal tasks (Flor-Henry, 1976; 1983; Goldstein, Filskov, Weaver, & Ives, 1977; Kronfol, Hamsher, Digre, & Wazir, 1978). Siiberman, Weingartner, and Post (1983a) have also suggested that the pattern of errors in depressed subjects closely resembles that of right temporal lobectomized patients and that the degree of impairment is correlated with the overall severity of depression. However, as cautioned by Weingartner and Silberman (1982), the impaired verbal learning and memory performance that is frequently observed in depressed patients also points to left hemisphere dysfunction. Because of decrements observed on left as well as right hemisphere tasks, Weingartner and Siloerman (1982) suggested that it is best to describe the deficits observed as a







24



relative impairment on right hemisphere tasks as compared to left hemisphere tasks.

Several studies have also reported the occurrence of "reversed

lateralization" or "functional delateralization" in depressed patients on tasks of verbal and nonverbal processing. Bruder (1983) published a review of dichotic listening studies and concluded that depressed patients demonstrated decreased lateralization on both verbal as well as nonverbal dichotic listening tasks. Several authors, however, have suggested that decreased lateralization is present primarily on nonverbal tasks (Coulbourn & Lishman, 1974; Johnson & Crockett, 1982). Evidence of reversed lateralization has also been reported by some authors. Silberman, Weingartner, Stillman, Chen, & Post (1983b) have reported a left visual field superiority on a verbal task in a sample of depressed females. Similarly, research in depressed patients has also found (Flor-Henry, 1979; Flor-Henry & Koles, 1980) increased parietal activity during rest, witn left temporal activation during spatial tasks and right parietal activation during verbal tasks. These findings are at variance with predicted asymmetries in normal populations. Hommes and Panhuysen (1971), using a small sample of depressed patients, have reported that right sided sodium amytal injections resulted in a degree of aphasia for all subjects. Furthermore, this finding was significantly correlated with the severity of depressive symptomatology.

In summary, a large body of the neuropsychology literature on

emotional processing has suggested a right hemisphere dominance in the comprehension and expression of emotional information. These findings







25



have been consistently demonstrated in neurologic and psychiatric, as well as normal, populations using a range of tasks including affective facial judgements and production, emotional prosody judgements and production, emotional memory, emotional verbal judgements, production of affect, and humor appreciation.



Left Hemisphere Superiority for Positive Affect; Right Hemisphere Superiority for Negative Affect

Early clinical reports of emotional/mood changes following brain injury have also been interpreted as supporting a distinction between the superiority of the left hemisphere in the processing of positive affect and the superiority of the right hemisphere in the processing of negative affect. Interpretation of these studies is based on the notion of reciprocal inhibition which states that each hemisphere exerts some degree of inhinition on the contralateral hemisphere (Kinsbourne, 1973). In this view, damage to the left hemisphere would result in disinhibition of the right hemisphere's negative affective bias. Right hemisphere lesions would result in disinhibition of the left hemisphere's positive affective bias.

Early observations suggested that RHD patients often appeared indifferent or euphoric (Babinski, 1914; Denny-Brown et al., 1952). In contrast, several authors reported that left hemisphere damage was more associated with a "catastrophic" emotional response (Goldstein, 1952; Hecaen, 1962). These observations were later corroborated by Gainotti (1972) in a large scale study of 160 patients. Based on systematic investigation of patients' symptomatology, Gainotti







26



reported a consistent relationship between (a) depressive-catastrophic reaction and left hemisphere damage and (b) indifference/minimization of deficits and right hemisphere damage. Recently, Heilman and colleagues (Heilman et al., 1975; Heilman, Schwartz, & Watson, 1978; Heilman, Watson, & Bowers, 1983) have noted indifference reactions occur with striking frequency in RHD patients with the neglect syndrome suggesting that the two syndromes may be associated in some manner. This also suggests that right hemisphere changes associated with inappropriate euphoria and indifference may represent either (a) two points on a continuum or (b) two distinct emotional reactions following brain injury, one of them sharing a common mechanism with the unilateral neglect syndrome.

Using the Depression scale of the Minnesota Multiphasic

Personality Inventory as an index of depressive symptomatology, Gasparrini, Satz, Heilman, dnd Coolidge (1978) reported significantly elevated scores for LHD but not for RHD patients. More recently, Robinson and colleagues (Robinson, 1983; Robinson & Price, 1982) have also found that LHD patients were more likely to become clinically depressed and that RHD patients were more likely to be inappropriately euphoric. In addition, these results were not correlated with overall cognitive impairment, suggesting that the patient's emotional reactions to their deficits cannot solely account for these findings. Location of damage within the hemisphere has been found to be important in that these emotional reactions are more frequently associated with damage to anterior, frontal regions (Kolb & Milner, 1981; Robinson & Benson, 1981).







27



Findings from studies of unilateral carotid injection of sodium amytal (WADA procedure) also support reported differences in emotional changes following LHD and RHD. Terzian (1964) and Rossi and Rosadini (1967) reported depressive-catastrophic reactions following left sided injections and inappropriate euphoria following right sided injections. These findings are also supported by other reports (Alema, Rosadini, & Rossi, 1961; Perria, Rosadini, & Rossi, 1961). However, Milner (cited in Rossi & Rosadini, 1967) failed to replicate these findings. In her investigation, only 5% of patients displayed depressive type responses, while the majority displayed euphoric reactions. This discrepancy between Milner's study and those of other investigators may be related to the significantly higher doses of sodium amytal used in the Milner study (Silberman & Weingartner, 1986).

Investigation of normal, neurologically intact subjects has also provided some support for the relative superiority of the left hemisphere in the processing of positive affect and the relative superiority of the right hemisphere in the processing of negative affect. Davidson ana colleagues (1979) recorded EEG responses while subjects viewed television programs of varying emotional content and subsequently indicated their emotional responses. Greater left hemispheric activity was found in response to positive emotional content, and greater right hemispheric activation was found in response to negative emotional content. Interestingly, this difference was only apparent on more anterior, frontal recordings while posterior, parietal activity suggested relative right hemisphere







28



activation during all periods of felt emotion. These results again suggest the significance of the anterior-posterior dimension in processing of emotional information.

Asymmetries of emotional facial expressions have also tended to support the relative superiority of the left hemisphere in processing of positive affect and the relative superiority of the right hemisphere in processing of negative affect. Sackeim et al. (1978) have reported a tendency for facial expressions to be greater on the left side of the face. Furthermore, these authors suggest that this asymmetry was more pronounced for negative than positive facial expressions. Similarly, Schwartz, Ahern, and Brown (1979) have investigated facial expressions during spontaneous mood fluctuations. Right sided contractions were stronger during periods of happiness or excitement, while left sided contractions were stronger during facial expressions of sadness and fear.

Ahern and Schwartz (1979) have reported more right LEM (left

hemisphere activation) when subjects respond to questions that evoked happiness or excitement. In contrast, more left LEM (right hemisphere activation) occurred when subjects responded to questions that evoked sad or fearful affects. As previously discussed, the questionable validity of LEM as an index of cerebral activation, however, must be considered in any interpretation of this study.

Studies have also investigated left and right visual field

differences for positive and negative emotions. Using a contact lens system that restricts visual input to the RVF (left hemisphere) or LVF (right hemisphere), Dimond, Farrington, and Johnson (1976) reported







29



that unpleasant films were rated as more unpleasant when presented to the LVF than when presented to the RVF. More recently, Reuter-Lorenz, Givis, and Moskovitcn (1983) have also reported shorter reaction times to RVF presentations of happy faces and LVF presentations of sad faces. These findings are congruent with proposed left hemispherepositive affect and right hemisphere-negative affect distinctions.

Although numerous studies are consistent with the hypothesis that the right hemisphere preferentially mediates negative emotions and the left hemisphere mediates positive emotions, an equal number of studies find no support for this hemispheric valence hypothesis. Rather both positive and negative stimuli seem to be preferentially mediated by the right hemisphere. To account for these discrepant views on the hemispheric processing of positive versus negative stimuli, Bryden and Ley (1983) have argued that methodological differences across studies might contribute to the discrepant findings. For example, ReuterLorenz and Davidson (1981) report faster reaction times for LVF presentations of sad faces and RVF presentations of happy faces. In this study, subjects were required to identify which of two laterally presented faces (one neutral and one emotional) showed an affective expression. In investigations which show a significant overall right hemisphere effect, the task is quite different. In these studies, the task is usually to determine whether a laterally presented face is the same or different than a centrally presented face. It is possible that these differences in task requirements may, in some way, account for the findings.







30





A second possibility is that negative emotional faces may be more configurationally complex (requiring greater right hemispheric processing), a feature which may be of even greater significance in the task requirements of the Reuter-Lorenz task. This possibility does not, however, account for the large number of studies which have actually analyzed for type of emotion and still failed to find any laterality effect due to emotional valence (Bowers et al., 1985; Bryden et al., 1982; Buchtel et al., 1978; Heilman et al., 1984; Ley & Bryden, 1979).

A third possibility accounting for these discrepant findings is that studies which do find emotion specific hemispheric effects (i.e., left hemisphere-positive and right hemisphere-negative) are those which deal primarily with mood and/or experiential phenomena. In contrast, studies which do not find emotion specific hemispheric effects but instead do find right hemisphere superiority are those which involve cognitive encoding of emotional stimuli (i.e., "cold," cognitive tasks).

In their chapter, Bryden and Ley (1983) conclude that less evidence exists to strongly support the notion that the left hemisphere is more involved in the processing of positive affect and the right hemisphere is more involved in the processing of negative affect. The available evidence, however, provides strong support for the role of the right hemisphere in processing both positive and negative affective material.







31



Mechanisms of Emotional Processing: The Role of Arousal

Schacter and colleagues (Schacter, 1964; Schacter & Singer, 1962) proposed a theory of emotions termed the Cognitive-Arousal model. According to this model, an emotional state is the product of an interaction between arousal and cognition. An important assumption is that both arousal and cognition are necessary components of emotion. In this view, arousal is viewed as important in determining the felt intensity of the emotion while the cognitive element is important in determining the specific quality of the emotion.

Early support for the joint roles of arousal and cognition were provided by Maranon (1924, cited in Fehr and Stern, 1970) who artificially aroused subjects with administration of drugs that stimulated the sympathetic nervous system. Maranon's subjects did not report feeling emotions although some did report feeling "as if" emotions. In contrast, if subjects were given a congitive set (induction of an affective memory) they did report emotional reactions when artifically induced arousal was present. Scnacter (1970) has also provided support for this notion in a study which looked at the specific effects of pharmacologically induced arousal in neutral and stressful situation. Schacter demonstrated that physiological arousal alone (neutral condition) was not sufficient to evoke emotional responses from subjects but that the combination of arousal and availability of a cognitive label (stressful situation) was necessary. Although Schacter's research has been heavily criticized, especially on methodological grounds, this does not refute the







32



assumption that arousal and cognition may play an important role in emotion.

Early investigations in neurophysioiogy laid much of the

groundwork for our current knowledge of physiological arousal. In 1933, Berger reported that the electroencephalographic (EEG) pattern during behavioral arousal displayed decreases in amplitude and increases in frequency. This "electroencephalographic desynchronization" observed during periods of behavioral arousal was also later reported to occur during emotional states (Lindsley, 1970).

Studies have also identified critical neuroanatomic structures involved in the elicitation of arousal responses. Stimulation in nonspecific thalamic nuclei or the mesencephalic reticular formation (MRF) result in behavioral manifestations of arousal as well as EEG desynchronization (Moruzzi & Magoun, 1949). Similarly, stimulation of frontal or temporoparietal cortex activates the MRF (French, Hernandez-Peon, & Livingston, 1955) and elicits an arousal response (Sequndo, Nasuet, & Buser, 1955). Another pathway by which cortical stimulation can produce arousal is via limbic system projections to cortex and MRF (Heilman & Valenstein, 1972; Watson, Heilman, Cauthen, & King, 1973).

This conceptualization of reciprocal connections between MRF

system and cortical regions is central to a model of arousal proposed by Sokolov (1963). Sokolov described a specific pattern of physiological changes which occurred in response to novel or "significant" stimuli. This specific pattern of physiological changes was termed the orientating response (OR). At the behavioral level,







33



the organism may realign the head and/or body toward the source of stimulation. At the neurophysiological level, several changes occur which include a transient increase in skin conductance, pupil dilation, heart rate deceleration, pauses of respiration, and EEG desynchronization. The presumed functional value of these collective components of the OR is to make the organism more receptive to incoming stimuli as well as to prepare the organism for action.

A second component of Sokolov's model is that of the defensive response (DR), which is likely to be of equal, if not greater, potential significance in the processing of emotional stimuli. In Sokolov's view, when high intensity or aversive stimuli are presented, the orienting response is soon replaced by the defensive response. This response is characterized by greater increases in sympathetic activity across response systems including heart rate acceleration and cephalic vasoconstriction. The functional value of this response at the behavioral level is avoidance of the stimulus.

It is interesting to note that both orienting and defensive

responses are conceptualized as arousal responses, yet each results in characteristically distinct patterns of responding. This occurrence presents difficulty for the view of arousal as a unidimensional phenomenon. Subsequent investigators have looked at these seemingly paradoxical heart rate responses and attempted to correlate them with psychological processes.

Lacey (1967) reconceptualized this "directional f-actionation" of cardiac activity in terms of the conditions under which stimulation occurred and their effects on the organism's processing of stimuli.







34



Lacey proposed that cardiac deceleration is associated with environmental intake while heart rate acceleration is associated with environmental rejection (the intake-rejection hypothesis). Lacey suggested that cardiac deceleration served to facilitate sensory processing while cardiac acceleration served to inhibit sensory processing. In this view, stimuli which elicit attention and interest are associated with cardiac deceleration and environmental intake. In contrast, stimuli which are painful or aversive or which require a significant amount of mental activity such as problem solving or arithmetic are associated with cardiac acceleration and environmental rejection. Lacey (1967; 1972) also proposed a neurophysiological mechanism for such cardiac changes whereby cardiac responses altered cortical activity indirectly by means of a visceral afferent feedback loop mediated by the baroreceptors.

An alternative explanation for heart rate changes has been offered by Obrist and colleagues (1974) who have emphasized the relationship between motor requirements and cardiac activity. According to Obrist et al., there is a positive correlation between changes in cardiac activity and changes in level of somatic activity and both are controlled by integrative mechanisms in the central nervous system (cardiac-somatic coupling). Obrist et al. (1974) also noted that instances occur in which the cardiac-somatic coupling is dissociated, whereby increases in heart rate (termed cardiac preparatory responses) are observed without related overt changes in somatic activity. Interestingly, this occurs specifically in situations related to active avoidance of aversive stimuli.







35



Similarly, Freyschuss (1970) has observed heart rate acceleration when subjects are instructed either to tense or move an arm even though such movement is impossible because of experimentally induced paralysis. These observations suggest that cardiac activity is not solely coupled with overt somatic activity per se, but that cardiac activity is coupled with real as well as intended somatic activity.

While orienting and defensive responses result in

characteristically distinct patterns of autonomic responding, they have both been conceptualized as arousal responses. However, this view presents difficulty for the view of arousal as a unidimensional phenomenon. It also provides some support for the notion that autonomic reactivity is not as uniform as once suggested (Cannon, 1927; Schacter & Singer, 1962).

Ax (1953) provided some of the first evidence to suggest that various affective states may be associated with distinct autonomic patterning. Ax reported that diastolic blood pressure increased more during anger than fear imagery, while heart rate and systolic blood pressure increased with equal magnitude. More recent studies have replicated these findings (Schacter, 1957; Weerts & Roberts, 1976). Schwartz, Weinberger, and Singer (1981) have recently reported cardiovascular differentiation between imagery induced happiness, sadness, fear, and anger. Several investigators have also found significantly greater heart rate accelerations in response to fearful stimuli such as mutiliation slides, spiders, and fearful imagery (Hare & Blevings, 1975; Klorman & Ryan, 1980; Klorman, Weissberg, & Weisenfeld, 1977; Vrana, Cuthbert, & Lang, 1986). Recently, Ekman,







36



Levenson, and Friesen (1983) have also reported heart rate increases in response to production of emotional facial expressions of anger, fear, and sadness and heart rate decreases in response to disgust, surprise, and happiness.

In summary, there is some evidence to suggest that autonomic

arousal is not as uniform as once suggested. In fact, several studies support the notion that different patterns of autonomic arousal may be associated with different types of emotional states. In addition, recent conceptualizations of neart rate arousal responses have suggested that such changes may be linked to overt and/or covert motoric responses. This view is also consistent with the bioinformational theory proposed by Lang (1979). Lang proposes that emotional imagery results in patterns of autonomic activity very similar to those found in tne actual emotional situation. This raises the possibility that certain emotions, by virtue of their strong motor components, may be more associated with heart rate acceleration while others with less motor demands may be more associated with heart rate deceleration.



Neuropsychological Models of Emotional Processing

Several different neuropsychological motels of emotional processing have been proposed to account for findings of investigations in brain impaired and neurologically intact subjects. The models of Fox and Davidson (1984), Kinsbourne (Kinsbourne & Bemporad, 1984), Tucker (1981), and Heilman (Heilman et al. 1983; Heilman, 1988, personal communication) will be briefly reviewed.







37



The Model of Fox and Davidson

Fox and Davidson (1984) have proposed a developmental model of neural substrates underlying affective response systems. Their model is based on evolutionary considerations of the critical role of basic approach and avoidance systems. These authors propose that approach and avoidance comprise the two underlying behavioral dimensions upon which a11 subsequent affective subsystems and responses have evolved. In addition, they suggest that hemispheric specialization constitutes the critical, neuroanatomical substrate of approachavoidance behavior. More specifically, these authors propose that the left hemisphere is specialized for approach behaviors or positive affects while the right hemisphere is more specialized for avoidance behaviors or negative affects.

In support of tnis model, these authors relate the development of hemispheric specialization and interhemispheric transfer to the development of affective response systems. These authors argue that all of the primary emotions emerge over the first year of life (Bowlby, 1972; Charlesworth, 1964; Izard, Hubner, Risser, McGuiness, & Dougherty, 1980; Sroufe & Wunsch, 1972; Steinberg & Campos, 1983). Subsequent to this, they propose that primary emotions are modified by three processes: (a) the addition of new behaviors to the response repertoire of the "affect program," (b) regulation in the form of inhibition and appraisal, and (c) blending of primary emotions.

Interest and disgust are reliably elicited in the neonate (Izard, 1977), and Fox and Davidson (1984) cite evidence from EEG findings in infants to suggest that these emotions are lateralized. They found







38



more interest expressions and greater relative left-sided EEG activity following sucrose administrations compared with citric acid administrations in infants and argued that these findings provide support for left hemisphere superiority in processing of positive affect. They further proposed that these emotions are under unilateral hemispheric control since little functional interconnection between the hemispheres exists at birth.

Through the course of development, changes in interhemispheric

communication are proposed as the necessary substrate for emergence of fear and sadness in the emotional repertoire. In support of this, the authors cite evidence that the onset of locomotion, a behavior associated with commissural transfer, is tightly coupled to the emergence of fear (Bayley, 1968; 1969; Rader, Bausano, & Richards, 1980). In addition, the authors argue that the expression of sadness is often associated with alternating sequences of approach and avoidance, again implicating a critical role for interhemispheric communication (Ainsworth, Blehar, Waters, & Wall, 1978; Izard, 1977). The capacity to inhibit negative affective responses, which emerge during the second year, are also presumed to be linked to the functional integrity of the commissural system. In addition, these authors propose that the left hemisphere normally exerts an inhibitory influence on the right hemisphere through transcallosal connections, resulting in attenuation of negative affect in the normal state. Kinsbourne's Model

Kinsbourne and Bemporad (1984) proposeJ a separate model also based on the behavioral dichotomy that he terms action-approach and







39



inaction-avoidance. This model, however, invokes both anteriorposterior and lateral specialization of cerebral processing. Posterior regions are proposed to provide data necessary to maintain homeostasis and anterior regions are proposed to exert control to stabilize as needed. In addition, anterior systems are viewed as exerting inhibitory control over its posterior source of information. As in the conceptualization of Fox and Davidson, this model also proposes a left hemispheric approach and right hemispheric avoidance dichotomy. However, in Kinsbourne's model the left hemisphere is specialized for processing of external change and ongoing action (approach) and the right hemisphere is specialized for processing of internal changes, interruption of action (avoidance) as well as control of emotional arousal. This conceptualization establishes, in effect, four quadrants providing different mechanisms of processing:


Left-Anterior Action control over external change

Left-Posterior Enables left-anterior action control to make
contact with necessary exteroceptive information Right-Anterior Emotional control over internal arousal

Right-Posterior Enables right-anterior emotional control to make contact with interoceptive information With regard to the two control systems (action control and emotional control), Kinsbourne argues that the two hemispheres are not in inhibitory, but rather in compensatory interaction. Furthermore, whether the control is under left hemisphere-approach or right hemisphere-avoidance processing depends on the stimulus circumstances







40



and on the status of the organism's attempt to exert control over its environment or itself.

Kinsbourne provides a very interesting model of emotional

processing which is based, to some extent, on the approach-avoidance model of Fox and Davidson and further elaborated to include an anterior-posterior dimension. However, less evidence exists to support or refute his claims.

Tucker's Model

Tucker (1981) has also suggested a neuropsychological model of emotional processing based on lateralized neuroanatomical systems. These systems control (a) tonic activation and motor readiness and (b) phasic arousal responses to perceptual input. In contrast to notions of reciprocal inhibition of the hemispheres, Tucker invokes mechanisms of subcortical release of lateralized arousal systems to account for left-negative versus right-positive valence findings.

According to Tucker, the left hemisphere is specialized for

activation and complex motor operations. Support for this is provided by the preponderance of right hand motor dominance in the general population as well as observed deficits in both right and left hand production of learned, skilled motor movements (apraxias) following left hemisphere lesions (Geschwind, 1975). The presumed neurochemical substrate for this specialization is the dopamine system which several investigations suggest is predominantly a left lateralized system (Glick, Meibnach, Cox, & Maayani, 1979; Wagner et al., 1983). Tucker argues that activation operates in a tonic fashion to increase informational redundancy. This view is supported by the observation







41



that increased dopaminergic activity is associated with restriction of behavioral output by the production of motor stereotypies in humans and other animals (Ellinwood, 1967; Iversen, 1977). Tucker also cites evidence from psychiatric literature to suggest that negative emotions of anxiety as well as ritualized, stereotyped behaviors associated with obsessive-compulsive disorder and, to some extent, left partial complex seizure disorder represent subcortical release and subsequent overactivity of this left hemisphere activation system.

In contrast, Tucker proposes a right hemispheric specialization for phasic arousal responses to perceptual input which exerts its control through habituation. The presumed neurochemical substrate for this specialization is the norephinephrine system which some investigations have suggested is represented to a greater extent in the right hemisphere (Oke, KelLer, Mefford, & Adams, 1978; Oke, Lewis, & Adams, 1980). Lateralized norephinephrine pathways are known to show a pattern of widespread distribution throughout the brain providing the necessary substrate for arousal responses and facilitation of orienting to novelty. In support of this, Tucker notes greater right hemisphere ability in tasks of "global" versus "local" processing, requiring integration of perceptual input (Levy, 1969; Nebes, 1974). In addition, he cites evidence from the psychiatric literature suggesting greater right hemisphere involvement in hysteric personalities euphoric and often indifferent emotional responses which appear analogous to the responses of right hemisphere damaged patients (Galin, Diamond, & Braff, 1977; Gur & Gur, 1975; Smokler & Shevrin, 1979).







42



Tucker suggests that the two hemisphere's differing modes of processing may be the primary factor in lateralized valence effects reported in the literature. He hypothesizes that what we experience as emotions arise from operation of these arousal and attentional modulatory processes.

Heilman's Model

From investigations of indifference reaction associated with

right hemisphere damage and unilateral neglect syndrome, Heilman and colleagues (1983) have suggested a model of emotional processing based on hemispheric differences in arousal-activation responses. Heilman has suggested that right hemisphere damaged patient's difficulties in emotional expression may be a result of (a) deficits in arousalactivation and (b) an ability to develop an appropriate cognitive state due to basic deficits in comprehension of prosodic elements of speech and affective facial expressions. Patients with indifference reaction often have the unilateral neglect syndrome in which they may fail to orient, report, or respond to stimuli in the contralateral side of space (Denny-Brown et al., 1952; Gainotti, 1972; Heilman & Valenstein, 1972). Heilman et al. have suggested that unilateral neglect is a defect in attenuation-arousal-activation due to disruption of a corticolimbic-reticular loop (Heilman & Van Den Abel, 1979). Based on the fact that neglect occurs most often following right hemisphere damage, he has proposed that the right hemisphere may be dominant for mediating attention-arousal-activation responses.

To investigate arousal responses in brain impaired patients,

Heilman et al. (1978) stimulated the forearm ipsilateral to the side







43



of lesion in RHD and LHD patients while recording galvanic skin response from the same side. The authors noted that RHD patients had significantly smaller GSR arousal responses than LHD patients or nonbrain damaged control subjects. Similarly, Morrow, Urtunski, Kim, and Boiler (1981) presented LHD and RHD patients with neutral and emotional stimuli. Right hemisphere patients showed decreased galvanic skin responses to both neutral as well as emotional stimuli relative to LHD patients.

More recently, Yokoyama, Jennings, Ackles, Hood, and Boller (1987) have looked at heart rate and reaction time responses in patients with right unilateral hemispheric lesions. These authors found that RHD patients had significantly slower reaction times and decreased heart rate responses (both deceleratory as well as acceleratory) relative to LHD patients. These findings indicate that the greater role of the right hemisphere in attention may be reflected in both reaction time as well as anticipatory heart rate changes.

Investigations in neurologically intact subjects corroborate these findings. Hugdahl, Franzon, Anderson, and Walldebo (1983) report greater anticipatory heart rate accelerations for emotional stimuli presented to the LVF (right hemisphere) compared with RVF (left hemisphere) trials. Similarly Walker and Sandman (1982) also report greater right hemisphere activity (as measured by the P100 component of the average evoked potential) when the heart was spontaneously accelerated. In addition, Hugdahl, Wahlgren, and Wass (1982) found delayed habituation of the electrodermal orienting







44



response to visual stimuli initially projected to the right hemisphere.

Heilman and Van Den Abel (1979) have also suggested a right hemisphere superiority for activation. Using a neutral warning stimulus paradigm, these authors reported that warning stimuli projected to the right hemisphere reduced reaction times of the right hand more than warning stimuli projected to the left hemisphere. In addition, warning stimuli projected to the right hemisphere reduced reaction times of the right hand more than warning stimuli projected to the left hemisphere reduced reaction times of the left hand. Based on these findings, it can be seen that warning stimuli projected to the right hemisphere reduced reaction times of both hands to a greater extent than left hemisphere presentations, suggesting that the right hemisphere was better able to activate responses in both hands relative to the left hemisphere.

In addition, Verfaellie, Bowers, and Heilman (1987) reported a study of neurologically intact subjects which provides some support for the right hemisphere dominance of activation. By manipulating preliminary intentional warning cues (which hand to use in responding), they found faster left hand versus right hand responses suggesting that the left hand (right hemisphere) was better able to benefit from this preparatory information than the right hand (left hemisphere). In support of this, Verfaellie and Heilman (1987) also report the performance of two patients with right and left supplementary motor area (SMA) damage on this paradigm. They report that the patient with left SMA damage (intact right hemisphere) was







45



able to benefit from preparatory information while the patient with right SMA damage was unable to benefit from preparatory information, again suggesting a greater role of the right hemisphere in activation of response.

More recently, Heilman (1988, personal communication) has suggested that emotion specific hemispheric effects (i.e., left hemisphere-positive, right hemisphere-negative) reported in the literature may be artifactual and actually represent hemispheric differences in arousal and preparation for action. Because the right hemisphere is dominant for mediating arousal/activation, it may therefore be more intrinsically involved in processing emotional stimuli that have greater "preparatory" significance for survival (i.e., fight-flight emotions such as anger and fever). In contrast, the left hemisphere may be more involved in mediating nonpreparatory emotions (i.e., happiness, sadness, disgust) that place less immediate or "phasic" attentional demands on the organism for survival.

In summary, recent investigations in brain impaired and

neurologically intact subjects suggests a greater role of the right hemisphere in arousal-activation responses. Furthermore, this finding is observed across several indices of arousal-activation including heart rate, skin conductance, and reaction time measures.



Critical Issues

As reviewed in the introduction, there appears to be general consensus that the two hemispheres in man differ in terms of their contribution to emotional processing. However, the precise role







46



played by each remains unclear. Some investigators have argued that the right hemisphere is globally involved in all aspects of emotional processing including the cognitive encoding/decoding of emotional stimuli, arousal-activation responses to emotional stimuli and behavioral responses to these stimuli (Heilman et al., 1983; Ley & Bryden, 1979). Other investigators have argued that the two hemispheres differ in terms of the type of emotions that are preferentially mediated by each (Fox & Davidson, 1984; Kinsbourne & Bemporad, 1984; Tucker, 1981). The most popular version of this view is that the left hemisphere is dominant for positive (approach) emotions, whereas the right hemisphere is dominant for negative (avoidance) emotions.

Still others have argued that this positive-negative dichotomy in hemispheric processing of emotions is artifactual and actually elatess to hemispheric differences in arousal and preparation for action (e.g., activation) (Heilman, 1988, personal communication). In this view, the right hemisphere is dominant for mediating arousal/activation and as such, is more intrinsically involved in processing emotional stimuli that have greater "preparatory" significance for survival (i.e., fight-flight emotions such as anger and fear). In contrast, the left hemisphere is more involved in mediating nonpreparatory emotions (i.e., happiness, sadness, disgust) that place less immediate or "phasic" attentional demands on the organism for survival.

In order to distinguish among these models, it would be necessary to determine whether different categories of emotional stimuli result







47



in differential patterns of arousal/activation, depending on the hemisphere to which they were presented. According to the global right hemisphere emotion model, emotional stimuli of any valence directed to the right hemisphere should result in greater arousal/activation responses than those directed to the left hemisphere. According to valence models, negative emotional stimuli (anger, fear, disgust) would induce greater arousal/activation responses when directed to the right versus left hemisphere, whereas the opposite should occur when positive stimuli (happiness) are used. Finally, according to the preparatory/nonpreparatory model, emotional stimuli having preparatory significance (anger, fear) should result in greater arousal/activation responses when directed to the right versus left hemisphere. The opposite should occur with nonpreparatory emotional stimuli (happiness, disgust, neutral).

The focus of the present study was to further examine these divergent views regarding the hemispheric processing of emotional stimuli. The basic paradigm was one in which neut-al and emotional stimuli of different valence were laterally presented to either the left or right hemisphere (using a tachistoscopic procedure). The purposes of this study were to determine (a) the extent to which laterally presented emotional/nonemotional stimuli might result in differential patterns of behavioral activation (as assessed by reaction time responses) as well as differential patterns of autonomic arousal (as assessed by HR and SCR responses); (b) whether there were hemispheric asymmetries in mediating arousal and/or activation in response to these emotional/nonemotional stimuli; and (c) whether







48



certain categories of emotion (positive/negative; preparatory/ nonpreparatory) induce asymmetric arousal/activation, depending on the hemisphere to which they are initially presented.

In order to address these issues, two experiments were

completed. In the first study, laterally presented emotional stimuli of different valences served as warning stimuli to the subjects who then made manual RT responses to a neutral midline stimulus. This warning stimulus paradigm was chosen because it enables one to determine the extent to which lateralized emotional warning stimuli serve to behaviorally activate and prepare the individual to respond to a subsequent stimulus (Lansing, Schwartz, & Lindsey, 1959). In the second study, laterally presented emotional stimuli were also shown to subjects, and autonomic indices of arousal (HR, SCR) were measured. Although it would have been more "ideal" to obtain both autonomic and RT measures to the lateralized emotional stimuli in the same study, this was not realistically feasible. The "slow" rise time of the SCR (2-4 seconds) in conjunction with the relatively short lived activating effects of warning stimuli (500-2000 msec) precluded such a direct manipulation. Thus, two separate experiments were conducted.

Four different emotional categories were chosen for the present investigation. Two categories which have previously been shown to result predominantly in cardiac deceleratory responses (happy, disgust) and two categories which have previously been shown to result predominantly in cardiac acceleratory responses (fear, anger) (Ekman et al., 1983). Due to the relative paucity of discernable positive emotions among the wide range of emotional categories (Ekman, 1972),







49



it was not possible to equate occurrence of positive versus negative emotions.



Hypotheses and Predictions

According to the right hemisphere emotion model, the right

hemisphere plays a greater role than the left hemisphere in mediating arousal/activation responses to emotional materials. If this model is correct, then one would predict faster RT responses to a midline neutral stimulus when it is preceded by an emotional warning stimulus

(WS) directed to the right hemisphere (LVF) than by an emotional WS directed to the left hemisphere (RVF). Additionally, RTs should also be faster when emotional WS versus nonemotional WS are directed to the right hemisphere (LVF). These predictions were examined in Experiment I. Similarly, one would also predict that autonomic responsivity to emotional versus nonemotional stimuli should be greater when they are directed to the right hemisphere (LVF) versus the left hemisphere (RVF). These predictions were examined in Experiment II.

According to hemispheric valence models of emotional processing, negative emotional stimuli are preferentially mediated by the right hemisphere, and positive emotional stimuli are mediated by the left hemisphere. If tnis hypothesis is correct, then one would predict that negative emotional WS directed to the right hemisphere (LVF) should result in faster RTS to a neutral midline stimulus than when the negative WS is directed to the left hemisphere (RVF). Conversely, positive emotional WS directed to the left hemisphere (RVF) snould result in faster RTs than positive WS directed to the right hemisphere







50



(LVF). Similarly, in Experiment II, one would predict greater autonomic responsivity (HR acceleratory and deceleratory responses, SCR) to negative stimuli that are directed to the right hemisphere (LVF) versus stimuli that are directed to the right hemisphere (LVF) versus those directed to the left hemisphere (RVF). The opposite pattern of autonomic arousal should occur for positive emotional stimuli.

According to the hemispheric preparatory model of emotion, the

right hemisphere is dominant for mediating emotional stimuli that have a greater preparatory significance for survival (i.e., fight-flight emotions such as anger and fear). In contrast, the left hemisphere is dominant for mediating nonpreparatory stimuli that, place less "phasic" demands on the individual for immediate survival (i.e., happiness, disgust, neutrality). If tnis modei i3 correct, then in the first experiment one would predict that anger and fear WS directed to the right hemisphere (LVF) snould result in faster RTs than when they are directed to the left hemisphere (RVF). Conversely, happy, disgust, and neutral WS should result in faster RTs when they are directed to the left versus right hemisphere. Likewise, if one assumes that behavioral activation and autonomic responsivity are strongly coupled, then similar predictions would be made for Experiment II. That is, one would predict greater autonomic responsivity (HR acceleratory and deceleratory responses) to anger and fear stimuli (assumed to be preparatory) when they are directed to the right versus left hemisphere. The opposite hemispheric pattern of autonomic arousal







51



should occur for neutral, disgust, and happy stimuli (assumed to be nonpreparatory).

Alternatively, it is possible that the right hemisphere may be dominant for mediating arousal/activation responses to stimuli, regardless of their emotional-nonemotional content. In this view, stimuli directed to the right hemisphere should result in greater arousal/activation responses than stimuli directed to the left hemisphere. However, any differences in arousal/activation responses to emotional versus neutral stimuli should be comparable across the left and right hemispheres. Thus, in Experiment I, WS directed to the right hemisphere (LVF) should result in faster RTs than WS directed to the left hemisphere (RVF). Any RT differences between emotional versus nonemotional WS snould be comparable for LVF and RVF presentations. Likewise, in Experiment II, one would predict greater autonomic responsivity (HR, SCR) to stimuli directed to the right versus left hemisphere. Again, however, any differences in arousal responses to emotional versus neutral stimuli should be comparable across LVF and RVF presentations of the stimuli.

















METHOD


Subjects

A total of 60 (30 male, 30 female) students at the University of Florida served as subjects (Ss) in the present investigation. Subjects were given either course credit or paid for their participation. All Ss were right handed according to self-report and their performance on the Briggs and Nebes (1975) Handedness Questionnaire. An overall score of +9 or above (right hand preference) was used as the criterion for participation in this study.

Different groups of 30 Ss each (15 male, 15 female) participated in Experiment I and 1l. Subjects were randomly assigned to Experiment I or Experiment I!. Due to the potential confounding effects of prior exposure to the emotional stimuli on subsequent RT and psychophysiological responses, a between groups comparison was felt to be advantageous. The mean age of Ss in Experiment I was 20.7 with a range of 18-26 years. The mean age of Ss in Experiment II was 21.0 with a range of 19-27 years.



Experiment I: Reaction Time Task Stimuli

Stimuli consisted of 192 black and white slides. These slides depicted 96 neutral and 96 emotional scenes which included 24 slides





52







53



of each of four different valences (happy, angry, fearful, and disgusting). The emotional scenes were selected from a variety of materials including magazines and photography books. The scenes used in this experiment did not include familiar landmarks or personalities in an effort to avoid possible confounding effects of familiarity on RT and HR responses to these stimuli. The stimuli were rated for type and intensity of affect by 20 (10 male, 10 female) University of Florida students who did not participate in the present experiments. All stimuli averaged at least 91% agreement. Mean intensity ratings (on a scale of 1 to 5) for the five categories were happy, 3.4; angry,

3.6; fearful, 3.9; disgusting, 3.8; and neutral, .3. Apparatus

Slides were projected onto a 40x35-cm Kodak milk-glass, rear view projection screen. A 5-mm diameter red light-emitting diode (LED) was placed at the center of the screen to serve as a central fixation point and imperative stimulus in the RT task. Two spring loaded keys were placed 30 cm to the left and right of body midline. The timing, presentation of stimuli, and recording of Ss' responses from release of spring loaded keys were accomplished by an IBM-PC microcomputer interfaced with BRS logic. Slides were projected at 5 degrees of visual angle lateral to the central fixation LED. Attached to the projector lens were Uniblitz 325B high speed shutters which allowed maximum rise and fall time of slide presentation. The room was dimly lit to avoid the effects of visual startle during stimulus presentation.







514



Eye movements were continuously monitored by electro-oculography (EOG) to ensure maintained fixation and lateralized presentation of slides. The EOG signal was detected by Beckman Ag/AgC1 miniature electrodes attached at the temporal canthus of the left and right eye. The EOG signal was filtered and fed into a DC amplifier and recorded on a Grass Model 78B polygraph. Event marker input from IBM-PC microcomputer was also fed into the polygraph recording to identify occurrence of slide presentation and to facilitate subsequent identification of eye movements during this interval. Procedure

Subjects were seated 91 cm from the projection screen with left and right hands placed on left and right sided keys, respectively. The task was a choice reaction time task. Suojects viewed a laterally presented warning stimulus (neutral or emotional scene) of 500 msec. This was followed 500, 1000, or 1500 msec later by a centrally presented neutral, imperative stimulus (red LED) with interstimulus interval randomly varied across trials. Half of the Ss were instructed to release the left key following onset of the LED if the preceding warning stimulus was a neutral scene or the right key if the warning stimulus was an emotional scene. The remaining Ss were instructed to release the left key following onset of the LED if the preceding warning stimulus was an emotional scene and the right key if the preceding warning stimulus was a neutral scene. Half-way through the experiment, hand order was reversed for all subjects. Subjects received a total of 192 trials. Response hand and visual field of







55



presentation were randomized and counterbalanced across stimulus type. Subjects were given 10 practice trials prior to the experiment.



Experiment II: Psychophysiological Responses to
Laterally Presented Emotional Material Stimuli

The stimuli were identical to those used in Experiment I. They included 24 neutral and 96 emotional scenes [24 of each of four differing valences (happy, angry, fearful, disgusting)]. Apparatus

Slides were projected onto a 40x35-cm Kodak milk-glass, rear view projection screen. A 5-mm diameter adhesive circle was placed at the center of the screen to serve as a central fixation point. Slides were projected at 5 degrees of visual angle lateral to the central fixation point. Lafayette Model #43016 shutters were attached to the projector lens to maximum the rise and fall time of slide presentation. The timing and presentation of stimuli were accomplished by IBM-PC microcomputer interfaced with BRS logic. Eye movements were continuously monitDred in the same fashion as Experiment I. The room was again dimly lit to avoid the effects of visual startle during stimulus presentation.

Psychophysiological measures (HR, SCR, and respiration depth)

were recorded for the 3 seconds prior to stimulus onset and 8 seconds of stimulus presentation. Heart rate responses were recorded by two Beckman Ag/AgCl electrodes attached to the right and left lateral margins of the chest. A third electrode was attached to the







56



sternum. Electrode sites were prepared by mild abrasion of the skin with Hewlett Packard Redux paste. Electrodes were fastened by the use of adhesive collars and were filled with Hewlett Packard Jel Redux cream as the electrolyte. The electrocardiogram (ECG) was amplified by a Colbourn S75-03 high gain bioamplifier. This signal was input to a Colbourn 575-38 Ban-Pass Biofilter with subsequent detection of the R-wave component which interrupted the computer to provide inter-beat intervals. A-D conversion was accomplished by Colbourn R65-17 Data Translation Board and signal was downloaded to an IBM-PC interfaced with data acquisition modules.

Skin conductance was recorded with Met-Associates electrodes placed on the thenar and hypothenar eminences of the left and right hands. Electrode sites were wiped clean with distilled water. Electrodes were attached by the use of adhesive collars filled with KY jelly. The analog SC signal was fed into a Colbourn Model S71-22 Skin Conductance Module. This signal was digitized by a Colbourn R65-17 Data Translation Board and the signal was downloaded to an I3M-PC microcomputer interfaced with the data acquisition modules.

Respiration depth was also recorded to detect and subsequently exclude those trials in which unusually large inhalations or exhalations may have confounded HR or SCR. Respiration depth was measured by means of a Colbourn Model T41-91 Aneroid Chest Bellows, and processed by a Colbourn S72-25 Module. A-D conversion and computer interface utilized the same equipment as HR and SCR measures.







57



Procedure

Prior to the experiment, Ss were requested to list specific episodes from their own lives which they considered happy, angry, fearful, disgusting, or neutral. Subjects were requested to list five episodes for each category for a total of 25.

The experiment took place in a quiet, dimly lit room. Subjects were seated in a comfortable, recliner chair positioned 152 cm from the projection screen. After electrode placement and a 20-minute adaptation period, Ss were instructed to refrain from any unnecessary movement during the experiment.

Subjects were presented with a neutral or emotional slide in left or right visual field for 8 seconds. Subjects were instructed to maintain fixation on the centrally positioned circle throughout slide presentation. To decrease Ss' habituation and encourage continued processing of the stimuli during this time, Ss were also instructed to recall one of the five episodes of the same valence as the slide presented. Approximately 10 seconds after slide offset, Ss were asked to indicate the valence of the slide and responses were recorded manually on a separate sheet. Inter-trial interval varied randomly from 23 to 45 seconds to minimize occurrence of anticipatory HR and SCR. Subjects received a total of 120 trials with visual field of presentation randomized and counterbalanced across stimulus type. Subjects also received 10 practice trials for maintained fixation prior to the experiment utilizing neutral stimuli only.

















RESULTS


Experiment I: Reaction Time Task Reaction Time Responses

Subjects' reaction time responses served as the data for analysis in a repeated measures analysis of variance (ANOVA). The between subjects factor was Sex (male, female). The within subject factors were Visual Field (left, right), Hand (left, right), and Stimulus Type (happy, angry, fearful, disgusting, neutral) as within subjects factors. Incorrect trials and trials in which eye movements occurred were not included in the analysis. The remaining 81% of trials served as the data for analyses. This included 83% of happy trials, 82% of angry trials, 82% of fearful trials, 74% of disgusting trials, and 85% of neutral trials. A log transformation was used to correct for skewness in data distribution and to adequately meet homogeneity of variance assumptions for the ANOVA (Kirk, 1968; Winer, 1971).

A summary of the results of this ANOVA are depicted in Table 1. Findings revealed a main effect of Sex [F (1, 28) = 4.44, p = .0442)] with mean RT responses of male (M = 481.20) Ss significantly faster than RT responses of female (M = 603.48) Ss, depicted in Figure 1. A significant Sex x Visual Field effect [F (1, 28) = 4.06, p = .0535)] was also obtained and is depicted in Figure 2. Post-hoc simple effects testing was performed to clarify the nature of this interaction (Kirk, 1978; Winer, 1971). Findings revealed no


58







59






Table 1
Summary of Analysis of Variance: Experiment I, Reaction Time


Source df SS F p


Sex 1, 28 6.0277 4.44 Visual Field 1, 28 .0016 .25 Visual Field x Sex 1, 28 .0267 4.06 Hand 1, 28 .0375 .87 Hand x Sex 1, 28 .0032 .07 Stimulus Type 4, 112 .0874 1.38 Stimulus Type x Sex 4, 112 .1674 2.64 Visual Field x Hand 1, 28 .0000 .01 Visual Field x Hand x Sex 1, 28 .0008 .05 Visual Field x Stimulus Type 4, 112 .0334 .35 Visual Field x Stimulus Type x Sex 4, 112 .0966 1.02 Hand x Stimulus Type 4, 112 .1497 1.38 Hand x Stimulus Type x Sex 4, 112 .0386 .36 Visual Field x Hand x Stimulus Type 4, 112 .1995 2.49 Visual Field x Hand x Stimulus Type
x Sex 4, 112 .0357 .45


*p < .05.







60






510 506.28


500

E
P 49 483.47 Males only
. 479.51
_ 480 Ca 468.94 468.29
470


460
Happy Angry Fearful Disgusting Neutral

Emotion Figure 1. Experiment I--Reaction Time Analysis, Sex Main Effect.





700
Females

614.67
O 611.06 603.80 593.30 594.54 E 60o


Males
506.28
a 500 48.29 468.94 483.47 479.51




400 i
Happy Angry Fearful Disgusting Neutral Emotion

Figure 2. Experiment I--Reaction Time Analysis, Sex x Visual Field
Interaction.







61



significant differences across LVF and RVF trials for female Ss [F (1, 14) = 3.71, p = .0644)]. However, for male Ss, RVF (M = 477.10) were significantly faster than LVF (M = 485.49) trials [F (1, 14) =

5.23, p = .0299)].

Findings also revealed a significant interaction of Sex x

Stimulus Type [F (4, 112) = 2.64, p = .0376)], depicted in Figure 3. Post-hoc simple effects testing revealed no significant differences across Stimulus Type for female Ss [F (4, 56) = 1.26, p = .294)]. However, male Ss showed a significant effect for Stimulus Type [F (4, 56) = 2.60, p = .0454)]. Duncan's post-hoc comparisons revealed that for male Ss, RT responses to disgusting slides (M = 506.28) were significantly slower than happy (M = 468.29) or angry (M = 483.47) slides at p < .05.

A significant Hand x Visual Field x Stimulus Type interaction [F (4, 112) = 2.49, p = .0473)] was also obtained and is depicted in Figure 4. Post-hoc simple effects testing of RVF trials only revealed no statistically significant differences across Hand and Stimulus Type conditions [F (4, 116) = .97, p = .4249)]. In LVF, however, post-hoc simple effects tests revealed a significant Hand x Stimulus Type interaction [F (4, 116) = 2.61, p = .0390)]. Duncan's post-hoc comparisons revealed that happy trials in LVF were significantly faster when using the right (M = 507.45) versus left (M = 562.94) hand at p < .05.

Further inspection of the data revealed that the significant Sex effects observed in the preceding analysis may have been influenced by the markedly slowed performance of two female Ss. When compared to







62






650

603.48 600
E


(D
.o 550


Cr 500 481.20


450
Females Males Sex

(a)









700


601.97 604.99 Females 0E o0-

0

cu485.49 S500 477.10 Males




400
LVF RVF Visual Field

(b) Figure 3. Experiment I--Reaction Time Analysis, Sex x Stimulus Type
Interaction: (a) males only and (b) males and females.







63




Left Visual Field Trials
580- 578.99

562.94
s560 557.91 Right Hand 547.83
E 546.28 543.63 543.55 =- Left Hand C 540
.0 527.34

520
cc 520 507.45 521.39


500 1 1
Happy Angry Fearful Disgusting Neutral Emotion

(a)




Right Visual Field Trials 560o Right Hand

549.20
549,35 535
550 549.35 545.06
E
-*
C 540.82 O 540 Left Hand CU539.12 533.31 530
525.63
520.98
520
Happy Angry Fearful Disgusting Neutral Emotion
(b)

Figure 4. Experiment I--Reaction Time Analysis, Hand x Visual Field x
Stimulus Type Interaction: (a) left visual field trials
and (b) right visual field trials.







64



the overall mean RT of females for each of the 20 Visual Field x Hand x Stimulus Type conditions, these two Ss possessed 8 (40% of total) and 10 (50% of total) mean reaction times which fell two standard deviations above the overall mean performance of female Ss. Of the remaining 13 female Ss, no S possessed a single mean reaction time greater than 2 standard deviations above the overall mean of female Ss.

For this reason, a second analysis of the RT data was conducted in which the data from the two female Ss noted above was excluded. A summary of the results of this analysis are depicted in Table 2. Findings revealed no significant main effects. The main effect of Sex observed on the preceding analysis was no longer significant suggesting that this effect may have been significantly influenced by the markedly slowed performance of the two female Ss noted above. Findings, however, did reveal a significant Visual Field x Sex interaction LF (1, 26) = 5.20, p < .0311)] as in the preceding analysis, depicted in Figure 5. Simple effects testing revealed a pattern of findings similar to those noted in the first analysis with no significant difference between LVF and RVF performance for female Ss. Males, however, performed significantly faster to RVF stimuli relative to LVF stimuli. A trend for the Sex x Stimulus Type interaction was also noted [F (4, 104) = .0722, p = .0722)] which had previously attained significance in the first analysis. This interaction is depicted in Figure 6. The Hand x Visual Field x Stimulus Type interaction significant, in the first analysis, failed to reach significance in the present analysis.







65






Table 2
Summary of Analysis of Variance: Experiment I, Reaction Time (minus outliers)


Source df SS F P


Sex 1, 26 2.3201 2.17 Visual Field 1, 26 .0002 .03 Visual Field x Sex 1, 26 .0332 5.20 Hand 1, 26 .0441 1.15 Hand x Sex 1, 26 .0060 .16 Stimulus Type 4, 104 .1194 1.96 Stimulus Type x Sex 4, 104 .1350 2.22 Visual Field x Hand 1, 26 .0019 .12 Visual Field x Hand x Sex 1, 26 .0000 .00 Visual Field x Stimulus Type 4, 104 .0187 .20 Visual Field x Stimulus Type x Sex 4, 104 .0699 .76 Hand x Stimulus Type 4, 104 .1714 1.59 Hand x Stimulus Type x Sex 4, 104 .0357 .33 Visual Field x Hand x Stimulus Type 4, 104 .1514 1.86 Visual Field x Hand x Stimulus Type
x Sex 4, 104 .0362 .45


"* < .05.







66





600

556 556.84 Females 552.16
49 550
E


o 500 485.49 477.10 Males


450


400
Left Right Visual Field


Figure 5. Experiment I--Reaction Time Analysis (minus outliers),
Visual Field x Sex Interaction.





Females
580
567.11 565.08
560 5
544.02 54435 E 540

C
520
506.28
500o Males cc 483.47 479.51
480 468.29 468.29

460 I
Happy Angry Fearful Disgusting Neutral Emotion

Figure 6. Experiment I--Reaction Time Analysis (minus outliers), Sex
x Stimulus Type Interaction.







67



Percent Correct Responses

A separate ANOVA was conducted which used proportion of correct

responses for each S as the dependent variable. An arcsin square root transformation was performed to correct for lack of a normal distribution inherent in proportion data and to meet homogeneity of variance assumptions for the ANOVA (Kirk, 1968; Winer, 1971).

Results of this analysis utilizing all 30 Ss are depicted in

Table 3. Findings revealed only a significant main effect of Stimulus Type [F (4, 112) = 7.60, p = .0001)], depicted in Figure 7. Duncan's post-hoc comparisons revealed that percent correct identification of disgusting (M = .804) trials was significantly worse than all four remaining categories (happy, M = .907; angry, M = .904; fearful, M = .903; neutral, M = .938) at p < .05. In addition, a trend for Stimulus Type x Visual Field [F (4, 112) = 2.29, p = .064)] was also revealed, depicted in Figure 8. This pattern of findings suggest that happy trials were more accurate in the LVF while neutral trials were more accurate in the RVF.

A second ANOVA was conducted for percent correct data which

excluded the two female Ss noted above. Results of this analysis are depicted in Table 4. This analysis revealed only a significant main effect of Stimulus Type [F (4, 104) = 6.13, p = .0002)], depicted in Figure 9. Duncan's post-hoc comparisons revealed that percent correct identification of disgusting (M = .810) trials was significantly worse than all four remaining categories (happy, M = .911; angry, M = .904; fearful, M = .899; neutral, M = .940) at p < .05. No other effects reached significance or trend status.







68







Table 3
Summary of Analysis of Variance: Experiment I, Percent Correct


Source df SS F p


Sex 1, 28 .8665 2.28 Visual Field 1, 28 .0181 .22 Visual Field x Sex 1, 28 .1683 2.01 Hand 1, 28 .0757 1.23 Hand x Sex 1, 28 .0203 .33 Stimulus Type 4, 112 2.5844 7.60 ** Stimulus Type x Sex 4, 112 .2799 .82 Visual Field x Hand 1, 28 .0009 .01 Visual Field x Hand x Sex 1, 28 .0004 .01 Visual Field x Stimulus Type 4, 112 .4369 2.29 Visual Field x Stimulus Type x Sex 4, 112 .1908 1.00 Hand x Stimulus Type 4, 112 .4237 1.44 Hand x Stimulus Type x Sex 4, 112 .1018 .35 Visual Field x Hand x Stimulus Type 4, 112 .0793 .45 Visual Field x Hand x Stimulus Type
x Sex 4, 112 .1774 1.02


** p < .01.







69








0.95 .938
0 .907 .904 .903

O


T 0.85


.804


0.75
Happy Angry Fearful Disgusting Neutral Emotion


Figure 7. Experiment I--Percent Correct Analysis, Stimulus Type Main
Eft'ect.




Trend 100
.957 RVF .925
3 .903 LF S 90 .919
0 .904
0 .819


* 80a, o
... .789


70
Happy Angry Fearful Disgusting Neutral Emotion

Figure 8. Experiment I--Percent Correct Analysis, Stimulus Type x
Visual Field Trend.







70







Table 4
Summary of Analysis of Variance: Experiment I, Percent Correct (minus outliers)


Source df SS F p


Sex 1, 26 .9794 2.42 Visual Field 1, 26 .0009 .01 Visual Field x Sex 1, 26 .0552 .87 Hand 1, 26 .0498 .76 Hand x Sex 1, 26 .0091 .14 Stimulus Type 4, 104 2.0531 6.13 ** Stimulus Type x Sex 4, 104 .2958 .88 Visual Field x Hand 1, 26 .0205 .27 Visual Field x Hand x Sex 1, 26 .0180 .24 Visual Field x Stimulus Type 4, 104 .3694 1.86 Visual Field x Stimulus Type x Sex 4, 104 .2289 1.15 Hand x Stimulus Type 4, 104 .3254 1.16 Hand x Stimulus Type x Sex 4, 104 .1308 .47 Visual Field x Hand x Stimulus Type 4, 104 .0452 .28 Visual Field x Hand x Stimulus Type
x Sex 4, 104 .0866 .54


** p < .01.








71


















0.95 -94
0 .911
() .904 .899
O

0.810 a 0.85

CL



0.75
Happy Angry Fearful Disgusting Neutral
Emotion




Figure 9. Experiment I--Percent Correct Analysis, Stimulus Type Main
Effect.







72



Experiment II

Heart Rate Data Reduction

Heart rate responses were edited by a computer program which

converted all trials from interbeat interval data to beats per minute format for each second of the sampling period. Each interbeat interval was weighted proportionally to the fraction of the second it occupied according to method recommended by Graham (1980). Baseline for each trial was defined as the average HR for the 2 seconds preceding stimulus onset. For each trial, baseline HR was then subtracted from each of eight post-stimulus HR values, yielding second by second post-stimulus HR changes from baseline. Trials in which eye movements occurred were not included in the analysis. The remaining 77% of trials served as the data for analysis. This included 78% of happy trials, 77% of angry trials, 78% of fearful trials, 75% of disgusting trials, and 78% of neutral trials.

From this, two separate data sets were generated. The first, selected, for each trial, was the maximum deceleratory HR response from the eight post-stimulus second by second HR changes from baseline. This maximum deceleratory response then served as the dependent variable in a repeated measures ANOVA. The second data set selected, for each trial, was the maximum acceleratory HR response from the eight post-stimulus second by second HR changes from baseline. This maximum acceleratory response also served as the dependent variable in a separate repeated measures ANOVA.







73



Maximum deceleratory responses

A repeated measures ANOVA was performed with Sex (male, female) as a between subjects factor and Visual Field (left, right), and Stimulus Type (happy, angry, fearful, disgusting, neutral) as the within subjects factors. A summary of the results of this ANOVA are depicted in Table 5. Results of this analysis revealed a main effect of Sex [F (1, 28) = 5.16, p = .0310)]. Males (M = -9.153) showed significantly greater maximum deceleratory HR responses than females (M = -7.756). These findings are depicted in Figure 10. Results also revealed a trend for a main effect of Visual Field [F (1, 28) = 3.33, p = .0789)] with a pattern of greater deceleratory HR responses to LVF trials (M = -8.700) relative to RVF trials (M = 8.209) (see Figure 11).

This analysis also yielded a main effect of Stimulus Type [F

(4, 112) = 4.96, p = .0010)]. Duncan's post-hoc comparisons revealed happy slides (M = -9.016) elicited significantly greater HR aecelerations than angry (M = -7.969) or fearful (M = -7.663) slides at p < .05. Neutral slides (M = -8.984) also elicited significantly greater HR decelerations than angry or fearful slides at p < .05. These relationships are depicted in Figure 12.

Findings also revealed a significant Visual Field x Sex

interaction [F (1, 28) = 7.86, p = .0091)]. Simple effects testing revealed that males showed significantly greater deceleratory HR changes in LVF (M = -9.776) relative to RVF (M = -8.530) trials, depicted in Figure 13. Results also revealed a significant Sex x







74








Table 5
Summary of Analysis of Variance: Experiment II, Heart Rate Deceleration


Source df SS F p


Sex 1, 28 146.4789 5.16 Visual Field 1, 28 18.0850 3.33 Visual Field x Sex 1, 28 42.7167 7.86 ** Stimulus Type 4, 112 89.6031 4.96 ** Stimulus Type x Sex 4, 112 60.4664 3.35 ** Visual Field x Stimulus Type 4, 112 7.7819 .47 Visual Field x Stimulus Type x Sex 4, 112 12.6660 .77


*p < .05. ** p < .01.








75









-7

0
a -7.756


O




E -9 -9.153
0










-8.1

8.290







C
cu
C -8.5
-82








4 -8.6E -8.700
-8.7

-8.8
LVF VFM Visual Field

Figure 11. Experiment II--Maximum Deceleratory Heart Rate Analysis, Visual Field Trend.







76







-7
.o
-7.663

Q -7.969
S-8.64-8


= Ca-8.984


E -9.016

x
-10 ,
Happy Angry Fearful Disgusting Neutral Emotion


'Figure 12. Experiment II--Maximum Deceleratory Heart Rate Analysis,
Stimulus Type Main Effect.




-7
o
-7.888
7.624 v Females S -8-8.530



E
Males


E


-10

LVF RVF Visual Field


Figure 13. Experiment II--Maximum Deceleratory Heart Rate Analysis,
Visual Field x Sex Interaction.







77



Stimulus Type interaction [F (4, 112) = 3.35, p = .0125)]. Simple effects testing revealed no significant effects across Stimulus Type for males [F (4, 56) = 1.86, p = .130)]. However, for female subjects this Stimulus Type effect was significant [F (4, 56) = 5.32, p = .001)]. Duncan's post-hoc comparisons revealed that deceleratory HR responses to happy (M = -8.306), disgusting (M = -8.345) and neutral (M = -8.638) were all significantly greater than deceleratory HR responses to angry (M = -6.120) slides at p < .05. These relationships are depicted in Figure 14. Maximum acceleratory responses

A repeated measures ANOVA was performed with Sex (male, female) as the between subjects factor and Visual Field (left, right) and Stimulus Type (happy, angry, fearful, disgusting, neutral) as the within subjects factors. A summary of the results of this analysis are depicted in Table 6. Results revealed a significant interaction of Stimulus Type x Sex [F (4, 112) = 3.48, p = .0102)], depicted in Figure 15. Simple effects testing revealed no significant effect of Stimulus Type for males [F (4, 56) = 1.86, p = .130)]. For female Ss, however, this effect was significant [F (1, 14) = 5.32, p = .001)]. Duncan's post-hoc comparisons revealed that angry (M = 3.473) and fearful (M = 3.888) slides elicited significantly greater HR accelerations than neutral (M = 2.158) slides at p < .05.

This analysis also revealed a trend for an interaction between Visual Field x Sex [F (1, 28) = 3.36, p = .0776)], with a pattern of greater acceleratory HR responses for males in LVF (M = 2.531) trials







78














-6.120



-7
0 -7.370


lii -8.306
3 0-8 3 -8.345 Females S-8.567-8
-8.567 -8.638
0 2 -9.205 -8.937
E 9 -9.726 -9.331 Males
-10
Happy Angry Fearful Disgusting Neutral Emotion



Figure 14. Experiment II--Maximum Deceleratory Heart Rate Analysis,
Sex x Stimulus Type Interaction.







79








Table 6
Summary of Analysis of Variance: Experiment II, Heart Rate Acceleration


Source df SS F p


Sex 1, 28 28.2236 .65 Visual Field 1, 28 .5214 .13 Visual Field x Sex 1, 28 13.2770 3.36 Stimulus Type 4, 112 28.6964 1.74 Stimulus Type x Sex 4, 112 57.5180 3.48 ** Visual Field x Stimulus Type 4, 112 13.3162 .73 Visual Field x Stimulus Type x Sex 4, 112 9.4844 .52


** p < .01.








80







S 4 -3.888


0
23.473
2 .158 Males S 3- 2.584 < c 2.358 Cfl

I"E Females E o 2 2.302 .z 2.097 2.838 S1.763

M I I I I I !
Happy Angry Fearful Disgusting Neutral Emotion


Figure 15. Experiment II--Maximum Acceleratory Heart Rate Analysis, Stimulus Type x Sex Interaction.




3.2
3.016 END
c TI:END
3.0

8 e 28- 2.763

S2 Females X M .6 2.531
E
0
2.24

E 2.2
2.027
Males
2.0
LVF RVF Visual Field

Figure 16. Experiment II--Maximum Acceleratory Heart Rate Analysis, Visual Field x Sex Interaction.







81




relative to RVF (M = 2.027) trials. This pattern of findings is depicted in Figure 16.

Second by second responses

Second by second post-stimulus HR changes for male and female Ss are displayed for each Visual Field x Stimulus Type condition in Figures 17 and 18, respectively. These values represent the average change from baseline for each of the eight post-stimulus seconds. Post-stimulus HR changes from baseline for both male and female Ss suggest that, overall, Ss showed primarily HR deceleratory changes throughout the presentation of the stimuli. This may be accounted for, in part, by task factors which required perceptual intake throughout the presentation of the stimulus. Perceptual intake has been associated with HR deceleration as suggested by Lacey (1967).

It is also of note that post-stimulus HR changes from baseline do not reflect occurrence of acceleratory HR responses (i.e., positive HR change from baseline). It may be that the emotional stimuli themselves were not inherently "strong" enough to elicit acceleratory HR responses. However, findings from the analysis of maximum acceleratory HR responses do indicate that, at some point during the 8 post-stimulus seconds, Ss are experiencing HR acceleration. The most plausible explanation for the lack of post-stimulus HR accelerations in graphic representation of the 8 post-stimulus seconds is that these acceleratory responses are occurring at different points in time across different trials and possibly across different Ss. That is, accelerations may be "averaged out" by decelerations occurring at the same time on separate but like trials (i.e., same VF x Stimulus Type








82



Males Males
0 0

-1 "- LVF-Happy -1 LVF-Angry 7 -5 RVF-Angry
-2- RVF-Happy -2+ a -3- E -30-
0


-3 -6
D1 02 D3 0405 06 07 08 D1 02 D3 D4 D5 6 07 D08

Post-Stimulus Seconds Post-Stimulus Seconds



MaeMales









00
-1 LVF-NeuraLVF-isgusting
-a- LVF-Fearful 2 4 RVF-Oisgusting
SRVF-Fearful -2








5 -2
S-4- -4


-5 -6

01 02 03 04 05 D6 D 0 D7 D8
Post-Stimulus Seconds Post-Stimulus Seconds


Malesss









T 0ye Condtion
-1- -o-LVF-Neutrai
t n cRVF-Neurai -2 -3

-4
C


01 02 03 04 05 06 07 08 Post-Stimulus Seconds

Figure 17. Experiment II--Second x Second Post-Stimulus Heart Rate
Changes in Male Subjects for Each Visual Field x Stimulus
Type Condition.







33




oFemales Females
S0
S-1 LVF-Happy -1 .- LVF-Angry
-2 RVF-Happy +0 RVF-Angry
2 -2 0 -3 E -3-5 -4 -4

D1 D2 D3 04 D05 0D6 07 08 D1 02 D3 04 D5 06 D7 08
Post-Stimulus Seconds Post-Stimulus Seconds




Females -- LVF-Fearful Females
-+ RVF-Fearful 0 .- LVF-Disgusting
0 + RVF-Disgusting



.-2
a -2
. -3

-4 4

-5 -5
D1 02 03 D4 D5 D6 D7 8 02 3 D4 0506 D7 D8 Post-Stimulus Seconds
Post-Stimulus Seconds




o Females

1- LVF-Neutral
-2 + RVF-Neutral E -3


-5 1 1 1 I Di 02 D3 04 05 06 07 D8 Post-Stimulus Seconds

Figure 18. Experiment II--Second x Second Post-Stimulus Heart Rate
Changes in Female Subjects for Each Visual Field x
Stimulus Type Condition.







84



condition). In the present experiment, Ss were required to also image a personal episode of the same valence as the stimulus. The time course of the Ss' imaging could not be controlled and it may be the case that Ss differed in onset of their imaging and occurrence of affective aspects of the image.

One observations which is of some interest is the fact that

female Ss do show some indications of "autonomic patterning" in their HR changes across the 8 post-stimulus seconds. That is, females appeared to show less deceleration for fearful trials relative to other affective categories. Furthermore, this pattern appears to be reflected to a greater extent for LVF (right hemisphere) than RVF (left hemisphere) presentations.

Skin Conductance Data Reduction

For each trial, SCRs were depicted as the difference between the average of the skin conductance level during the 2 seconds preceding stimulus onset (tonic, baseline level) and the maximum skin conductance level during the 8 post-stimulus seconds (phasic level). This SCR value for each trial served as the dependent variable in a repeated measures ANOVA. Sex was the between subjects factor and Visual Field (left, right), Hand (left, right), and Stimulus Type (happy, angry, fearful, disgusting, neutral) were the within subjects factors.

Results of this ANOVA yielded no significant main effects or

interaction effects. Table 7 depicts a summary of this analysis. A weak trend, however, was observed for a Sex x Visual Field x Stimulus






85







Table 7
Summary of Analysis of Variance: Experiment II, Skin Conductance Responses


Source df SS F p


Sex 1, 28 .6410 2.60 Visual Field 1, 28 .0011 .08 Visual Field x Sex 1, 28 .0004 .03 Hand 1, 28 .0056 .13 Hand x Sex 1, 28 .0399 .94 Stimulus Type 4, 112 .0167 .36 Stimulus Type x Sex 4, 112 .0628 1.35 Visual Field x Hand 1, 28 .0007 .19 Visual Field x Hand x Sex 1, 28 .0018 .43 Visual Field x Stimulus Type 4, 112 .0664 1.61 Visual Field x Stimulus Type x Sex 4, 112 .0868 2.10 Hand x Stimulus Type 4, 112 .0186 1.31 Hand x Stimulus Type x Sex 4, 112 .0097 .69 Visual Field x Hand x Stimulus Type 4, 112 .0249 1.59 Visual Field x Hand x Stimulus Type
x Sex 4, 112 .0112 .71 -







86



Type interaction [F (4, 112) = 2.10, p = .0875)], depicted in Figure 19. This pattern of findings revealed for male Ss only a Visual Field x Stimulus Type interaction which approached significance. This pattern suggested greater SCR in LVF (M = .1302) relative to RVF (M = .0651) for angry slides. In addition, happy slides elicited greater SCR in RVF (M = .1268) compared to LVF (M = .0736) trials.







87













Trend
0.14 .137
.133
.126 .13Males Only c .125
0.12
.110
C LVF

o..107

RVF
o 0.08 .087 .C=
( .07 .065
0.06 1 1
Happy Angry Fearful Disgusting Neutral Emotion



Figure 19. Experiment II--Skin Conductance Response Analysis, Sex x
Visual Field x Stimulus Type Trend.

















DISCUSSION


Several models have been proposed to account for lateral

symmetries observed on tasks of emotional processing. One model, the right hemisphere model, suggests that the right hemisphere is globally more involved in all aspects of emotional processing including the cognitive encoding/decoding, arousal-activation, and behavioral responses to emotional stimuli. In this view, emotional stimuli presented initially to the right hemisphere via the left sensory channel (left visual field, left ear) elicit significantly greater arousal responses and result in significantly quicker/more accurate detection than emotional stimuli presented initially to the left hemisphere via the right sensory channel (right visual field, right ear).

A second model, the hemispheric valence model, proposes that the left hemisphere is more adept at processing positive emotions and the right hemisphere is more adept at processing negative emotions. Within this framework, positive emotional stimuli initially presented to the left hemisphere would elicit significantly greater arousal responses and result in significantly quicker/more accurate detection than positive emotional stimuli presented initially to the right hemisphere. Likewise, negative emotional stimuli presented to the right hemisphere would elicit significantly greater arousal responses





88






89



and result in significantly quicker/more accurate detection than negative emotional stimuli presented initially to the left hemisphere.

A third model, the preparatory model, argues that this positivenegative dichotomy in hemispheric processing of emotions is artifactual and actually relates to differences in arousal and activation/preparation for action. According to this model, the right hemisphere is dominant for mediating arousal/activation and as such, is more intrinsically involved in processing emotional stimuli that have greater "preparatory" significance for survival (i.e., fightflight emotions). In contrast, the left hemisphere is more involved in mediating nonpreparatory emotions that place less immediate attentional demands on the organism for survival.

Lastly, it is possible that the right hemisphere is dominant for mediating arousal/activation responses to stimuli irrespective of tneir emotional/nonemotional content. That is, both emotional and nonemotional stimuli should result in significantly greater arousal! activation responses when projected to the right hemisphere than the left hemisphere.

The present study sought to further investigate these different conceptualizations of the hemispheric processing of emotional stimuli. The purposes of the study were to (a) determine the extent to which laterally presented emotional/nonemotional stimuli might result in differential patterns of behavioral activation (reaction time responses) as well as differential patterns of autonomic arousal (HR, SCR responses); (b) to determine whether there were hemispheric asymmetries in mediating arousal and/or activation responses to these






90



emotional/nonemotional stimuli; and (c) to determine whether certain categories of emotion (positive/negative; preparatory/nonpreparatory) induce asymmetric arousal/activation, depending on the hemisphere to which they are initially presented.

Findings from Experiment I, in which RTs were made to midline

neutral stimuli that were preceded by lateralized stimuli of different emotional valences, failed to support any of the laterality models of emotion. For example, no overall superiority for lateralized emotional warning stimuli presented to the right versus left hemisphere was found. Likewise, no hemispheric specific emotional valence effects were observed. Similarly, no evidence was present for the view of hemispheric differences in "preparatory" versus "nonpreparatory" emotions. What was found, however, was the following: (3) females showed no late ality effects of any kind; and

(b) males, on the other hand, had overall faster RTs to neutral stimuli that were preceded by emotional warning stimuli in the RVF (left hemisphere) versus LVF (right hemisphere). This finding, which suggests that emotional stimuli induce greater behavioral activation when presented to the left hemisphere than to the right hemisphere, is the opposite of that predicted by any model arguing for superiority of the right hemisphere in mediating emotional responsivity.

There are several possibilities which might account for these findings. In Experiment I, Ss were required to make a left-right decision based on the emotional/nonemotional nature of the warning stimulus. That is, they had to respond with one hand to emotional stimuli and with the other hand to neutral stimuli. In other words,






91



Ss had to make a left-right discrimination judgement during the interstimulus interval between the WS and the imperative stimulus. It is well known that left-right discrimination seems to fall within the domain of left hemisphere functions (Benton, 1968; Gertsmann, 1940; Saugeut, Benton, & Hecaen, 1971). Consequently, it is possible that this left-right discrimination inherent in the task demands may be related to the finding of faster RTs for male Ss when stimuli were presented to the left hemisphere versus the right hemisphere. The use of a go/no paradigm using a single hand for response would circumvent this possible confounding factor.

Secondly, a variety of task strategies and stimulus factors can significantly influence the magnitude or even the direction of perceptual asymmetries (Moskovitch, 1986). Factors such as stimulus duration, spatial frequency, stimulus clarity, and number and configuration of stimulus features can significantly influence observed lateral asymmetries (Bryden, 1978; Bryden & Allard, 1976; Moskovitch, 1983; Sergent, 1983; Sergent & Bindra, 1981). For example, Patterson and Bradshaw (1975) have reported that decreasing the dimensions along which facial stimuli Iiffered decreased and even reversed the expected LVF superiority on a simultaneous face matching task.

It may be the case that in the present investigation the stimuli were sufficiently complex to warrant a change in processing strategy which relied perhaps to a greater degree on left hemispheric analysis of details and significant features in distinguishing among the different categories of stimuli. If this is the case, then the






92



significant processing demands of the warning stimulus may have mitigated any potential activation or preparatory effect that the emotional stimuli may have had because processing of the warning stimulu continued through the interstimulus interval. The use of less complex warning stimuli such as emotional and neutral faces is one way of testing this hypothesis.

In addition, it may also be of informational value to use such warning stimuli in a simple reaction time task to look at the general activation effect of emotional and neutral stimuli on simple reaction time to a neutral imperative stimuli. This paradigm would serve to eliminate the processing demands of the warning stimulus which may have interfered with the potential activation effect of emotional warning stimuli on right hemisphere processing.

Lastly, it may be the case that emotional warning stimuli used in the present investigation were not arousing enough to provide adequate activation effects in response to the imperative stimulus.

Findings the second experiment, in which measures of autonomic arousal were obtained to laterally presented emotional/nonemotional stimuli, were more in line with current views regarding hemispheric differences in processing emotional stimuli. With SCRs, there were no significant effects in terms of SCRs to lateralized emotional stimuli. However, a trend was observed (Sex x VF, p = .087), whereby male Ss had greater SCRs to happy stimuli when they were presented to tne RVF (left hemisphere) than to the LVF (right hemisphere); greater SCRs occurred when angry stimuli were presented to LVF (right hemisphere) than to the RVF (left hemisphere). Although this pattern




Full Text
RESULTS
Experiment I: Reaction Time Task
Reaction Time Responses
Subjects' reaction time responses served as the data for analysis
in a repeated measures analysis of variance (ANOVA). The between
subjects factor was Sex (male, female). The within subject factors
were Visual Field (left, right), Hand (left, right), and Stimulus Type
(happy, angry, fearful, disgusting, neutral) as within subjects
factors. Incorrect trials and trials in which eye movements occurred
were not included in the analysis. The remaining 81 of trials served
as the data for analyses. This included 83* of happy trials, 82% of
angry trials, 82% of fearful trials, 74 of disgusting trials, and 85%
of neutral trials. A log transformation was used to correct for
skewness in data distribution and to adequately meet homogeneity of
variance assumptions for the ANOVA (Kirk, 1968; Winer, 1971).
A summary of the results of this ANOVA are depicted in Table 1 .
Findings revealed a main effect of Sex [_F (1, 28) = 4.44, £ = .0442)]
with mean RT responses of male (M = 481.20) Ss significantly faster
than RT responses of female (14 = 603.48) Ss, depicted in Figure 1. A
significant Sex x Visual Field effect [F_ (1, 28) = 4.06, £ = .0535)]
was also obtained and is depicted in Figure 2. Post-hoc simple
effects testing was performed to clarify the nature of this
interaction (Kirk, 1978; Winer, 1971). Findings revealed no
58


Maximum Heart Rale Decelerlalion '3. Maximum Heart Rate Deceleration
from Baseline c from Baseline
76
Emotion
12. Experiment II--Maximum Deceleratory Heart Rate Analysis,
Stimulus Type Main Effect.
Visual Field
Figure 13. Experiment IIMaximum Deceleratory Heart Rate Analysis,
Visual Field x Sex Interaction.


REFERENCES
Ahern, G. L., & Schwartz, G. E. (1979)- Differential lateralization
for positive versus negative emotions. Neuropsychologia, 17,
693-697.
Ainsworth, M. D. S., Blehar, M., Waters, E., & Wall, E. (1978).
Patterns of attachment. Hillsdale, NJ: Erlbaum.
Alema, G., Rosadini, G., & Rossi, G. F. (1961). Psychic reactions
associated with intracarotid amytal injections and relation to
brain injury. Excerpta Medica, 37, 154-155.
Arabian, J. M., & Furedy, J. J. (1983). Individual differences in
imagery ability and pavlovian heart rate decelerative
conditioning. Psychophysiology 20, 325-331.
Ax, A. F. (1953). The physiological differentiation between fear and
anger in humans. Psychosomatic Medicine, 15, 433-442.
Babinski, J. (1914). Contribution a 1'tude des troubles mentaux dans
1'hemisplegie organique crbrale (anosognosie). Revue
Neurologique, 27, 845-848.
Bauer, R. M., *%. Craighead, W. E. ( 1979). Psychophysiological
responses to the imagination of fearful and neutral situations:
The effects of imagery instruction. Behavior Therapy, 10,
389-403.
Bayley, N. (1968). Behavioral correlates of mental growth: Birth to
36 years. American Psychologist, J_, 1-17.
Bayley, N. (1969). Bayley Scales of infant development^ Birtji to two
years. New York: Psychological Corporation.
Bear, D. M., & Fedio, P. (1977). Quantitative analysis of interictal
behavior in temporal lobe epilepsy. Archives of Neurology, 34,
454-467.
Benson, D. F., & Geschwind, N. (1971). Aphasia and related cortical
disturbances. In: A. B. Baker & L. H. Baker (Eds.), Clinical
Neurology. New York: Harper & Row.
Benton, A. (1968). Right-left discrimination. Pediatric Clinics of
North America, 15, 747~758.
109


METHOD
Subj ects
A total of 60 (30 male, 30 female) students at the University of
Florida served as subjects (Ss) in the present investigation.
Subjects were given either course credit or paid for their
participation. All Ss were right handed according to self-report and
their performance on the Briggs and Nebes (1975) Handedness
Questionnaire. An overall score of +9 or above (right hand
preference) was used as the criterion for participation in this study.
Different groups of 30 Ss each (15 male, 15 female) participated
in Experiment I and II. Subjects were randomly assigned to Experiment
I or Experiment II. Due to the potential confounding effects of
prior exposure to the emotional stimuli on subsequent RT and
psychophysiological responses, a between groups comparison was felt to
be advantageous. The mean age of Ss in Experiment I was 20.7 with a
range of 18-26 years. The mean age of Ss in Experiment II was 21.0
with a range of 19-27 years.
Experiment I: Reaction Time Task
Stimuli
Stimuli consisted of 192 black and white slides. These slides
depicted 96 neutral and 96 emotional scenes which included 24 slides
52


22
primarily in frontal and central regions (d'Elia & Perris, 1973; 1974;
Perris, 1975). These studies have also found that increased right
sided activity correlated significantly with the severity of
depression as well as performance on a verbal learning task. These
findings have been replicated by other investigators as well
(Rockford, Swartzburg, Chaudberg, 4 Goldstein, 1976). Perris (1974)
has also reported lower left to right amplitudes of visual evoked
responses in depressed patients relative to schizophrenics and normal
controls. Taken together, these findings suggest greater right
hemisphere relative to left hemisphere activity in depressed mood.
Two interpretations of these findings have been suggested. One
maintains that such hemispheric differences represent a relative left
hemisphere underresponsivenes3 (Rockford et al., 1976). A second
possibility is that such differences represent a right hemisphere
overresponsiveness (d'Elia & Perris, 1974).
Recently, subtle left sided neurological signs have been reported
in depressed patients, suggesting right hemisphere involvement. The
first report was presented by Brumback and Staton (1981). They
described two depressed children who presented with pronator drift of
the left arm, hyperreactive left deep tendon reflexes, and left
extensor plantar responses; these symptoms resolved following
treatment with tricyclic antidepressants. Similarly, Freeman,
Galaburda, Cabal, and Geschwind (1985) reported a case of a 62-year-
old depressed female who had left sided facial weakness, a gaze
preference to the right, and limited use of her left arm. Again,
these symptoms resolved following treatment with ECT.


16
production, emotional prosody judgements and production, humor
appreciation, and emotional memory.
Right Hemisphere Dominance in Regulation of Mood and Affect
In addition to studies which provide evidence that the right
hemisphere is superior for recognizing emotional aspects of
information, recent investigations have also suggested that the right
hemisphere is dominant in regulating mood and affect. These studies
fail into two major categories. The first category includes those
studies which suggest that the right hemisphere is preferentially
activated during episodes of felt emotion, primarily in normal
subjects. The second category includes those studies which correlate
psychiatric disorders of mood and affect with decrement in right
hemisphere functions.
Smotional Activation Studies
Investigations which have looked at the right hemisphere's
activation during period of felt emotion have used several different
indices of cerebral activation. These have included such measures as
electrocortical activity (usually in terms of decreased alpha power),
measures of lateral eye movements, and asymmetries of facial
expression.
Davidson and Schwartz (1976) reported that subjects showed
greater right than left hemisphere EEG activity when recalling past
events associated with anger or relaxation and during self-reported
emotional reactions to visual material (Davidson, Schwartz, Saron,


91
Ss had to make a left-right discrimination judgement during the
interstimulus interval between the WS and the imperative stimulus. It
is well known that left-right discrimination seems to fall within the
domain of left hemisphere functions (Benton, 1968; Gertsmann, 1940;
Saugeut, Benton, & Hecaen, 1971). Consequently, it is possible that
this left-right discrimination inherent in the task demands may be
related to the finding of faster RTs for male Ss when stimuli were
presented to the left hemisphere versus the right hemisphere. The use
of a go/no paradigm using a single hand for response would circumvent
this possible confounding factor.
Secondly, a variety of task strategies and stimulus factors can
significantly influence the magnitude or even the direction of
perceptual asymmetries (Moskovitch, 1986). Factors such as stimulus
duration, spatial frequency, stimulus clarity, and number and
configuration of stimulus features can significantly influence
observed lateral asymmetries (Bryden, 1978; Bryden & Allard, 1976;
Moskovitch, 1983; Sergent, 1983; Sergent & Bindra, 1981). For
example, Patterson and Bradshaw (1975) have reported that decreasing
the dimensions along which facial stimuli differed decreased and even
reversed the expected LVF superiority on a simultaneous face matching
task.
It may be the case that in the present investigation the stimuli
were sufficiently complex to warrant a change in processing strategy
which relied perhaps to a greater degree on left hemispheric analysis
of details and significant features in distinguishing among the
different categories of stimuli. If this is the case, then the


35
Similarly, Freyschuss (1970) has observed heart rate acceleration when
subjects are instructed either to tense or move an arm even though
such movement is impossible because of experimentally induced
paralysis. These observations suggest that cardiac activity is not
solely coupled with overt somatic activity per se, but that cardiac
activity is coupled with real as well as intended somatic activity.
While orienting and defensive responses result in
characteristically distinct patterns of autonomic responding, they
have both been conceptualized as arousal responses. However, this
view presents difficulty for the view of arousal as a unidimensional
phenomenon. It also provides some support for the notion that
autonomic reactivity is not as uniform as once suggested (Cannon,
1927; Schacter & Singer, 1962).
Ax (1953) provided some of the first evidence to suggest that
various affective states may be associated with distinct autonomic
patterning. Ax reported that diastolic blood pressure increased more
during anger than fear imagery, while heart rate and systolic blood
pressure increased with equal magnitude. More recent studies have
replicated these findings (Schacter, 1957; Weerts & Roberts, 1976).
Schwartz, Weinberger, and Singer (1981) have recently reported
cardiovascular differentiation between imagery induced happiness,
sadness, fear, and anger. Several investigators have also found
significantly greater heart rate accelerations in response to fearful
stimuli such as mutiliation slides, spiders, and fearful imagery (Hare
& 31evings, 1975; Klorman & Ryan, 1980; Klorman, Weissberg, &
Weisenfeld, 1977; Vrana, Cuthbert, & Lang, 1986). Recently, Ekman,


34
Lacey proposed that cardiac deceleration is associated with
environmental intake while heart rate acceleration is associated with
environmental rejection (the intake-rejection hypothesis). Lacey
suggested that cardiac deceleration served to facilitate sensory
processing while cardiac acceleration served to inhibit sensory
processing. In this view, stimuli which elicit attention and interest
are associated with cardiac deceleration and environmental intake. In
contrast, stimuli which are painful or aversive or which require a
significant amount of mental activity such as problem solving or
arithmetic are associated with cardiac acceleration and environmental
rejection. Lacey (1967; 1972) also proposed a neurophysiological
mechanism for such cardiac changes whereby cardiac responses altered
cortical activity indirectly by means of a visceral afferent feedback
loop mediated by the baroreceptors.
An alternative explanation for heart rate changes has been
offered by Obrist and colleagues (1974) who have emphasized the
relationship between motor requirements and cardiac activity.
According to Obrist et al., there is a positive correlation between
changes in cardiac activity and changes in levei of somatic activity
and both are controlled by integrative mechanisms in the central
nervous system (cardiac-somatic coupling). Obrist et al. (1974) also
noted that instances occur in which the cardiac-somatic coupling is
dissociated, whereby increases in heart rate (termed cardiac
preparatory responses) are observed without related overt changes in
somatic activity. Interestingly, this occurs specifically in
situations related to active avoidance of aversive stimuli.


72
Experiment II
Heart Rate Data Reduction
Heart rate responses were edited by a computer program which
converted all trials from interbeat interval data to beats per minute
format for each second of the sampling period. Each interbeat
interval was weighted proportionally to the fraction of the second it
occupied according to method recommended by Graham (1980). Baseline
for each trial was defined as the average HR for the 2 seconds
preceding stimulus onset. For each trial, baseline HR was then
subtracted from each of eight post-stimulus HR values, yielding second
by second post-stimulus HR changes from baseline. Trials in which eye
movements occurred were not included in the analysis. The remaining
77% of trials served as the data for analysis. This included 73? of
happy trials, 77? of angry trials, 73$ of fearful trials, 75? of
disgusting trials, and 78? of neutral trials.
From this, two separate data sets were generated. The first,
selected, for each trial, was the maximum deceleratory HR response
from the eight post-stimulus second by second HR changes from
baseline. This maximum deceleratory response then served as the
dependent variable in a repeated measures ANOVA. The second data set
selected, for each trial, was the maximum acceleratory HR response
from the eight post-stimulus second by second HR changes from
baseline. This maximum acceleratory response also served as the
dependent variable in a separate repeated measures ANOVA.


39
inaction-avoidance. This model, however, invokes both anterior-
posterior and lateral specialization of cerebral processing.
Posterior regions are proposed to provide data necessary to maintain
homeostasis and anterior regions are proposed to exert control to
stabilize as needed. In addition, anterior systems are viewed as
exerting inhibitory control over its posterior source of
information. As in the conceptualization of Fox and Davidson, this
model also proposes a left hemispheric approach and right hemispheric
avoidance dichotomy. However, in Kinsbourne's model the left
hemisphere is specialized for processing of external change and
ongoing action (approach) and the right hemisphere is specialized for
processing of internal changes, interruption of action (avoidance) as
well as control of emotional arousal. This conceptualization
establishes, in effect, four quadrants providing different mechanisms
of processing:
Left-Anterior
Left-Posterior
Right-Anterior
Right-Posterior
Action control over external change
Enables left-anterior action control to make
contact with necessary exteroceptive information
Emotional control over internal arousal
Enables right-anterior emotional control to make
contact with interoceptive information
With regard to the two control systems (action control and emotional
control), Kinsbourne argues that the two hemispheres are not in
inhibitory, but rather in compensatory interaction. Furthermore,
whether the control is under left hemisphere-approach or right
hemisphere-avoidance processing depends on the stimulus circumstances



PAGE 1

PSYCHOPHYSIOLOGICAL AND REACTION TIME RESPONSES TO LATERALLY PRESENTED EMOTIONAL STIMULI BY CYNTHIA RODRIGUES CIMINO A DISSERTATION PRESENTED TO THE GRADUATE SCHOOL THE UNIVERSITY OF FLORIDA IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF DOCTOR OF PHILOSOPHY UNIVERSITY OF FLORIDA 1988

PAGE 2

ACKNOWLEDGMENTS I would like to extend my gratitude to those who provided the encouragement and support that enabled the completion of this project. First, I am grateful to my dissertation committee. I would like to thank my chair, Dr. Dawn Bowers, for her time, hard work, and patience over the years; for her skill at making me really think; and especially for her belief in my ability. I am also grateful to Dr. Kenneth Heilman for providing material and moral support and for sharing his knowledge and never-ending enthusiasm, creativity, and wonderment. I thank Dr. Rus Bauer for his advice on psychophysiological technique and statistical method and for his infrequent but apt clinical interpretations. I would also like to tnank Dr. Eileen FenneLl for ner discerning comments on methodology, her ground-rooted advice on professional development, and her kindness. I am also grateful to Dr. Ed Valenstein for his support in my defense and for his thought-provoking question at the end of my dissertation copy (humble as always). Finally, I am grateful to Dr. Hugh Davis for so many things, but especially for his guidance in development of my clinical abilities, his warmth and playfulness, and his love of language and verbal tapestries. I owe a debt of gratitude to Cindy Zimmerman for her skill in organizing the typing and completion of the manuscript. My appreciation also goes to Dr. Roger Blashfield for his unconditional li

PAGE 3

3upport and friendship and for his honest and direct means of pushing tne to grow, personally and professionally. I thank Mieke Verfaeilie and Karen Froming for their solid support and friendship over the years. With love and deepest appreciation, I also thank my husband, Pat, for his continued encouragement and enduring love and support.

PAGE 4

TABLE OF CONTENTS Page ACKNOWLEDGMENTS ii ABSTRACT vi INTRODUCTION 1 Right Hemisphere Superiority for Recognizing Emotional Aspects of Stimuli 3 Emotional Prosody Studies 3 Affective Faces Studies 8 Related Research 13 Right Hemisphere Dominance in Regulation of Mood and Affect.... 16 Emotional Activation Studies 16 Mood and Affect Studies 20 Left Hemisphere Superiority for PodLtive Affect; Right Hemisphere Superiority for Negative Affect 25 Mechanisms of Emotional Processing: The Role of Arousal 31 Neuropsychological Models of Emotional Processing 36 The Model of Fox and Davidson 37 Kinsbourne's Model 38 Tucker 's Model 40 Heilman's Model 42 Critical Issues 45 Hypotheses and Predictions 49 METHOD 52 Subjects 52 Experiment I: Reaction Time Task 52 Stimuli 52 Apparatus 53 Procedure 54 Experiment II: Psychophysiological Responses to Laterally Presented Emotional Material 55 Stimuli 55 Apparatus 55 Procedure 57 iv

PAGE 5

RESULTS 58 Experiment I: Reaction Time Task 58 Reaction Time Responses 58 Percent Correct Responses 67 Experiment II 72 Heart Rate Data Reduction 72 Skin Conductance Data Reduction 84 DISCUSSION 88 Critical Issues 95 Conclusions 1 05 REFERENCES 109 BIOGRAPHICAL SKETCH 127 V

PAGE 6

Abstract of Dissertation Presented to the Graduate School of the University of Florida in Partial Fulfillment of the Requirements for the Degree of Doctor of Philosophy PSYCHOPHYSIOLOGICAL AND REACTION TIME RESPONSES TO LATERALLY PRESENTED EMOTIONAL STIMULI BY CYNTHIA RODRIGUES CIMINO December, 1988 Chair: Dawn Bowers Major Department: Clinical and Health Psychology Neuropsychological investigations of brain injured and neurologically Intact subjects have suggested that the two hemispheres differ in terms of their contribution to emotional processing. Several different models have been proposed to account for these observed hemispheric asymmetries. One model, the right hemisphere emotion model, suggests that the right hemisphere (RH) is globally involved in all aspects of emotional processing. A second model, the hemispheric valence model, suggests that the left hemisphere (LH) is dominant for processing positive emotions, whereas the RH is dominant for processing negative emotions. A third model, the preparatory model, suggests the RH is dominant for mediating arousal/activation and, as such, is more intrinscally involved in processing emotional stimuli that have greater "preparatory" significance for survival (i.e., fight-flight emotions such as anger and fear). In contrast, the LH is more involved in mediating nonprepar atory emotions (i.e., happiness, sadness, disgust) that place less immediate or "phasic" vi

PAGE 7

demands on the organism for survival. Alternatively, it is possible that the RH may be dominant for mediating arousal/activation responses to stimuli, regardless of their emotional/nonemotional content. The focus of the present study was to further examine these different conceptualizations of the hemispheric processing of emotional stimuli in neurologically intact male and female subjects. This was accomplished in two separate experiments: (a) a choice reaction time task was used to investigate subjects' activation responses to a centrally presented, neutral stimuli .vhen it was preceded by neutral or emotional warning stimuli and (b) heart rate (HR) and skin conductance (SC) measures were used to investigate subjects' responses to laterally presented neutral and emotional stimuli. Findings from Experiment I, the reaction time experiment, failed to support any of the laterality models of emotion. Findings from Experiment II, using HR and SC measures, were more congruent with models of hemispheric differences in processing of emotional stimuli. Males showed lateralized effects of HR arousal responses which support greater RH involvement in production of arousal responses. However, this effect was present regardless of the emotional/nonemotional content of the stimulus. Skin conductance responses in male subjects did provide some support for hemispheric specific emotional valence effects but this did not reach statistical significance. In contrast, to the hemispheric asymmetries in arousal responses observed in male subjects, female subjects did not show significant differences in arousal responses across left and right visual fields. However, females were differentially impacted by the vi i

PAGE 8

emotional valence of the stimuli. Female subjects showed autonomic patterning effects across emotional categories, a finding which may be accounted for by imagery differences across male and female subjects. Taken together, these findings point to the relative importance of considering both sex and imagery ability of subjects in future investigations of emotional processing and autonomic responding vi ii

PAGE 9

INTRODUCTION Neuropsychological approaches to the investigation of emotional processing have evolved, in large part, from early clinical observation of brain injured patients and subsequent systematic investigation of their performance on a variety of emotional tasks. One of the earliest reports was provided by Babinski (191'^) in which he noted that patients with right hemisphere damage appeared indifferent or euphoric. Denny-Brown, Meyer, and Horenstein (1952) also reported evidence of such "indifference" reactions after right hemisphere lesion and noted its co-occurrence with unilateral neglect syndrome in whicn patients failed to orient, report, or respond to the left side of their body. In 1952, Goldstein published his observation that "catastrophic" emotional responses often accompanied left hemisphere damage. These reports were later corroborated by Hecaen (1962), who also noted that catastrophic reactions most often followed left hemisphere insult whereas indifference reactions were more frequent following right hemisphere damage. In 1972, Gainotti reported a large scale study of 160 patients who had sustained left or right sided lesions. Based on systematic observation of the frequency and type of symptomatology, Gainotti reported a consistent relationship between behaviors indicative of a 1

PAGE 10

2 depressive-catastrophic reaction and left hemisphere damage (LHD). Behaviors of emotional indifference and minimization of deficits were associated with right hemisphere damage (RHD). Similarly, Terzian (196^) and Rossi and Rosadini (1967) reported findings from unilateral carotid injection of sodium amytal and found depressive-catastrophic reaction following left sided injection and inappropriate euphoria following right sided injection. Milner (cited in Rossi & Rosadini, 1967) attempted to replicate these findings without success. In 1982, Sackeim and colleagues reported 119 cases of pathological laughing and crying in response to unilateral lesions. Results were congruent with Gainotti's findings with laughing outbursts more frequent following RHD and crying more often following LHD. Goldstein (1948) as well as Gainotti (1972) have interpreted such depressive-catastrophic reactions following LHD as a "normal" response to a significant loss of physical as well as psychological function. In contrast, emotional changes seen with RHD were interpreted by Gainotti (1972) as an abnormal response associated with anosognosia or denial of illness. More recent interpretations of such findings, however, have suggested that the indifference reaction may be due to the RHD patient's inability to accurately comprehend and/or express affect. This led to the hypothesis that the right hemisphere, in the intact state, is superior for the perception and/or expression of emotional material In the years which followed, more systematic investigation was applied to the understanding of the right hemisphere's role in the

PAGE 11

3 processing of emotional stimuli. Two major areas of research have addressed this specific question: those which examine the processing of the prosodic elements of speech and those which examine the processing of affective faces. Related areas of investigation will also be discussed. Right Hemisphere Superiority for Recognizing Emotional Aspects of Stimuli Emotional Prosody Studies It is well known that in right handers, the left hemisphere is more adept than the right hemisphere in decoding the linguistic content (semantic and phonemic elements) of speech (Benson & Geschwind, 1971). However, speech may carry at least two levels of information content: the linguistic content which conveys what is said and the prosodic content which conveys the way in which it is said. Prosodic elements which are defined as pitch, tempo, and rhythm, carry information about the emotional as well as nonemotional content of prosodic speech (Paul, 1909). Nonemotional prosody is important for conveying whether a sentence is a question, a statement, or a command. Emotional prosody is critical for conveying affective information In 1975, HeiLman, Scholes, and Watson studied the ability of right temporo-parietal and left temporo-par ietal damaged patients to identify affective prosody. Patients were presented with semantically neutral sentences which were read in one of four emotional tones — happy, sad, angry, or indifferent. In this study, the subject's task

PAGE 12

was to identify the emotional tone of the speaker. Results demonstrated that the RHD performed significantly worse on this task than the LHD group. In a subsequent study, Tucker, Watson, and Heilman (1977b) investigated patient's ability to discriminate between different affectively toned sentences. In this task, subjects were not required to identify the affective tone but to discriminate between sentences presented with the same or different affective tone. RHD patients, again, performed more poorly than LHD patients, providing further evidence for the right hemisphere's greater role in the processing of emotional aspects of speech. Weintraub, Mesulam, and Kramer (1981) have suggested that RHD patients have difficulty in both emotional and nonemotional aspects of prosodio speech. In their study, RHD patients had significantly greater difficulty in distinguishing whether filtered sentences were questions, commands, or statements. Based on their findings, Weintraub et al suggested that the poor performance of RHD patients on emotional prosody tasks may be accounted for by a more general deficit in the processing of prosodic information. No LHD group was reported in this investigation. In a subsequent study which addressed this question, Heilman, Bowers, Speedie, and Coslett (1984) investigated the ability of RHD and LHD patients to comprehend filtered sentences which contained either emotional (happy, sad, angry) or nonemotional (declarative, imperative, or interrogative) prosody. Results demonstrated that both RHD and LHD patients were impaired on the nonemotional prosody task

PAGE 13

5 compared to control subjects. In contrast, RHD patients performed significantly worse on the emotional prosody task relative to LHD patients, suggesting a greater role for the right hemisphere in the comprehension of emotional prosody. In addition to these reports of deficits in the comprehension and discrimination of affective prosody. Tucker et al (1977b) also found that RHD patients had difficulty in producing affectively intoned speech. Patients were asked to say a semantically neutral sentence using either a happy, sad, angry, or indifferent tone. RHD patients performed significantly worse than LHD patients suggesting that their deficits include not only the comprehension and discrimination of affectively intoned speech but also the expression of affectively intoned speech. This finding was later supported by Ross and Mesulam (1979) who reported two patients who could not express affectively intoned speech but could comprehend affective speech. In addition, Ross (1981) has also reported patients who could not comprehend affective intonations but could repeat affectively intoned speech. Ross has suggested that the right hemisphere may mediate the comprehension, repetition, and production of affective speech much in the same way as the left hemisphere does for prepositional speech with anterior lesions producing primarily production defects and posterior lesions producing primarily comprehension defects. With the advent of experimental procedures such as tachistoscopic presentation and dichotic listening in which stimulus processing is initially restricted to the left or right hemisphere, investigations

PAGE 14

6 of emotional processing in normal, neurologically intact subjects have also examined the right hemisphere's role in the processing of emotional prosody. Using a dichotic listening (DL) procedure, Haggard and Parkinson (1971) paired speech babble with short sentences spoken in one of four emotional tones. They found that accuracy in identifying the emotional tone was significantly better on left ear trials, suggesting a right hemisphere superiority. Similarly, Safer and Leventhal (1977) used monoaural presentation of sentences with positive, negative, and neutral content spoken in a positive, negative, or neutral tone. Results demonstrated that subjects who listened to sentences in their left ear tended to use the intonation in making their judgements whereas subjects who listened to sentences in their right ear tended to use the content of the sentence in making judgement. Interpretation of these findings, however, has been questioned because of the use of a between groups comparison for each ear presentation. The findings are suggestive, nevertheless. Ley and Bryden (1982), using a DL procedure, had subjects report on the emotional tone and content of a sentence arriving at either the left or right ear (specified on each trial). Subjects were more accurate in judging the emotional tone of the sentence when monitoring the left ear and more accurate in judging the content of the sentences when monitoring the right ear. In a subsequent experiment, Bryden, Ley, and Sugarman (1982) investigated hemispheric differences in ability to judge the emotional tone of musical stimuli. They did this by taking advantage of the fact that in Western culture music written in a major key is generally

PAGE 15

7 described as happy while music written in a minor is more often described as sad (Davies, 1978). In a DL procedure, subjects were required to identify the emotional tone of a short seven-note passage while monitoring either the left or right ear. Subjects were more accurate when identifying the emotional tone of passages presented to the left ear relative to those presented to the right ear, again supporting the notion of the right hemisphere dominance in processing of emotional stimuli. Dichotic listening procedures in normal subjects have also demonstrated left ear advantages in recognition of other nonverbal, emotional aspects of human speech such as laughing and crying (Carmon & Nachson, 1973; King & Kimura, 1972). In a study which used a variant of the monoaural paradigm in which subjects heard spoken captions and laughter in either the left or right ear, cartoons were judged as funnier when the Laughter was heard by the left ear relative to the right ear (DeWitt, 1978). Recently, Mahoney and Sainsbury (1987) investigated hemispheric asymmetries in perception of human, nonspeech emotional sounds. During conditions of divided attention, a left ear advantage emerged during the second block of trials. Under conditions of selective attention, however, this left ear advantage was seen on the first olock of trials. In addition to providing support for a right hemisphere superiority in processing of emotional nonspeech sounds, these findings also suggest that effects of attention influenced the rate and development of observed laterality effects but not the direction of these effects.

PAGE 16

8 In summary, investigations in both normal and brain impaired subjects have provided evidence that the right hemisphere is preferentially involved in the comprehension and expression of emotional prosodic elements of speech and other nonverbal vocalizations. Similarly, a large body of literature has also investigated the role of the right hemisphere in the processing of affective faces. Affective Faces Studies In 1980, DeKosky, Heilman, Bowers, and Valenstein reported a study which investigated RHD and LHD patients' ability to make neutral facial discriminations as well as affective facial discriminations. They found that RHD patients performed more poorly than LHD patients on both facial affect judgements as well as neutral facial discrimination. In fact, when the two groups were statistically equated for performance on the neutral discrimination task, differences between RHD and LHD patients on the affective facial discrimination task disappeared. These findings suggested that RHD patients' poor performance on facial affect judgements can be solely accounted for by their poor performance in facial discrimination ability. This has led to the question of whether processing the emotionality of a face, in fact, involves a "stimulus-content dimension" in its own right or whether such processing merely involves an increase in the conf igurational complexity of the stimulus and consequently increases the demand on right-hemisphere mediated visuospatial skills.

PAGE 17

9 More recent investigations have challenged this latter interpretation. Support for such a dissociation has recently been provided in a study by Kolb and Taylor (1981). Their findings revealed that patients with parietal excision were impaired on both matching of facial affect and matching of facial identity. However, these authors also report that patients with damage restricted to right temporal and right frontal regions were more impaired on processing of facial affect relative to processing of facial identity. Similarly, a study by Freid, Mateu, Ojemann, Wohns and Fedio (1982) reported that neutral facial matching was disrupted by stimulation of right parietal-occipital regions, whereas naming the affect depicted on pictures of faces was disrupted by stimulation of right posterior middle temporal gyrus. In contrast, Cicone, Wapner and Gardner (1980) found no relationship between RHD patients' performance on an emotional perception task and a facial identity task, suggesting that deficits associated with RHD cannot be accounted for by facial recognition deficits alone. More recently, Bowers and Heilman (1984) reported a case which demonstrated a dissociation between processing of affective and noneffective faces. Although this RHD patient was able to perform well on neutral facial tasks and on same-different emotional faces tasks, he was impairea on naming and comprehension of verbal labels for facial expressions. In a subsequent investigation. Bowers, Bauer, Coslett, and Heilman (1985) again addressed the question of whether defects shown

PAGE 18

10 by RHD patients on facial affect tasks are dissociable from defects in visuoperceptual processing. They cite two criticisms of the DeKosky study which directly address this issue. First, they noted that a small subset of RHD patients in the DeKosky study performed normally on the neutral visuoperceptual task, yet, were impaired on the facial affect tasks. This suggested that visuoperceptual deficits alone cannot account for impaired processing of affective faces in all RHD patients Secondly, they suggested that the use of same actors on affective faces trials may have allowed subjects to rely on a pure template matching strategy in which judgements about emotionality could have been made on the basis of whether the two faces had the same physiognomic configuration. A defect in this type of perceptual process could then potentially affect performance on both facial identity as well as affective facial tasks. As an alternative, Bowers et al required subjects to make affective facial judgements across different actors so that such judgements would take place in an "associative" context with less reliance on potential defective perceptual matching. Results of this study revealed that when patient groups were statistically equated on visuoperceptual ability (facial identity task), RHD patients still performed worse than LHD patients and control subjects on (a) emotional discrimination of different actors, (b) naming the emotion of a single face, and (c) picking the named emotion from four pictures of the same actor. These findings provide strong evidence that differences in LHD and RHD patients' abilities to

PAGE 19

1 1 make affective judgements cannot be accounted for solely by differences in the visuoperceptual processes underlying facial identity discrimination. Recent support for the relative dissociation of facial identity judgements from facial affect judgements has also been provided by Tranel Damasio, and Damasio (1988). Tranel et al. described four patients with bilateral lesions of occipitotemporal or temporal regions whose performance on facial affect tasks were significantly better than their performance on facial identity tasks. Research in normal subjects has also investigated the role of the right henisphere in the processing of affective faces. Ley and Bryden (1979) tachistoscopically presented faces to the left visual field (LVF) and right visual field (RVF), and subjects made either facial identity judgements or facial affective judgements. They found a LVF superiority for both tasks. However, when performance on the facial affect task was reanalyzed using performance on the facial Identity task as a covarlate, tne LVF superiority for making affective facial judgements remained. These findings, which are similar to those of Bowers et al (1985), suggest that the right hemisphere superiority for processing facial affect exists above and beyond the superiority for processing facial identity. In 1977, Suoeri and McKeever reported a reaction time (RT) task in which subjects responded to previously memorized emotional or nonemotional faces presented in LVF or RVF. Subjects who memorized emotional faces showed significantly faster reaction times to LVF targets than subjects who memorized nonemotional faces. Based on

PAGE 20

12 these findings, Suberi and McKeever argued that the LVF effect for processing nonemotional faces is significantly enhanced by presentation of emotional faces. McKeever and Dixon (1981) used emotional imagery and neutral faces to investigate right hemisphere effects in processing of affective material. They instructed subjects to imagine that something very sad happened to a number of predetermined target faces. In a subsequent target/nontarget discrimination task with lateralized presentations, they report that the use of emotional imagery significantly enhanced LVF (right hemisphere) performance. This effect, however, was demonstrated in female subjects only. Safer (1981) reported a study in which subjects memorized faces by either empathizing with their emotional expressions or by labeling the emotional expressions. Results demonstrated that subjects who used empathy recognized more faces presented to the LVF than RVF No laterality effect was demonstrated for those who labeled faces. This laterality effect for the empathy condition, however, was found for male subjects only. Similarly, Buchtel Campari, DeRisio, and Rota (1978) reported faster responding to both positive and negative stimuli presented in LVF relative to neutral targets. Hansch and Pirozzolo (1980) and Strauss and Moskovitch (1981) also reported a LVF effect for neutral and emotional faces. In summary, investigations in both normal and brain impaired subjects have supported the notion that the right hemisphere is preferentially involved in the processing of affective faces. In addition, several studies have also suggested that right hemisphere

PAGE 21

13 advantage for processing of facial affect exists above and beyond the right hemisphere's advantage for processing of facial identity. Evidence about possible differences due to sex of subjects, however, remains equivocal. Related Research Several studies have implicated the right hemisphere in memory for emotionally charged materials. Weschler (1973) reported one of the few studies of emotional memory in brain impaired subjects. Right hemisphere and LHD patients were presented with two types of stories-one emotional and the other nonemot ional When asked for subsequent recall, RHD subjects made significantly more errors in recalling emotional stories relative to LHD patients. Cimino, VerfaeLiie, Bowers, and Heilman (1988) investigated whether RHD patients have difficulty remembering past affective episodes by asking them to recall prior emotional and neutral experiences. Findings revealed that RHD patients produced significantly less emotional reports than control subjects as judged by independent raters. However, their own emotionality ratings were no different from those of control subjects suggesting some discordance between their actual production of emotional memories versus their own perceived emotionality of such memories. Unfortunately, most patients with LHD are aphasic and could not be used in this study. Therefore, this report cannot conclude that this defect is specific to RHD. An investigation in normal subjects (Gage & Safer, 1979) looked at hemispheric differences in mood-state dependent effects for

PAGE 22

11 recognit:!!! of emotional faces. The mood-state dependent effect states that stimuli will be better recalled when subject's mood at encoding and retrieval are congruent than when they are disparate (Bower, 1981). Using a mood induction procedure, Gage and Safer demonstrated that recognition of faces initially presented in a discrepant mood was significantly worse when presented to the LVF (right hemisphere) than when presented to the RVF (left hemisphere). Based on these observations, the authors suggest that the right hemisphere stores the subject's mood as an integral part of the memory representation to a greater extent than the left hemisphere. Although it is well known that the left hemisphere is specialized for mediating speech and linguistic stimuli in the vast majority of right handed adults, the right hemisphere also appears to have some reduced language capacity (Geschwind, 1969; Papanicolaou, Moore, Deutsch, Levin, & Eisenberg, 1988). Several recent studies have suggested that the right hemisphere may better process emotional versus nonemotional verbal stimuli, at least at the single word level. For example. Graves and coworkers (Graves, Landis, 4 Goodglass, 1980) found that aphasic patients with alexia due to left hemisphere lesions could read emotional words better than nonemotional words. In a subsequent study, these same investigators (Graves, Landis, & Goodglass, 1981) found that neurological ly intact males subjects better recognized emotional words than nonemotional words when these stimuli were presented to the LVF (right hemisphere). Other researchers (Brody, Goodman, Holm, Krinzman, & Sebrechts 1987) have looked at the effects of lateralized affective priming

PAGE 23

15 stimuli (faces/words) on subsequent judgements of the affective value of laterally presented emotional and nonemotional verbal target stimuli. They found that affective primes presented to the RVF (left hemisphere) resulted in decreased accuracy judgements of the target stimuli that were also presented to that hemisphere. In contrast, affective primes presented to the LVF (right hemisphere) resulted in increased accuracy judgements regarding the affective value of verbal stimuli presented to the right hemisphere. Taken together, findings with both aphasic and neurologically intact individuals suggest a right hemisphere advantage in processing emotional verbal stimuli. However, this view is not entirely clearcut in that other studies have failed to replicate the "right hemisphere" laterality effect for identifying emotional versus nonemotional verbal stimuli (Strauss, 1983). Another area of investigation concerning the role of the right hemisphere in processing of emotional stimuli Is that of humor appreciation. Brownell et al (1984) have recently reported that RHD patients have significant difficulty in understanding narrative humor as portrayed in short story joKes. Similarly, Bihrle et al (1986) have reported that RHD patients performed significantly worse than LHD patients on a nonverbal cartoon completion task. In summary, a large body of literature suggests that the right hemisphere plays a greater role in the comprehension and expression of emotional information. These findings have been consistently demonstrated in both neurologic as well as normal populations across a wide range of tasks including affective facial judgements and

PAGE 24

16 production, emotional prosody judgements and production, humor appreciation, and emotional memory. Right Hemisphere Dominance in Regulation of Mood and Affect In addition to studies which provide evidence that the right hemisphere is superior for recogni zi ng emotional aspects of information, recent investigations have also suggested that the right hemisphere is dominant in regulat ing mood and affect. These studies fall into two major categories. The first category includes those studies which suggest that the right hemisphere is preferentially activated during episodes of felt emotion, primarily in normal subjects. The second category includes those studies which correlate psychiatric disorders of mood and affect with decrement in right hemisphere functions. Emotional Activation Studies Investigations which have looked at the right hemisphere's activation during period of felt emotion have used several different Indices of cerebral activation. These have included such measures as electrocortical activity (usually in terms of decreased alpha power), measures of lateral eye movements, and asymmetries of facial expression Davidson and Schwartz (1976) reported that subjects showed greater right than left hemisphere EEC activity when recalling past events associated with anger or relaxation and during self-reported emotional reactions to visual material (Davidson, Schwartz, Saron,

PAGE 25

17 Bennett, & Golemena, 1979). Right hemisphere activation has also been reported during hypnotically induced depression (Tucker, Stenslie, Roth, & Shearer, 1981), during generation of emotional imagery and during painful stimulation (Karlin, Weinapple, Rochford, & Goldstein, 1979). In addition to comparisons of left versus right sided activation, several authors also emphasize the importance of relative differences in level of activation in anterior versus posterior regions. Tucker (1981) reported frontal activation but not posterior activation in depressed mood. Likewise, Davidson et al (1979) reported that mood valence varied with right versus left activation in frontal regions, but that posterior regions showed right hemisphere activation irrespective of valence. Lateral eye movements (LEM) as indices of hemispheric activation have also been used to assess the role of the right hemisphere in regulation of mood and affect. Prior investigations have revealed a tendency toward right LEM (left hemisphere activation) with verbal processing and left LEM (right hemisphere activation) with visuospatial processing (Kinsbourne, 1972). Schwartz, Davidson, and Maier (1975) have reported a greater frequency of left LEM in subjects performing emotional versus neutral mental tasks. Similarly, Borod, Vingiano, and Cytryn (1988b) measured LEM while subjects were asked to generate emotional images of positive and negative valence in auditory, visual, and tactile modalities. Overall, subjects looked significantly more to the left than to the

PAGE 26

18 right in response to emotional instructions. These findings suggest a greater role of the right hemisphere in generating emotional imagery. In addition. Tucker, Roth, Arneson, and Buckingham (1977a) have reported more left LEM in anxious than nonanxious subjects. Woods (1977) has also suggested that habitual left eye movers are higher in intensity and frequency of emotional reactions than right eye movers. While findings from LEM studies may appear to be conceptually apparent, interpretations of findings from such investigations must be cautioned in terms of the questionable reliability and validity of LEM as indicators of hemispheric activation. Berg and Harris (1980) were unable to replicate previous findings in LEM studies and concluded that the validity of the LEM procedure as a measure of hemispheric activation has yet to be established. Ehrlichman and Weinberger (1978) in a detailed review of the LEM literature, similarly concluded that the use of LEM in investigations of hemispheric functioning was questionable at best. Recently, lacrimal flow has also been utilized as an index of hemispheric involvement in production of mood states. Delp and Sackeim (1987) looked at lacrimal flow following sadness and happiness mood manipulation in male and female subjects. For female subjects, the sadness manipulation resulted in greater relative left eye lacrimal flow, whereas the happiness manipulation resulted in a shift toward greater relative reduction in left eye flow. Although these findings may be interpreted as support for greater right hemisphere involvement in lacrimal flow, this interpretation must be observed with some caution as the assumed lateralization of specific

PAGE 27

19 neuroanatomical pathways regulating lacrimal flow have not been clearly established. Measures of facial asymmetries observed during emotional expression have also provided support for the hypothesis of right hemisphere dominance in regulating affect and mood states. Musculature of the lower part of the face is contralaterally innervated and asymmetries observed with respect to facial expression may be used to infer relative hemispheric involvement in production of emotional expression. Studies with normal subjects have revealed that the left hemiface moves more extensively during posed facial expression (Borod & Caron, 1980; Borod, Caron, ^ Koff, 1981; Borod, Kent, Koff, Martin, & Alpert, 1988a; Borod, Koff, & White, 1983; Moskovitch &01ds, 1982). One criticism of such studies, however, is that they have used posed facial expressions which may not actually reflect the underlying affect or mood of the subject. In response to this criticism, several authors have regarded spontaneous facial expression as a more valid index of the subject's affective state. Ekman, Hager, and Friesen (1981) failed to find such asymmetries of emotional expression during spontaneous facial expressions. In contrast, other investigators (Borod et al 1983; Moskovitch & Olds, 1982) have observed greater left sided (right hemisphere) involvement for both spontaneous and posed facial expressions. Several studies have also used composite photographs in which the mirror image of the left or right half of the face is combined with the original ipsilateral image. This process results in a complete

PAGE 28

20 facial photo composed entirely of the left or right half of the face. These studies report that left sided composites of posed facial expressions are judged as more intense than right sided composites (Heller & Levy, 1981; Rubin & Rubin, 1980; Sackeim, Gur & Saucy, 1978). This effect has also been reported for spontaneous expressions as well (Dopson, Beckwith, TucKer, & Bullard-Bates 1984). Clinical reports have indicated that spontaneous emotional facial expressions are less likely to occur in RHD patients compared to LHD patients (Borod, Koff Perliman, Lorch, & Nicholas, 1986; Buck & Duffy, 1980; Ross & Mesulam, 1979). Kolb and Miiner (1981), however, were unable to find differences between RHD and LHD patients in spontaneous facial expressions and movements (only some of which were emotional). They did find, however, differences with respect to the anterior versus posterior distribution of the lesion, with anterior lesions resulting in less spontaneous movements than more posterior lesions. Borod et al (1986) recently reported the only systematic investigation, to date, of posed and spontaneous facial expressions in brain impaired patients. They found that RHD patients produced fewer posed as well as spontaneous emotional facial expressions than did LHD patients Mood and Affect Studies A second group of studies has also examined disorders of mood and affect associated with right hemisphere function. These investigations have studied mood and affective changes in patients with known hemispheric pathology. Patient groups have included

PAGE 29

21 epileptic patients with unilateral foci, head injury patients, and patients with unilateral subcortical infarction or surgical removal. In a factor analytic study which assessed interictal personality changes associated with right temporal lobe epilepsy, Bear and Fedio (1977) reported that right foci were more associated with affective changes while left foci were more associated with cognitive changes. Similarly, Taylor (1972) described a predominance of right sided foci in a sample of epileptics with associated symptoms of depression, anxiety, and phoDias. In 1969, Flor-Henry reported that of patients with unilateral epileptic foci, manic-depressive psychosis was found twice as many times in individuals with right sided foci than in those with left sided foci Lishman (1968), in his review of 1 cases of head injury, found that affective disturbances were more common following right hemisphere injury while cognitive/intellectual changes were more often seen following left sided injury. These findings parallel those of Bear and Fedio (1977) which looked at changes associated with left and right epileptic foci. Recently, mania has also been reported following right thalamectomy (Whitlock, 1982) and right thalamic infarct (Cummings & Mendez, 1984). Investigators have also looked at signs of hemispheric dysfunction in patients with primary mood disturbance as a means of investigating lateralized hemispheric functioning in relation to disorders of mood and affect. Several investigators, utilizing EEC as a measure of hemispheric activity, have reported low left versus right hemisphere activity in depressed patients with differences occurring

PAGE 30

22 primarily in frontal and central regions (d'Elia & Perris, 1973; 197'^; Ferris, 1975). These studies have also found that increased right sided activity correlated significantly with the severity of depression as well as performance on a verbal learning task. These findings have been replicated by other investigators as well (Rockford, Swartzburg, Chaudberg, & Goldstein, 1976). Perris (1974) has also reported lower left to right amplitudes of visual evoked responses in depressed patients relative to schizophrenics and normal controls. Taken together, these findings suggest greater right hemisphere relative to left hemisphere activity in depressed mood. Two interpretations of these findings have been suggested. One maintains that such hemispheric differences represent a relative left hemisphere underresponsi veness (Rockford et al 1976). A second possibility is that such differences represent a right hemisphere overresponsi veness (d'^lia & Perris, 197^). Recently, subtle left sided neurological signs have been reported in depressed patients, suggesting right hemisphere involvement. The first report was presented by Brumback and Staton (1981). They described two depressed children who presented with pronator drift of the left arm, hyperreactive left deep tendon reflexes, and left extensor plantar responses; these symptoms resolved following treatment with tricyclic antidepressants. Similarly, Freeman, Galaburda, Cabal, and Geschwind (1985) reported a case of a 62-yearold depressed female who had left sided facial weakness, a gaze preference to the right, and limited use of her left arm. Again, these symptoms resolved following treatment with EOT.

PAGE 31

23 Findings from investigations of galvanic skin responses (GSR) have provided only partial support for the presence of lateralized dysfunction in depressed patients. Schneider (1983) reported lower right handed GSRs in a depressed sample. In addition, Myslobodsky and Horesch (1978) have noted higher left handed GSRs in depressed subjects. Toone Cooke, and Lader (1981), however, were unable to replicate these findings. Such discrepancies may be accounted for by conflicting reports which suggest that GSR responses may be controlled ipsilaterally contralaterally or bilaterally (Holloway & Parsons, 1969; Lacroix & Comper, 1979; Myslobodsky 4 Rattok, 1977). The performance of depressed patients on tests likely to require greater right hemisphere processing has also been investigated. Several studies sugggest that depressed subjects perform more poorly on visuospatial tasks than verbal tasks (Flor-Henry, 1976; 1983; Goldstein, Filskov, Weaver, & Ives, 1977; Kronfol Hamsher Digre, & Wazir, 1978). Siiberman, Weingartner, and Post (1983ai have also suggested that the pattern of errors in depressed subjects closely resembles that of right temporal lobectomized patients and that the degree of impairment is correlated with the overall severity of depression. However, as cautioned by Weingartner and Siiberman (1982), the impaired verbal learning and memory performance that is frequently observed in depressed patients also points to left hemisphere dysfunction. Because of decrements observed on left as well as right hemisphere tasks, Weingartner and Siloerman (1982) suggested that it is best to describe the deficits observed as a

PAGE 32

24 relative impairment on right hemisphere tasks as compared to left hemisphere tasks. Several studies have also reported the occurrence of "reversed lateralization" or "functional delateralization" in depressed patients on tasks of verbal and nonverbal processing. Bruder (1983) published a review of dichotic listening studies and concluded that depressed patients demonstrated decreased lateralization on both verbal as well as nonverbal dichotic listening tasks. Several authors, however, have suggested that decreased lateralization is present primarily on nonverbal tasks (Coulbourn & Lishman, 197^; Johnson & Crockett, 1982). Evidence of reversed lateralization has also been reported by some authors. Silberman, Weingartner, Stillman, Chen, & Post (1983b) have reported a left visual field superiority on a verbal task in a sample of depressed females. Similarly, research in depressed patients has also found (Flor-Henry, 1979; Flor-Henry & Koles, 1980) increased parietal activity during rest, witn left temporal activation during spatial tasks and right parietal activation during verbal tasks. These findings are at variance with predicted asymmetries in normal populations. Hommes and Panhuysen (1971), using a small sample of depressed patients, have reported that right sided sodium amytal injections resulted in a degree of aphasia for all subjects. Furthermore, this finding was significantly correlated with the severity of depressive symptomatology. In summary, a large body of the neuropsychology literature on emotional processing has suggested a right hemisphere dominance in the comprehension and expression of emotional information. These findings

PAGE 33

25 have been consistently demonstrated in neurologic and psychiatric, as well as normal, populations using a range of tasks including affective facial judgements and production, emotional prosody judgements and production, emotional memory, emotional verbal judgements, production of affect, and humor appreciation. Left Hemisphere Superiority for Positive Affect; Right Hemisphere Superiority for Negative Affect Early clinical reports of emotional/mood changes following brain injury have also been interpreted as supporting a distinction between the superiority of the left hemisphere in the processing of positive affect and the superiority of the right hemisphere In the processing of negative affect. Interpretation of these studies Is based on the notion of reciprocal inhibition which states that each hemisphere exerts some degree of inhibition on the contralateral hemisphere (.Kinsbourne 1973). In this view, damage to the left hemisphere would result in dis inhibit ion of the right hemisphere's negative affective bias. Right hemisphere lesions would result in dlsinhibition of the left hemisphere's positive affective bias. Early observations suggested that RHD patients often appeared indifferent or euphoric (Babinski, 1914; Denny-Brown et al 1952). In contrast, several authors reported that left hemisphere damage was more associated with a "catastrophic" emotional response (Goldstein, 1952; Hecaen, 1962). These observations were later corroborated by Gainotti (1972) in a large scale study of 160 patients. Based on systematic investigation of patients' symptomatology, Gainotti

PAGE 34

26 reported a consistent relationship between (a) depressive-catastrophic reaction and left hemisphere damage and (b) indifference/minimization of deficits and right hemisphere damage. Recently, Heilman and colleagues (Heilman et al 1 975; Heilman, Schwartz, 4 Watson, 1978; Heilman, Watson, & Bowers, 1933) have noted indifference reactions occur with striking frequency in RHD patients with the neglect syndrome suggesting that the two syndromes may be associated in some manner. This also suggests that right hemisphere changes associated with inappropriate euphoria and indifference may represent either (a) two points on a continuum or (b) two distinct emotional reactions following brain injury, one of them sharing a common mechanism with the unilateral neglect syndrome. Using the Depression scale of the Minnesota Multiphasic Personality Inventory as an index of depressive symptomatology, Gasparrini, Satz, Heilman, and Coolidge (1978) reported significantly elevated scores for LHD but not for RHD patients. More recently, Robinson and colleagues (Robinson, 1983; Robinson & Price, 1982) have also found that LHD patients were more likely to become clinically depressed and that RHD patients were more likely to be inappropriately euphoric. In addition, these results were not correlated with overall cognitive impairment, suggesting that the patient's emotional reactions to their deficits cannot solely account for these findings. Location of damage within the hemisphere has been found to be important in that these emotional reactions are more frequently associated with damage to anterior, frontal regions (Kolb & Milner, 1981; Robinson & Benson, 1981).

PAGE 35

27 Findings from studies of unilateral carotid injection of sodium amytal (WADA procedure) also support reported differences in emotional changes following LHD and RHD. Terzian (1964) and Rossi and Rosadini (1967) reported depressive-catastrophic reactions following left sided injections and inappropriate euphoria following right sided injections. These findings are also supported by other reports (Alema, Rosadini, & Rossi, 1961; Perria, Rosadini, & Rossi, 1961). However, Milner (cited in Rossi i Rosadini, 1967) failed to replicate these findings. In her investigation, only 5% of patients displayed depressive type responses, while the majority displayed euphoric reactions. This discrepancy between Milner's study and those of other investigators may be related to the significantly higher doses of sodium amytal used in the Milner study (Silberman & Weingartner, 1986) Investigation of normal, neurologicaliy intact subjects has also provided some support for the relative superiority of the left hemisphere in the processing of positive affect and the relative superiority of the right hemisphere in the processing of negative affect. Davidson ana colleagues (1979) recorded EEG responses while subjects viewed television programs of varying emotional content and subsequently indicated their emotional responses. Greater left hemispheric activity was found in response to positive emotional content, and greater right hemispheric activation was found in response to negative emotional content. Interestingly, this difference was only apparent on more anterior, frontal recordings while posterior, parietal activity suggested relative right hemisphere

PAGE 36

28 activation during all periods of felt emotion. These results again suggest the significance of the anterior-posterior dimension in processing of emotional information. Asymmetries of emotional facial expressions have also tended to support the relative superiority of the left hemisphere in processing of positive affect and the relative superiority of the right hemisphere in processing of negative affect. Sackeim et al (1978) have reported a tendency for facial expressions to be greater on the left side of the face. Furthermore, these authors suggest that this asymmetry was more pronounced for negative than positive facial expressions. Similarly, Schwartz, Ahern, and Brown (1979) have investigated facial expressions during spontaneous mood fluctuations. Right sided contractions were stronger during periods of nappiness or excitement, while left sided contractions were stronger during facial expressions of sadness and fear. Ahern and Schwartz (1979) have reported more right LEM (left hemisphere activation) when subjects respond to questions that evoked happiness or excitement. In contrast, more left LEM (right hemisphere activation) occurred when subjects responded to questions that evoked sad or fearful affects. As previously discussed, the questionable validity of LEM as an index of cerebral activation, however, must be considered in any interpretation of this study. Studies have also investigated left and right visual field differences for positive and negative emotions. Using a contact lens system that restricts visual input to the RVF (left hemisphere) or LVF (right hemisphere), Dimond, Farrington, and Johnson (1976) reported

PAGE 37

29 that unpleasant films were rated as more unpleasant when presented to the LVF than when presented to the RVF More recently, Reutei — Lorenz, Givis, and Moskovitch (1983) have also reported shorter reaction times to RVF presentations of happy faces and LVF presentations of sad faces. These findings are congruent with proposed left hemispherepositive affect and right hemisphere-negative affect distinctions. Although numerous studies are consistent with the hypothesis that the right hemisphere preferentially mediates negative emotions and the left hemisphere mediates positive emotions, an equal number of studies find no support for this hemispheric valence hypothesis. Rather both positive and negative stimuli seem to be preferentially mediated by the right hemisphere. To account for these discrepant views on the hemispheric processing of positive versus negative stimuli, Bryden and Ley (1983) have argued that methodological differences across studies might contribute to the discrepant findings. For example, ReuterLorenz and Davidson (1981) report faster reaction times for LVF presentations of sad faces and RVF presentations of happy faces. In this study, subjects were requirea to identify which of two laterally presented faces (one neutral and one emotional) showed an affective expression. In investigations which show a significant overall right hemisphere effect, the task is quite different. In these studies, the task is usually to determine whether a laterally presented face is the same or different than a centrally presented face. It is possible that these differences in task requirements may, in some way, account for the findings.

PAGE 38

30 A second possibility is that negative emotional faces may be more conf igurationally complex (requiring greater right hemispheric processing), a feature which may be of even greater significance in the task requirements of the Reuter-Lorenz task. This possibility does not, however, account for the large number of studies which have actually analyzed for type of emotion and still failed to find any laterality effect due to emotional valence (Bowers et al 1985; Bryden et al 1982; Buchtel et al 1978; Heiiman et al 1984; Ley & Bryden, 1979). A third possibility accounting for these discrepant findings is that studies which do find emotion specific hemispheric effects (i.e., left hemisphere-positive and right hemisphere-negative) are those which deal primarily with mood and/or experiential phenomena. In contrast, studies which do not find emotion specific hemispheric effects but instead do find right hemisphere superiority are those which involve cognitive encoding of emotional stimuli (I.e., "cold," cognitive tasks ) In their chapter, Bryden and Ley (1983) conclude that less evidence exists to strongly support the notion that the left hemisphere is more involved in the processing of positive affect and the right hemisphere is more involved in the processing of negative affect. The available evidence, however, provides strong support for the role of the right hemisphere in processing both positive and negative affective material.

PAGE 39

31 Mechanisms of Emotional Processing: The Role of Arousal Schacter and colleagues (Schacter, 1964; Schacter & Singer, 1962) proposed a theory of emotions termed the Cognitive-Arousal model. According to this model, an emotional state is the product of an interaction between arousal and cognition. An important assumption is that both arousal and cognition are necessary components of emotion. In this view, arousal is viewed as important in determining the felt intensity of the emotion while the cognitive element is important in determining the specific quality of the emotion. Early support for the joint roles of arousal and cognition were provided by Maranon (192H, cited in Fehr and Stern, 1970) who artificially aroused subjects with administration of drugs that stimulated the sympathetic nervous system. Maranon's subjects did not report feeling emotions although some did report feeling "as if" emotions. In contrast, if subjects were given a congitive set (induction of an affective memory) they did report emotional reactions when artifically induced arousal was present. Schacter (1970) has also provided support for this notion in a study which looked at the specific effects of pharmacologically induced arousal in neutral and stressful situation. Schacter demonstrated that physiological arousal alone (neutral condition) was not sufficient to evoke emotional responses from subjects but that the combination of arousal and availability of a cognitive label (stressful situation) was necessary. Although Schacter's research has been heavily criticized, especially on methodological grounds, this does not refute the

PAGE 40

32 assumption that arousal and cognition may play an important role in emotion Early investigations in neurophysiology laid much of the groundwork for our current knowledge of physiological arousal. In 1933, Berger reported that the e lectroencephalographic (EEC) pattern during behavioral arousal displayed decreases in amplitude and increases in frequency. This "electroencephalographic desynchroni zation" observed during periods of behavioral arousal was also later reported to occur during emotional states (Lindsley, 1970). Studies have also identified critical neuroanatomic structures involved in the elicitation of arousal responses. Stimulation in nonspecific thalamic nuclei or the mesencephalic reticular formation (MRF) result In behavioral manifestations of arousal as well as EEG desynchronization (Moruzzi & Magoun, 19^9). Similarly, stimulation of frontal or temporoparietal cortex activates the MRF (French, Hernandez-Peon, & Livingston, 1955) and elicits an arousal response (Sequndo, Nasuet & Buser, 1955). Another pathway by which cortical stimulation can produce arousal is via limbic system projections to cortex and MRF (Heilman & Valenstein, 1972; Watson, Heilman, Cauthen, & King, 1973). This conceptualization of reciprocal connections between MRF system and cortical regions is central to a model of arousal proposed by Sokolov (1963). Sokolov described a specific pattern of physiological changes which occurred in response to novel or "significant" stimuli. This specific pattern of physiological changes was termed the orientating response (OR). At the behavioral level,

PAGE 41

33 the organism may realign the head and/or body toward the source of stimulation. At the neurophys iological level, several changes occur which include a transient increase in skin conductance, pupil dilation, heart rate deceleration, pauses of respiration, and EEC desynchronization The presumed functional value of these collective components of the OR is to make the organism more receptive to incoming stimuli as well as to prepare the organism for action. A second component of Sokolov's model is that of the defensive response (DR), which is likely to be of equal, if not greater, potential significance in the processing of emotional stimuli. In Sokolov's view, when high intensity or aversive stimuli are presented, the orienting response is soon replaced by the defensive response. This response is characterized by greater increases in sympathetic activity across response systems including heart rate acceleration and cephalic vasoconstriction. The functional value of this response at the behavioral level is avoidance of the stimulus. It is interesting to note that both orienting and defensive responses are conceptualized as arousal responses, yet each results in characteristically distinct patterns of responding. This occurrence presents difficulty for the view of arousal as a unidimensional phenomenon. Subsequent investigators have looked at these seemingly paradoxical heart rate responses and attempted to correlate them with psychological processes. Lacey (1967) reconceptuali zed this "directional f •-actionation" of cardiac activity in terms of the conditions under which stimulation occurred and their effects on the organism's processing of stimuli.

PAGE 42

34 Lacey proposed that cardiac deceleration is associated with environmental intake while heart rate acceleration is associated with environmental rejection (the intake-rejection hypothesis). Lacey suggested that cardiac deceleration served to facilitate sensory processing while cardiac acceleration served to inhibit sensory processing. In this view, stimuli which elicit attention and interest are associated with cardiac deceleration and environmental intake. In contrast, stimuli which are painful or aversive or which require a significant amount of mental activity such as problem solving or arithmetic are associated with cardiac acceleration and environmental rejection. Lacey (1967; 1972) also proposed a neurophysiological mechanism for such cardiac changes whereby cardiac responses altered cortical activity indirectly by means of a visceral afferent feedback loop mediated by the baroreceptors An alternative explanation for heart rate changes has been offered by Obrist and col Leagues (1974) who have emphasized the relationship between motor requirements and cardiac activity. According to Obrist et al., there is a positive correlation between changes in cardiac activity and changes in levei of somatic activity and both are controlled by integrative mechanisms in the central nervous system (cardiac-somatic coupling). Obrist et al (1974) also noted that instances occur in which the cardiac-somatic coupling is dissociated, whereby increases in heart rate (termed cardiac preparatory responses) are observed without related overt changes in somatic activity. Interestingly, this occurs specifically in situations related to active avoidance of aversive stimuli.

PAGE 43

35 Similarly, Freyschuss (1970) has observed heart rate acceleration when subjects are instructed either to tense or move an arm even though such movement is impossible because of experimentally induced paralysis. These observations suggest that cardiac activity is not solely coupled with overt somatic activity per se but that cardiac activity is coupled with real as well as intended somatic activity. While orienting and defensive responses result in characteristically distinct patterns of autonomic responding, they have both been conceptualized as arousal responses. However, this view presents difficulty for the view of arousal as a unidimensional phenomenon. It also provides some support for the notion that autonomic reactivity is not as uniform as once suggested (Cannon, 1927; Schacter & Singer, ]962) Ax (1953) provided some of the first evidence to suggest that various affective states may be associated with distinct autonomic patterning. Ax reported that diastolic blood pressure increased more uuring anger than fear imagery, while heart rate and systolic blood pressure increased with equal magnitude. More recent studies have replicated these findings (Schacter, 1957; Weerts & Roberts, 1976). Schwartz, Weinberger, and Singer (1981) have recently reported cardiovascular differentiation between imagery induced happiness, sadness, fear, and anger. Several investigators have also found significantly greater heart rate accelerations in response to fearful stimuli such as mutiliation slides, spiders, and fearful imagery (Hare & Blevings, 1975; Klorman & Ryan, 1980; Klorman, Weissberg, & Weisenfeld, 1977; Vrana, Cuthbert, & Lang, 1986). Recently, Ekman,

PAGE 44

36 Levenson, and Friesen (1983) ^ave also reported heart rate increases in response to production of emotional facial expressions of anger, fear, and sadness and heart rate decreases in response to disgust, surprise, and happiness. In summary, there is some evidence to suggest that autonomic arousal is not as uniform as once suggested. In fact, several studies support the notion that different patterns of autonomic arousal may be associated with different types of emotional states. In addi':ion, recent conceptualizations of heart rate arousal responses have suggested that such changes may be linked to overt and/or covert motoric responses. This view is also consistent with the bioinformational theory proposed by Lang (1979). Lang proposes that emotional imagery results in patterns of autonomic activity very similar to those found in the actual emotional situation. This raises the possibility that certain emotions, by virtue of thei" strong motor components, may be more associated with heart rate acceleration while others with less motor demands may be more associated with heart rate deceleration. Neuropsychological Models of Emotional Processing Several different neuropsychological models of emotional processing have been proposed to account for findings of investigations in brain impaired and neurologically intact subjects. The models of Fox and Davidson (1984), Kinsbourne (Kinsbourne t Bemporad, 1984), Tucker (1981), and Heilman (Heilman et al 1983; Heilman, 1988, personal communication) will be briefly reviewed.

PAGE 45

37 The Model of Fox and Davidson Fox and Davidson (1984) have proposed a developmental model of neural substrates underlying affective response systems. Their model is based on evolutionary considerations of the critical role of basic approach and avoidance systems. These authors propose that approach and avoidance comprise the two underlying behavioral dimensions upon which all subsequent affective subsystems and responses have evolved. In addition, they suggest that hemispheric specialization constitutes the critical, neuroanatomical substrate of approachavoidance behavior. More specifically, these authors propose that the left hemisphere is specialized for approach behaviors or positive affects while the right hemisphere is more specialized for avoidance behaviors or negative affects. In support of tnis model, these authors relate the development of hemispheric specialization and interhemispheric transfer to the development of affective response systems. These authors argue that ail of the primary emotions emerge over the first year of life (Bowlby, 1972; Charlesworth 1964; Izard, Hubner, Risser, McGuiness, & Dougherty, 1980; Sroufe & Wunsch, 1972; Steinberg & Campos, 1983). Subsequent to this, they propose that primary emotions are modified by three processes: (a) the addition of new behaviors to the response repertoire of the "affect program," (b) regulation in the form of inhibition and appraisal, and (c) blending of primary emotions. Interest and disgust are reliably elicited in the neonate (Izard, 1977), and Fox and Davidson (1984) cite evidence from EEC findings in infants to suggest that these emotions are lateral ized. They found

PAGE 46

38 more interest expressions and greater relative left-sided EEG activity following sucrose administrations compared with citric acid administrations in infants and argued that these findings provide support for left hemisphere superiority in processing of positive affect. They further proposed that these emotions are under unilateral hemispheric control since little functional interconnection between the hemispheres exists at birth. Through the course of development, changes in interhemispheric communication are proposed as the necessary substrate for emergence of fear and sadness in the emotional repertoire. In support of this, the authors cite evidence that the onset of locomotion, a behavior associated with commissural transfer, is tightly coupled to the emergence of fear (Bayley, 1963; 1969; Rader Bausano, & Richards, 1980). In addition, the authors argue that the expression of sadness is often associated with alternating sequences of approach and avoidance, again implicating a critical role for interhemispheric communication (Ainsworth, Blehar Waters, &Wall, 1978; Izard, 1977). The capacity to inhibit negative affective responses, which emerge during the second year, are also presumed to be linked to the functional integrity of the commissural system. In addition, these authors propose that the left hemisphere normally exerts an inhibitory influence on the right hemisphere through transcallosal connections, resulting in attenuation of negative affect in the normal state. Kinsbourne's Model Kinsbourne and Bemporad (1984) proposed a separate model also based on the behavioral dichotomy that he terms action-approach and

PAGE 47

39 inaction-avoidance. This model, however, invokes both anteriorposterior and lateral specialization of cerebral processing. Posterior regions are proposed to provide data necessary to maintain homeostasis and anterior regions are proposed to exert control to stabilize as needed. In addition, anterior systems are viewed as exerting inhibitory control over its posterior source of information. As in the conceptualization of Fox and Davidson, this model also proposes a left hemispheric approach and right hemispheric avoidance dichotomy. However, in Kinsbourne' s model the left hemisphere is specialized for processing of external change and ongoing action (approach) and the right hemisphere is specialized for processing of internal changes, interruption of action (avoidance) as well as control of emotional arousal. This conceptualization establishes, in effect, four quadrants providing different mechanisms of processing: Left-Anterior Action control over external change Left-Posterior Enables left-anterior action control to make contact with necessary exteroceptive information Right-Anterior Emotional control over internal arousal Right-Posterior Enables right-anterior emotional control to make contact with interoceptive information With regard to the two control systems (action control and emotional control), Kinsbourne argues that the two hemispheres are not in inhibitory, but rather in compensatory interaction. Furthermore, whether the control is under left hemisphere-approach or right hemisphere-avoidance processing depends on the stimulus circumstances

PAGE 48

40 and on the status of the organism's attempt to exert control over its environment or itself. Kinsbourne provides a very interesting model of emotional processing which is based, to some extent, on the approach-avoidance model of Fox and Davidson and further elaborated to include an anterior-posterior dimension. However, less evidence exists to support or refute his claims. Tucker's Model Tucker (1981) has also suggested a neuropsychological model of emotional processing based on lateralized neuroanatomical systems. These systems control (a) tonic activation and motor readiness and (b) phasic arousal responses to perceptual input. In contrast to notions of reciprocal inhibition of the hemispheres, Tucker invokes mechanisms of subcortical release of lateralized arousal systems to account for left-negative versus right-positive valence findings. According to Tucker, the left hemisphere is specialized for activation and complex motor operations. Support for this is provided by the preponderance of right hand motor dominance in the general population as well as observed deficits in both right and left hand production of learned, skilled motor movements (apraxias) following left hemisphere lesions (Geschwind, 1975). The presumed neurochemical substrate for this specialization is the dopamine system which several investigations suggest is predominantly a left lateralized system (Click, Meibnach, Cox, & Maayani 1979; Wagner et al 1983). Tucker argues that activation operates in a tonic fashion to increase informational redundancy. This view is supported by the observation

PAGE 49

11 that increased dopaminergic activity is associated with restriction of behavioral output by the production of motor stereotypies in humans and other animals (Eilinwood, 1967; Iversen, 1977). Tucker also cites evidence from psychiatric literature to suggest that negative emotions of anxiety as well as ritualized, stereotyped behaviors associated with obsessive-compulsive disorder and, to some extent, left partial complex seizure disorder represent subcortical release and subsequent overactivity of this left hemisphere activation system. In contrast, Tucker proposes a right hemispheric specialization for phasic arousal responses to perceptual input which exerts its control through habituation. The presumed neurochemical substrate for this specialization is the norephinephr ine system which some investigations have suggested is represented to a greater extent in the right hemisphere (Oke, Keller, Mefford, 4 Adams, 1978; Oke, Lewis, & Adams, 1980). Lateralized norephinephrine pathways are known to show a pattern of widespread distrioution throughout the brain providing the necessary substrate for arousal responses and facilitation of orienting to novelty. In support of this. Tucker notes greater right hemisphere ability in tasks of "global" versus "local" processing, requiring integration of perceptual input (Levy, 1969; Nebes, ^97^). In addition, he cites evidence from the psychiatric literature suggesting greater right hemisphere involvement in hysteric personalities euphoric and often indifferent emotional responses which appear analogous to the responses of right hemisphere damaged patients (Galin, Diamond, & Braff, 1977; Gur & Gur 1975; Smokier & Shevrin, 1979).

PAGE 50

42 Tucker suggests that the two hemisphere's differing modes of processing may be the primary factor in lateralized valence effects reported in the literature. He hypothesizes that what we experience as emotions arise from operation of these arousal and attentional modulatory processes. Heilman's Model From investigations of indifference reaction associated with right hemisphere damage and unilateral neglect syndrome, Heilman and colleagues (1983) have suggested a model of emotional processing based on hemispheric differences in arousal-activation responses. Heilman has suggested that right hemisphere damaged patient's difficulties in emotional expression may be a result of (a) deficits in arousalactivation and (b) an ability to develop an appropriate cognitive state due to basic deficits in comprehension of prosodic elements of speech and affective facial expressions. Patients with indifference reaction often have the unilateral neglect syndrome in which they may fail to orient, report, or respond to stimuli in the contralateral side of space (Denny-Brown et al 1952; Gainotti, 1972; Heilman & Valenstein, 1972). Heilman et al have suggested that unilateral neglect is a defect in attenuation-arousal-activation due to disruption of a corticolimbic-reticular loop (Heilman & Van Den Abel, 1979). Based on the fact that neglect occurs most often following right hemisphere damage, he has proposed that the right hemisphere may be dominant for mediating attention-arousal-activation responses. To investigate arousal responses in brain impaired patients, Heilman et al (1978) stimulated the forearm ipsilateral to the side

PAGE 51

J I ^3 of lesion in RHD and LHD patients while recording galvanic skin response from the same side. The authors noted that RHD patients had significantly smaller GSR arousal responses than LHD patients or nonbrain damaged control subjects. Similarly, Morrow, Urtunski Kim, and Boiler (1981) presented LHD and RHD patients with neutral and emotional stimuli. Right hemisphere patients showed decreased galvanic skin responses to both neutral as well as emotional stimuli relative to LHD patients. More recently, Yokoyama, Jennings, Ackles, Hood, and Boiler (1987) have looked at heart rate and reaction time responses in patients with right unilateral hemispheric lesions. These authors found that RHD patients had significantly slower reaction times and decreased heart rate responses (both deceleratory as well as acceleratory ) relative to LHD patients. These findings indicate that the greater role of the right hemisphere in attention may be reflected in both reaction time as well as anticipatory heart rate changes. Investigations in neurological ly intact subjects corroborate these findings. Hugdahl, Franzon, Anderson, and Walldebo (1983) report greater anticipatory heart rate accelerations for emotional stimuli presented to the LVF (right hemisphere) compared with RVF (left hemisphere) trials. Similarly Walter and Sandman (1982) also report greater right hemisphere activity (as measured by the PI 00 component of the average evoked potential) when the heart was spontaneously accelerated. In addition, Hugdahl, Wahlgren, and Wass (1982) found delayed habituation of the electrodermal orienting

PAGE 52

44 response to visual stimuli initially projected to the right hemisphere Heilman and Van Den Abel (1979) have also suggested a right hemisphere superiority for activation. Using a neutral warning stimulus paradigm, these authors reported that warning stimuli projected to the right hemisphere reduced reaction times of the right hand more than warning stimuli projected to the left hemisphere. In addition, warning stimuli projected to the right hemisphere reduced reaction times of the right hand more than warning stimuli projected to the left hemisphere reduced reaction times of the left hand. Based on these findings, it can be seen that warning stimuli projected to the right hemisphere reduced reaction times of both hands to a greater extent than left hemisphere presentations, suggesting that the right hemisphere was better able to activate responses in both hands relative to the left hemisphere. In addition, Verfaellie, Bowers, and Heilman (1987) reported a 3tudy of neurologically intact subjects which provides some support for the right hemisphere dominance of activation. By manipulating preliminary intentional warning cues (which hand to use in responding), they found faster left hand versus right hand responses suggesting that the left hand (right hemisphere) was better able to benefit from this preparatory information than the right hand (left hemisphere). In support of this, Verfaellie and Heilman (1987) also report the performance of two patients with right and left supplementary motor area (SMA) damage on this paradigm. They report that the patient with left SMA damage (intact right hemisphere) was

PAGE 53

^5 able to benefit from preparatory information wnile the patient with right SMA damage was unable to benefit from preparatory information, again suggesting a greater role of the right hemisphere in activation of response. More recently, Heilman (1988, personal communication) has suggested that emotion specific hemispheric effects (i.e., left hemisphere-positive, right hemisphere-negative) reported in the literature may be artifactual and actually represent hemispheric differences in arousal and preparation for action. Because the right hemisphere is dominant for mediating arousal/activation, it may therefore be more intrinsically involved in processing emotional stimuli that have greater "preparatory" significance for survival (i.e., fight-flight emotions such as anger and fever). In contrast, the left hemisphere may be more involved in mediating nonpreparatory emotions (i.e., happiness, sadness, disgust) that place less immediate or "phasic" attentional demands on the organism for survival. In summary, recent investigations in braia impaired and neurologically intact subjects suggests a greater r^ole of the right hemisphere in arousal-activation responses. Furthermore, this finding is observed across several indices of arousal-activation including heart rate, skin conductance, and reaction time measures. Critical Issues As reviewed in the introduction, there appears to be general consensus that the two hemispheres in man differ in terms of their contribution to emotional processing. However, the precise role

PAGE 54

46 played by each remains unclear. Some investigators have argued that the right hemisphere is globally involved in all aspects of emotional processing including the cognitive encoding/decoding of emotional stimuli, arousal-activation responses to emotional stimuli and behavioral responses to these stimuli (Heilman et al 1983; I-.ey & Bryden, 1979). Other investigators have argued that the two hemispheres differ in terms of the type of emotions that are preferentially mediated by each (Fox i Davidson, 1984; Kinsbourne & Bemporad, 1984; Tucker, 1981). The most popular version of this view is that the left hemisphere is dominant for positive (approach) emotions, whereas the right hemisphere is dominant for negative (avoidance) emotions. Still others have argued that this positive-negative dichotomy in hemispheric processing of emotions is artifactual and actually relates to hemispheric differences in arousal and preparation for action (e.g., activation) (Heilman, 1988, personal communication). In this view, the right hemisphere is dominant for mediating arousal/activation and as such, is more intrinsically involved in processing emotional stimuli that have greater "preparatory" significance for survival (i.e., fight-flight emotions such as anger and fear). In contrast, the left hemisphere is more involved in mediating nonpreparatory emotions (i.e., happiness, sadness, disgust) that place less immediate or "phasic" attentional demands on the organism for survival. In order to distinguish among these models, it would be necessary to determine whether different categories of emotional stimuli result

PAGE 55

^7 in differential patterns of arousal/activation, depending on the hemisphere to which they were presented. According to the global right hemisphere emotion model, emotional stimuli of any valence directed to the right hemisphere should result in greater arousal/activation responses than those directed to the left hemisphere. According to valence models, negative emotional stimuli (anger, fear, disgust) would induce greater arousal/activation responses when directed to the right versus left hemisphere, whereas the opposite should occur when positive stimuli (happiness) are used. Finally, according to the preparatory/nonpreparatory model, emotional stimuli having preparatory significance (anger, fear) should result in greater arousal/activation responses when directed to the right versus left hemisphere. The opposite should occur with nonpreparatory emotional stimuli (happiness, disgust, neutral). The focus of the present study was to further examine these divergent views regarding the hemispheric processing of emotional stimuli. The basic paradigm was one in which neutral and emotional stimuli of different valence were laterally presented to either the left or right hemisphere (using a tachistoscopic procedure). The purposes of this study were to determine (a) the extent to which laterally presented eraotional/nonemotional stimuli might result in differential patterns of behavioral activation (as assessed by reaction time responses) as well as differential patterns of autonomic arousal (as assessed by HR and SCR responses); (b) whether there were hemispheric asymmetries in mediating arousal and/or activation in response to these emotional/nonemotional stimuli; and (c) whether

PAGE 56

48 certain categories of emotion (positive/negative; preparatory/ nonpreparatory ) induce asymmetric arousal/activation, depending on the hemisphere to which they are initially presented. In order to address these issues, two experiments were completed. In the first study, laterally presented emotional stimuli of different valences served as warning stimuli to the subjects who then made manual RT responses to a neutral midline stimulus. This warning stimulus paradigm was chosen because it enables one to determine the extent to which lateralized emotional warning stimuli serve to behaviorally activate and prepare the individual to respond to a subsequent stimulus (Lansing, Schwartz, & Lindsey, 1959). In the second study, laterally presented emotional stimuli were also shown to subjects, and autonomic indices of arousal (HR, SCR) were measured. Although it would have been more "ideal" to obtain both autonomic and RT measures to the lateralized emotional stimuli in the same study, this was not realistically feasible. The "slow" rise time of the SCR (2-4 seconds) in conjunction with the relatively short lived activating effects of warning stimuli (500-2000 msec) precluded such a direct manipulation. Thus, two separate experiments were conducted. Four different emotional categories were chosen for the present investigation. Two categories which have previously been shown to result predominantly in cardiac deceleratory responses (happy, disgust) and two categories which have previously been shown to result predominantly in cardiac acceleratory responses (fear, anger) (Ekman et al 1983). Due to the relative paucity of discernable positive emotions among the wide range of emotional categories (Ekman, 1972),

PAGE 57

49 it was not possible to equate occurrence of positive versus negative emotions Hypotheses and Predictions According to the right hemisphere emotion model, the right hemisphere plays a greater role than the left hemisphere in mediating arousal/activation responses to emotional materials. If this model is correct, then one would predict faster RT responses to a midline neutral stimulus when it is preceded by an emotional warning stimulus (WS) directed to the right hemisphere (LVF) than by an emotional WS directed to the left hemisphere (RVF). Additionally, RTs should also be faster when emotional WS versus nonemotional WS are directed to the right hemisphere (LVF). These predictions were examined in Experiment I. Similarly, one would also predict that autonomic responsivity to emotional versus nonemotional stimuli should be greater when they are directed to the right hemisphere (LVF) versus the left hemisphere (RVF). These predictions were examined in Experiment II. According to hemispheric valence models of emotional processing, negative emotional stimuli are preferentially mediated by the right hemisphere, and positive emotional stimuli are mediated by the left hemisphere. If this hypothesis is correct, then one would predict that negative emotional WS directed to the right hemisphere (LVF) should result in faster RTS to a neutral midline stimulus than when the negative WS is directed to the left hemisphere (RVF). Conversely, positive emotional WS directed to the left hemisphere (RVF) snould result in faster RTs than positive WS directed to the right hemisphere

PAGE 58

50 (LVF). Similarly, in Experiment II, one would predict greater autonomic responsivity (HR acceleratory and deceleratory responses, SCR) to negative stimuli that are directed to the right hemisphere (LVF) versus stimuli that are directed to the right hemisphere (LVF) versus those directed to the left hemisphere (RVF). The opposite pattern of autonomic arousal should occur for positive emotional stimul i According to the hemispheric preparatory model of emotion, the right hemisphere is dominant for mediating emotional stimuli that have a greater preparatory significance for survival (i.e., fight-flight emotions such as anger and fear). In contrast, the left hemisphere is dominant for mediating nonpreparatory stimuli that' place less "phasic" demands on the individual for immediate survival (i.e., happiness, disgust, neutrality). If this model is correct, then in the first experiment one would predict that anger and fear WS directed to the right hemisphere (LVF) snould result in faster RTs than when they are directed to the left hemisphere (RVF). Conversely, happy, disgust, and neutral WS should result In faster RTs when they are directed to the left versus right hemisphere. Likewise, if one assumes that behavioral activation and autonomic responsivity are strongly coupled, then similar predictions would be made for Experiment II. That is, one would predict greater autonomic responsivity (HR acceleratory and deceleratory responses) to anger and fear stimuli (assumed to be preparatory) when they are directed to the right versus left hemisphere. The opposite hemispheric pattern of autonomic arousal

PAGE 59

51 should occur for neutral, disgust, and happy stimuli (assumed to be nonpreparatory ) Alternatively, it is possible that the right hemisphere may be dominant for mediating arousal/activation responses to stimuli, regardless of their emotional-nonemotional content. In this view, stimuli directed to the right hemisphere should result in greater arousal/activation responses than stimuli directed to the left hemisphere. However, any differences in arousal/activation responses to emotional versus neutral stimuli should be comparable across the left and right hemispheres. Thus, in Experiment I, WS directed to the right hemisphere (LVF) should result in faster RTs than WS directed to the left hemisphere (RVF). Any RT differences between emotional versus nonemotional WS should be comparable for LVF and RVF presentations. Likewise, in Experiment II, one would predict greater autonomic responsivity (HR SCR) to stimuli directed to the right versus left hemisphere. Again, however, any differences in arousal responses to emotional versus neutral stimuli should be comparable across LVF and RVF presentations of the stimuli.

PAGE 60

METHOD SubJ ects A total of 60 (30 male, 30 female) students at the University of Florida served as subjects (Ss) in the present investigation. Subjects were given either course credit or paid for their participation. All Ss were right handed according to self-report and their performance on the Briggs and Nebes (1975) Handedness Questionnaire. An overall score of +9 or above (right hand preference) was used as the criterion for participation in this study. Different groups of 30 Ss each (15 male, 15 female) participated in Experiment I and 11. Subjects were randomly assigned to Experiment I or Experiment II. Due to the potential confounding effects of prior exposure to the emotional stimuli on subsequent RT and psychophysiological responses, a between groups comparison was felt to be advantageous. The mean age of Ss in Experiment I was 20.7 with a range of 18-26 years. The mean age of Ss in Experiment II was 21.0 with a range of 19-27 years. Experiment I: Reaction Time Task Stimuli Stimuli consisted of 192 black and white slides. These slides depicted 96 neutral and 96 emotional scenes which included 24 slides 52

PAGE 61

53 of each of four different valences ('nappy, angry, fearful, and disgusting) The emotional scenes were selected from a variety of materials including magazines and photography books. The scenes used in this experiment did not include familiar landmarks or personalities in an effort to avoid possible confounding effects of familiarity on RT and HR responses to these stimuli. The stimuli were rated for type and intensity of affect by 20 (10 male, 10 female) University of Florida students who did not participate in the present experiments. All stimuli averaged at least 9'\% agreement. Mean intensity ratings (on a scale of 1 to 5) for the five categories were happy, 3'^; angry, 3.6; fearful, 3.9; disgusting, 3.3; and neutral, .3. Apparatus Slides were projected onto a '40x35-Gm Kodak milk-glass, rear view projection screen. A 5-mm diameter red light-emitting diode (LED) was placed at the center of the screen to serve as a central fixation point and imperative stimulus in the RT task. Two spring loaded keys were placed 30 cm to the left and right of body midline. The timing, presentation of stimuli, and recording of Ss' responses from release of spring loaded keys were accomplished by an I3M-PC microcomputer interfaced with BRS logic. Slides were projected at 5 degrees of visual angle lateral to the central fixation LED. Attached to the projector lens were Uniblitz 325B high speed shutters which allowed maximum rise and fall time of slide presentation. The room was dimly lit to avoid the effects of visual startle during stimulus presentation

PAGE 62

54 Eye movements were continuously monitored by electro-oculography (EOG) to ensure maintained fixation and laterallzed presentation of slides. The EOG signal was detected by Beckman Ag/AgCl miniature electrodes attached at the temporal canthus of the left and right eye. The EOG signal was filtered and fed into a DC amplifier and recorded on a Grass Model 78B polygraph. Event marker input from IBM-PC microcomputer was also fed into the polygraph recording to identify occurrence of slide presentation and to facilitate subsequent identification of eye movements during this interval. Procedure Subjects were seated 91 cm from the projection screen with left and right hands placed on left and right sided keys, respectively. The task was a choice reaction time task. SuDjects viewed a laterally presented warning stimulus (neutral or emotional scene) of 500 msec. This was followed 500, 1000, or 1500 msec later by a centrally presented neutral, imperative stimulus (red LED) with interstimulus interval randomly varied across trials. Half of the Ss were Instructed to release the left key following onset of the LED if the preceding warning stimulus was a neutral scene or the right key if the warning stimulus was an emotional scene. The remaining Ss were instructed to release the left key following onset of the LED if the preceding warning stimulus was an emotional scene and the right key if the preceding warning stimulus was a neutral scene. Half-way through the experiment, hand order was reversed for all subjects. Subjects received a total of 192 trials. Response hand and visual field of

PAGE 63

55 presentation were randomized and counterbalanced across stimulus type. Subjects were given 10 practice trials prior to the experiment. Experiment II: Psychophysiological Responses to Laterally Presented Emotional Material Stimuli The stimuli were identical to those used in Experiment I. They included 24 neutral and 96 emotional scenes [24 of each of four differing valences (happy, angry, fearful, disgusting)]. Apparatus Slides were projected onto a 40x35-cm Kodak milk-glass, rear view projection screen. A 5-mm diameter adhesive circle was placed at the center of the screen to serve as a central fixation point. Slides were projected at 5 degrees of visual angle lateral to the central fixation point. Lafayette Model //1430I6 shutters were attached to the projector lens to maximum the rise and fall time of slide presentation. The timing and presentation of stimuli were accomplished by IBM-PC microcomputer interfaced with BRS logic. Eye movements were continuously monitored in the same fashion as Experiment I. The room was again dimly lit to avoid the effects of visual startle during stimulus presentation. Psychophysiological measures (HR, SCR, and respiration depth) were recorded for the 3 seconds prior to stimulus onset and 8 seconds of stimulus presentation. Heart rate responses were recorded by two Beckman Ag/AgCl electrodes attached to the right and left lateral margins of the chest. A third electrode was attached to the

PAGE 64

56 sternum. Electrode sites were prepared by mild abrasion of the skin with Hewlett Packard Redux paste. Electrodes were fastened by the use of adhesive collars and were filled with Hewlett Packard Jel Redux cream as the electrolyte. The electrocardiogram (ECG) was amplified by a Colbourn S75-03 high gain bioamplif ier This signal was input to a Colbourn 375-38 Ban-Pass Biofilter with subsequent detection of the R-wave component which interrupted the computer to provide inter-beat intervals. A-D conversion was accomplished by Colbourn R65-17 Data Translation Board and signal was downloaded to an IBM-PC interfaced with data acquisition modules. Skin conductance was recorded with Met-Associates electrodes placed on the thenar and hypothenar eminences of the left and right hands. Electr-ode sites were wiped clean with distilled water. Electrodes were attached by the use of adhesive collars filled with KY jelly. The analog SC signal was f^d into a Colbourn Model S71-22 Skin Conductance Module. This signal was digitized by a Colbourn R65-17 Data Translation Board and the signal was downloaded to an IBM-PC microcomputer interfaced with the data acquisition modules. Respiration depth was also recorded to detect and subsequently exclude those trials in which unusually large inhalations or exhalations may have confounded HR or SCR. Respiration depth was measured by means of a Colbourn Model Ti41-91 Aneroid Chest Bellows, and processed by a Colbourn S72-25 Module. A-D conversion and computer interface utilized the same equipment as HR and SCR measures.

PAGE 65

57 Procedure Prior to the experiment, Ss were requested to list specific episodes from their own lives which they considered happy, angry, fearful, disgusting, or neutral. Subjects were requested to list five episodes for each category for a total of 25. The experiment took place in a quiet, dimly lit room. Subjects were seated in a comfortable, reciiner chair positioned 152 cm from the projection screen. After electrode placement and a 20-minute adaptation period, Ss were instructed to refrain from any unnecessary movement during the experiment. Subjects were presented with a neutral or emotional slide in left or right visual field for 3 seconds. Subjects were instructed to maintain fixation on the centrally positioned circle throughout slide presentation. To decrease Ss' habituation and encourage continued processing of the stimuli during this time, Ss were also instructed to recall one of the five episodes of the same valence as the slide presented. Approximately 10 seconds after slide offset, Ss were asked to indicate the valence of the slide and responses were recorded manually on a separate sheet. Inter-trial interval varied randomly from 23 to '45 seconds to minimize occurrence of anticipatory HR and SCR. Subjects received a total of 120 trials with visual field of presentation randomized and counterbalanced across stimulus type. Subjects also received 10 practice trials for maintained fixation prior to the experiment utilizing neutral stimuli only.

PAGE 66

RESULTS Experiment I: Reaction Time Task Reaction Time Responses Subjects' reaction time responses served as the data for analysis in a r^epeated measures analysis of variance (ANOVA). The between subjects factor was Sex (male, female). The within subject factors were Visual Field (left, right). Hand (left, right), and Stimulus Type (happy, angry, fearful, disgusting, neutral) as within subjects factors. Incorrect trials and trials in which eye movements occurred were not included in the analysis. The remaining 3]% of trials served as the data for analyses. This included 83? of happy trials, 82% of angry trials, 82% of fearful trials, T^% of disgusting trials, and 85% of neutral trials. A log transformation was used to correct for skewness in data distribution and to adequately meet homogeneity of variance assumptions for the ANOVA (Kirk, 1968; Winer, 1971). A summary of the results of this ANOVA are depicted in Table 1 Findings revealed a main effect of Sex [_F (1, 28) = 4,44, p = .0442)] with mean RT responses of male (M = 481.20) Ss significantly faster than RT responses of female (M = 603.48) Ss depicted in Figure 1. A significant Sex x Visual Field effect [ F_ ( 1 28) = 4.06, _p = .0535)] was also obtained and is depicted in Figure 2. Post-hoc simple effects testing was performed to clarify the nature of this interaction (Kirk, 1978; Winer, 1971). Findings revealed no 58

PAGE 67

Table 1 Summary ot Analysis of Variance: Experiment I, Reaction Time Source df ss r P Sex 1 28 Visual Field 1 28 001 D 25 Visual Field X Sex 1 28 4 UO if Hand 1 28 0 J /t) o { Hand x Sex 1 28 A r\ o o 07 Stimulus Type 1 1 2 .087'! 1 .38 Stimulus Type x Sex H, 112 .1674 2.64 Visual Field X Hand 1 28 .0000 .01 Visual Field X Hand x Sex 1 28 .0008 .05 Visual Field x Stimuliis Type 4, 1 12 .0334 • 35 Visual Field X Stimulus Type x Sex 1 1 2 .0966 1 .02 Hand x Stimulus Type 4, 1 12 .1497 1 .38 Hand x Stimulus Type x Sex n, 1 1 2 .0386 .36 Visual Field X Hand x Stimulus Type 1 12 .1995 2. 49 Visual Field x Hand x Stimulus X Sex Type 4, 1 12 .0357 .45 p < .05.

PAGE 68

60 510 -I 500 H jl 490 C o 480 CO
PAGE 69

51 significant differences across LVF and RVF trials for female Ss [F_ (1, lU) = 3.71, P = .0644)]. However, for male Ss RVF (M = 477.10) were significantly faster than LVF (M = 485.49) trials CF_ (1 14) = 5.23, P = .0299)]. Findings also revealed a significant interaction of Sex x Stimulus Type CF_ (4, 112) = 2.64, p = -0376)], depicted in Figure 3. Post-hoc simple effects testing revealed no significant differences across Stimulus Type for female Ss [_F (4, 56) = 1.26, £ = .294)]. However, male Ss showed a significant effect for Stimulus Type [F (4, 56) = 2.60, _p = .0454)]. Duncan's post-hoc comparisons revealed that for male Ss RT responses to disgusting slides (M = 506.28) were significantly slower than happy (M = 468.29) or angry (M = 483.47) slides at p_ < .05. A significant Hand x Visual Field x Stimulus Type interaction [F (4, 112) = 2.49, £ = .0473)] was also obtained and is depicted in Figure 4. Post-hoc simple effects testing of RVF trials only revealed no statistically significant differences across Hand and Stimulus Type conditions [F_ (4, 116) = .97, £ = .4249)]. In LVF, however, post-hoc simple effects tests revealed a significant Hand x Stimulus Type interaction [_F (4, 116) = 2.61, £ = .0390)]. Duncan's post-hoc comparisons revealed that happy trials in LVF were significantly faster when using the right (M = 507.45) versus left (M = 562.94) hand at £ < .05. Further inspection of the data revealed that the significant Sex effects observed in the preceding analysis may have been influenced by the markedly slowed performance of two female Ss When compared to

PAGE 70

62 700 -I 600 601 .97 604.99 Females —a 500 485.49 477.10 Males 400 — I 1 — LVF RVF Visual Field (b) Figure 3. Experiment I--Reaction Time Analysis, Sex x Stimulus Type Interaction: (a) males only and (b) males and females.

PAGE 71

E c .g o CO Q GC 580 T 560 540 520 Left Visual Field Trials 578.99 562.94 543.63 543.55 557.91 507.45 521 .39 500 -I r— \ I 1 1 1 Happy Angry Fearful Disgusting Neutral Emotion (a) Right Hand Left Hand Right Visual Field Trials o E c o 1 560 -1 550 540 530 520 553.39 Right Hand 553.61 Left Hand 533.31 T r Happy Angry Fearful Disgusting Neutral Emotion (b) Figure ^4. Experiment IStimulus Type and (b) right -Reaction Time Analysis, Hand x Visual Field Interaction: (a) left visual field trials visual field trials.

PAGE 72

'f t 64 the overall mean RT of females for each of the 20 Visual Field x Hand X Stimulus Type conditions, these two Ss possessed 8 (40$ of total) and 10 (50$ of total) mean reaction times which fell two standard deviations above the overall mean performance of female Ss Of the remaining 13 female Ss no S possessed a single mean reaction time greater than 2 standard deviations above the overall mean of female Ss For this reason, a second analysis of the RT data was conducted in which the data from the two female Ss noted above was excluded. A summary of the results of this analysis are depicted in Table 2. Findings revealed no significant main effects. The main effect of Sex observed on the preceding analysis was no longer significant suggesting that this effect may have been significantly influenced by the markedly slowed performance of the two female Ss noted above. Findings, however, did reveal a significant Visual Field x Sex interaction [ F_ ( 1 26) = 5.20, £ < .03il)] as in the preceding analysis, depicted in Figure 5. Simple effects testing revealed a pattern of findings similar to those noted in the first analysis with no significant difference between LVF and RVF performance for female S3. Males, however, performed significantly faster to RVF stimuli relative to LVF stimuli. A trend for the Sex x Stimulus Type interaction was also noted [F_ (4, 104) = .0722, £ = .0722)] which had previously attained significance in the first analysis. This interaction is depicted in Figure 6. The Hand x Visual Field x Stimulus Type interaction significant, in the first analysis, failed to reach significance in the present analysis.

PAGE 73

65 Table 2 Summary of Analysis of Variance: Experiment I, Reaction Time (minus outliers) Source df ss F p 1 26 2.3201 2.17 Visual Field 1 26 .0002 .03 V isual F i6ld x Sex 1 26 .0332 5.20 Hand 1 26 .0441 1.15 H^i nfi Y Y 1 26 .0060 .16 Stimulus Type 4, 104 .1194 1 .96 Stimulus Type x Sex n, 104 .1350 2.22 Visual Field x Hand 1 26 .0019 .12 Visual Field x Hand x Sex 1 26 .0000 .00 Visual Field x Stimulus Type n, 104 .0187 .20 Visual Field x Stimulus Type x Sex 104 .0699 .76 Hand x Stimulus Type ij. 104 .1714 1 .59 Hand x Stimulus Type x Sex 4, 1 04 .0357 .33 Visual Field x Hand x Stimulus Type 4, 104 .1514 1.86 Visual Field x Hand x Stimulus x Sex Type 4, 104 .0362 .45 p < .05.

PAGE 74

66 600 T E 550 55Z16 EH556.34 Females a c o 1
PAGE 75

67 Percent Correct Responses A separate ANOVA was conducted which used proportion of correct responses for each S as the dependent variable. An arcsin square root transformation was performed to correct for lack of a normal distribution inherent in proportion data and to meet homogeneity of variance assumptions for the ANOVA (Kirk, 1968; Winer, 1971). Results of this analysis utilizing axl 30 Ss are depicted in Table 3. Findings revealed only a significant main effect of Stimulus Type [£ (4, 112) = 7.60, p = .0001)], depicted in Figure 7. Duncan's post-hoc comparisons revealed that percent correct identification of disgusting (M = .80'4) trials was significantly worse than all four remaining categories (happy, M = .907; angry, M = .90^4; fearful, M = .903; neutral, M = .938) at _p < -05. In addition, a trend for Stimulus Type x Visual Field [£ (4, 112) = 2.29, _p = .064)] was also revealed, depicted in Figure 8. This pattern of findings suggest that happy trials were more accurate in the LVF while neutral trials were more accurate in the RVF. A second ANOVA was conducted for percent correct data which excluded the two female Ss noted dbove. Results of this analysis are depicted in Table ^. This analysis revealed only a significant main effect of Stimulus Type [F (4, 104) = 6.13, P = .0002)], depicted in Figure 9. Duncan's post-hoc comparisons revealed that percent correct identification of disgusting (M = .810) trials was significantly worse than all four remaining categories (happy, M = .911; angry, M = .904; fearful, M = .899; neutral, M = .940) at p < .05. No other effects reached significance or trend status.

PAGE 76

68 Table 3 Summary of Analysis of Variance: Experiment I, Percent Correct Source df SS F P Sex 1, 28 .8665 2.28 Visual Field 1. 28 .0181 .22 Visual Field x Sex 1 28 .1683 2.01 Hand 1 28 .0757 1 .23 Hand x Sex 1 28 .0203 .33 Stimulus Type ^ 112 2.5844 7.60 Stimulus Type x Sex ^, 112 .2799 .82 Visual Field x Hand 1 28 .0009 .01 Visual Field x Hand x Sex 1 28 .0004 .01 Visual Field x Stimulus Type 1 12 .4369 2.29 Visual Field x Stimulus Type x Sex ^ 1 1 2 .1908 1 .00 Hand x Stimulus Type 1 12 4237 1 .44 Hand x Stimulus Type x Sex 112 .1 018 .35 Visual Field x Hand x Stimulus Type 4. 1 12 .0793 .45 Visual Field x Hand x Stimulus X Sex Type ^4. 112 .1774 1 ,02 p < .01

PAGE 77

0.75 I I I 1 I I I Happy Angry Fearful Disgusting Neutral Emotion Figure 7. Experiment I — Percent Correct Analysis, Stimulus Type Main Effect Figure 8. Experiment I--Percent Correct Analysis, Stimulus Type x Visual Field Trend.

PAGE 78

70 Table 4 Summary of Analysis of Variance: Experiment I, Percent Correct (minus outliers ) Source df ss F P Sex 1 26 c\'~r r\\\ 9794 2.42 Visual Field 1 26 0009 .01 Visual Field x Sex 1 26 .0552 .87 Hand 1 26 .0490 76 Hand x Sex 1 26 .0091 Stimulus Type 4, 104 2.0531 6.13 Stimulus Type x Sex 4, 104 .2958 .88 Visual Field x Hand 1 26 .0205 .27 Visual Field x Hand x Sex 1 26 .0180 .24 Visual Field x Stimulus Type 4, 104 .3594 1 .86 Visual Field x Stimulus Type x Sex 4, 104 .2289 1.15 Hand x Stimulus Type 4, 104 3254 1 .16 Hand x Stimulus Type x Sex 4, 1 04 1 308 .47 Visual Field x Hand x Stimulus Type 4, 104 .0452 .28 Visual Field x Hand x Stimulus X Sex Type 4, 104 .0866 .54 p < .01

PAGE 79

71 Figure 9. Experiment I --Percent Correct Analysis, Stimulus Type Main Effect

PAGE 80

72 Experiment II Heart Rate Data Reduction Heart rate responses were edited by a computer program which converted all trials from interbeat interval data to beats per minute format for each second of the sampling period. Each interbeat interval was weighted proportionally to the fraction of the second it occupied according to method recommended by Graham (1980). Baseline for each trial was defined as the average HR for the 2 seconds preceding stimulus onset. For each trial, baseline HR was then subtracted from each of eight post-stimulus HR values, yielding second by second post-stimulus HR changes from baseline. Trials in which eye movements occurred were not included in the analysis. The remaining 77% of trials served as the data for analysis. This Included 7^% of happy trials, 77% of angry trials, 73? of fearful trials, 75% of disgusting trials, and 78J of neutral trials. From this, two separate data sets were generated. The first, selected, for each trial, was the maximum decelerat ory HR response from the eight post-stimulus second by second HR changes from baseline. This maximum deceleratory response then served as the dependent variable in a repeated measures ANOVA. The second data set selected, for each trial, was the maximum acceleratory HR response from the eight post-stimulus second by second HR changes from baseline. This maximum acceleratory response also served as the dependent variable in a separate repeated measures ANOVA.

PAGE 81

73 Maximum deceleratory responses A repeated measures ANOVA was performed witn Sex (male, female) as a between subjects factor and Visual Field (left, right), and Stimulus Type (happy, angry, fearful, disgusting, neutral) as the witnin subjects factors. A summary of the results of this ANOVA are depicted in Table 5. Results of this analysis revealed a main effect of Sex [F (1, 28) = 5.16, p = .0310)]. Males (M = -9.153) showed significantly greater maximum deceleratory HR responses than females (M = -7.756). These findings are depicted in Figure 10. Results also revealed a trend for a main effect of Visual Field [ F_ ( 1 28) = 3-33. p = .0789)] with a pattern of greater deceleratory HR responses to LVF trials (M = -8.700) relative to RVF trials (M = 8.209) (see Figure 11). This analysis also yielded a main effect of Stimulus Type [F_ (4, 112) = 4.96, p = .0010)]. Duncan's post-hoc comparisons revealed happy slides (M = -9.016) elicited significantly greater HR decelerations than angry (M = -7.969) or fearful (M = -7.663) slides at £ < .05. Neutral slides (M = -8.984) also elicited significantly greater HR decelerations than angry or fearful slides at £ < .05. These relationships are depicted in Figure 12. Findings also revealed a significant Visual Field x Sex interaction [F_ (1, 28) = 7.86, £ = .0091)]. Simple effects testing revealed that males showed significantly greater deceleratory HR changes in LVF (M = -9.776) relative to RVF (M = -8.530) trials, depicted in Figure 13Results also revealed a significant Sex x

PAGE 82

74 Table 5 Summary of Analysis of Variance-. Experiment II, Heart Rate Deceleration Source df SS F P Sex 1 23 1 46.4789 5.16 Visual Field 1. 28 18.0850 3.33 Visual Field x Sex 1 28 42.7167 7.86 ** Stimulus Type 4. 1 12 89.6031 4.96 Stimulus Type x Sex 4, 112 60.4664 3.35 Visual Field x Stimulus Type 4, 1 12 7.7819 .47 Visual Field x Stimulus Type X Sex 4, 1 12 1 2.6660 .77 p < .05. p < .01

PAGE 83

Figure 10. Experiment II — Maximum Deceleratory Heart Rate Analysis Sex Main Effect Figure 11. Experiment II--Maximum Deceleratory Heart Rate Analysis Visual Field Trend.

PAGE 84

"Figure 12. Experiment II--Maximum Deceleratory Heart Rate Analysis Stimulus Type Main Effect. Figure 13. Experiment II--Maximum Deceleratory Heart Rate Analysis Visual Field x Sex Interaction.

PAGE 85

77 Stimulus Type interaction [F_ (4, 112) = 3.35, £ = .0125)]. Simple effects testing revealed no significant effects across Stimulus Type for males [F_ ( M 56) = 1 86 £ = .130)]. However, for female subjects this Stimulus Type effect was significant [£ (4, 56) = 5.32, £ = .001)]. Duncan's post-hoc comparisons revealed that deceleratory HR responses to happy (M = -8.306), disgusting (M = -8.345) and neutral (M = -8.638) were all significantly greater than deceleratory HR responses to angry (M^ = -6.120) slides at £ < .05. These relationships are depicted in Figure 14. Maximum acceleratory responses A repeated measures ANOVA was performed with Sex (male, female) as the between subjects factor and Visual Field (left, right) and Stimulus Type (happy, angry, fearful, disgusting, neutral) as the within subjects factors. A summary of the results of this analysis are depleted In Table 6. Results revealed a significant interaction of Stimulus Type x Sex [_F (4, 112) = 3-48, £ = .0102)], depicted in Figure 15. Simple effects testing revealed no significant effect of Stimulus Type for males [F_ ( 4, 56) = 1.86, £ = .130)]. For female Ss however, this effect was significant l7_ (1, 14) = 5.32, £ = .001)]. Duncan's post-hoc comparisons revealed that angry (M^ = 3.473) and fearful (M = 3.888) slides elicited significantly greater HR accelerations than neutral (M = 2.158) slides at p < .05. This analysis also revealed a trend for an interaction between Visual Field x Sex [_F ( 1 28) = 3-36, £ = .0776)], with a pattern of greater acceleratory HR responses for males in LVF (M = 2.531) trials

PAGE 86

igure 14. Experiment II--Maximum Deceleratory Heart Rate Analysis, Sex X Stimulus Type Interaction.

PAGE 87

79 Table 6 Summary of Analysis of Variance: Experiment II, Heart Rate Acceleration Source df SS F £ Sex 1 28 28.2236 .65 Visual Field 1 28 .5214 .13 Visual Field x Sex 1 28 13.2770 3.36 Stimulus Type ^, 1 12 28.6964 1 .74 Stimulus Type x Sex 4, 1 1 2 57.5180 3.48 Visual Field x Stimulus Type 1 12 13.3162 .73 Visual Field x Stimulus Type x Sex 112 9.4844 .52 p < .01

PAGE 88

Figure 15. Experiment II--Maximum Acceleratory Heart Rate Analysis Stimulus Type x Sex Interaction. Figure 16. Experiment II — Maximum Acceleratory Heart Rate Analysis Visual Field x Sex Interaction.

PAGE 89

81 relative to RVF (M = 2.027) trials. This pattern of findings is depicted in Figure 16. Second by second responses Second by second post-stimulus HR changes for male and female Ss are displayed for each Visual Field x Stimulus Type condition in Figures 17 and 18, respectively. These values represent the average change from baseline for each of the eight post-stimulus seconds. Post-stimulus HR changes from baseline for both male and female Ss suggest that, overall, Ss showed primarily HR deceleratory changes throughout the presentation of the stimuli. This may be accounted for, in part, by task factors which required perceptual intake throughout the presentation of the stimulus. Perceptual Intake has been associated with HR deceleration as suggested by Laeey (1967). It is also of note that post-stimulus HR changes fr'om baseline do not reflect occurrence of acceleratory HR responses (i.e., positive HR change from baseline). It may be that the emotional stimuli themselves were not inherently "strong" enough to elicit acceleratory HR responses. However, findings from the analysis of maximum acceleratory HR responses do indicate that, at some point during the 3 post-stimulus seconds, Ss are experiencing HR acceleration. The most plausible explanation for the lack of post-stimulus HR accelerations in graphic representation of the 8 post-stimulus seconds Is that these acceleratory responses are occurring at different points in time across different trials and possibly across different Ss That is, accelerations may be "averaged out" by decelerations occurring at the same time on separate but like trials (i.e., same VF x Stimulus Type

PAGE 90

82 Males 0 T -6 I I I I I I I I I D1 D2 D3 D4 05 D6 D7 D8 Post-Stimulus Seconds Males 0 -I 01 02 03 04 05 06 07 08 Post-Stimulus Seconds Males Males 0 n -1 -2-3^ -4-5•oLVF-Fearful RVF-Fearful I I I • I I I I I 01 02 03 04 05 06 07 08 Post-Stimulus Seconds c 3 Vi CO £ o i CO 0 -1 -2-3-4 -5 oLVF-Oisgusting -• RVF-Disgusting I I I I I I I I 01 02 03 04 05 06 07 08 Post-Stimulus Seconds Figure 17. Experiment II--SeGond x Second Post-Stimulus Heart Rate Changes in Male Subjects for Each Visual Field x Stimulus Type Condition.

PAGE 91

33 0 •1 -2-3Females -oLVF-Happy -•RVF-Happy I • I I I I I I I I D1 D2 D3 D4 DS D6 D7 08 Post-Stimulus Seconds (0 _c "3 in (0 CD E o i 0 -1 -2-3-4-5Females -oLVF-Angry RVF-Angry I I — 1 — I 1 1 I — D1 D2 D3 04 05 06 07 08 Post-Stimulus Seconds c 1 03 E o i a. m Females -QLVF-Fearful RVF-Fearful 0 -1 -2-3-4 -5 I I I I I I I I 01 02 03 04 OS 06 07 08 Post-Stimulus Seconds Females c "3 m E o i Q. 0 -1 -2-3-4 -5 LVF-Oisgusting RVF-Disgusting I I I I I • 1 I I I 0^ 02 03 04 05 06 07 08 Post-Stimulus Seconds Females 01 02 03 04 05 06 07 D8 Post-Stimulus Seconds Figure 18. Experiment lI--Second x Second Post-Stimulus Heart Rate Clianges in Female Subjects for Each Visual Field x Stimulus Type Condition.

PAGE 92

84 condition). In the present experiment, Ss were required to also image a personal episode of the same valence as the stimulus. The time course of the Ss' Imaging could not be controlled and it may be the case that Ss differed in onset of their imaging and occurrence of affective aspects of the image. One observations which is of some interest is the fact that female Ss do show some indications of "autonomic patterning" in their HR changes across the 8 post-stimulus seconds. That is, females appeared to show less deceleration for fearful trials relative to other affective categories. Furthermore, this pattern appears to be reflected to a greater extent for LVF (right hemisphere) than RVF (left hemisphere) presentations. Skin Conductance Data Reduction For each trial, SCRs were depicted as the difference between the average of the skin conductance level during the 2 seconds preceding stimulus onset (tonic, baseline level) and the maximum skin conductance level during the 8 post-stimulus seconds (phasic level). This SCR value for each trial served as the dependent variable in a repeated measures ANOVA. Sex was the between subjects factor and Visual Field (left, right). Hand (left, right), and Stimulus Type (happy, angry, fearful, disgusting, neutral) were the within subjects factors Results of this ANOVA yielded no significant main effects or interaction effects. Table 7 depicts a summary of this analysis. A weak trend, however, was observed for a Sex x Visual Field x Stimulus

PAGE 93

85 Table 7 Summary of Analysis of Variance: Sxperiment II, Skin Conductance Responses Source df SS F p Sex 1 28 .5410 2 .60 Visual Field 1 28 .001 1 .08 Visual Field x Sex 1 28 .0004 .03 Hand 1, 28 .0056 .13 Hand x Sex 1 28 .0399 .94 Stimulus Type ^. 1 12 .0167 .36 Stimulus Type x Sex ^. 112 .0628 1 .35 Visual Field x Hand 1 28 .0007 .19 Visual Field x Hand x Sex 1 28 .0018 .43 Visual Field x Stimulus Type ^4. 1 12 .0664 1 .61 Visual Field x Stimulus Type x Sex 4, 1 1 2 .0868 2 .10 Hand x Stimulus Type 1 12 .0136 1 .31 Hand x Stimulus Type x Sex 112 .0097 .69 Visual Field x Hand x Stimulus Type ^. 1 12 .0249 1 .59 Visual Field x Hand x Stimulus x Sex Type 1 12 .01 12 .71

PAGE 94

86 Type interaction [£ (4, 112) = 2.10, _p = .0875)], depicted in Figure 19. This pattern of findings revealed for male Ss only a Visual Field X Stimulus Type interaction which approached significance. This pattern suggested greater SCR in LVF (M = .1302) relative to RVF (M = .0651) for angry slides. In addition, happy slides elicited greater SCR in RVF (M = .1268) compared to LVF (M = .0736) trials.

PAGE 95

Trend Males Only LVF RVF Happy Angry Fearful Disgusting Neutral Emotion Figure 19. Experiment II--Skin Conductance Response Analysis, Sex x Visual Field x Stimulus Type Trend.

PAGE 96

DISCUSSION Several models have been proposed to account for lateral symmetries observed on tasks of emotional processing. One model, the right hemisphere model, suggests that the right hemisphere is globally more involved in all aspects of emotional processing including the cognitive encoding/decoding, arousal-activation, and behavioral responses to emotional stimuli. In this view, emotional stimuli presented initially to the right hemisphere via the left sensory channel (left visual field, left ear) elicit significantly greater arousal responses and result in significantly quicker/more accurate detection than emotional stimuli presented initially to the left hemisphere via the right sensory channel (right visual field, right ear ) A second model, the hemispheric valence model, proposes that the left hemisphere is more adept at processing positive emotions and the right hemisphere is more adept at processing negative emotions. Within this framework, positive emotional stimuli initially presented to the left hemisphere would elicit significantly greater arousal responses and result in significantly quicker/more accurate detection than positive emotional stimuli presented initially to the right hemisphere. Likewise, negative emotional stimuli presented to the right hemisphere would elicit significantly greater arousal responses 88

PAGE 97

89 and result in significantly quicker/more accurate detection than negative emotional stimuli presented initially to the left hemisphere. A third model, the preparatory model, argues that this positivenegative dichotomy in hemispheric processing of emotions is artifactual and actually relates to differences in arousal and activation/preparation for action. According to this model, the right hemisphere is dominant for mediating arousal/activation and as such, is more intrinsically involved in processing emotional stimuli that have greater "preparatory" significance for survival (i.e., fightflight emotions). In contrast, the left hemisphere is more involved in mediating nonpreparatory emotions that place less immediate attentional demands on the organism for survival. Lastly, it is possible that the right hemisphere Is dominant for mediating arousal/activation responses to stimuli irrespective of tneir emotional/nonemotional content. That is, both emotional and nonemotional stimuli should result in significantly greater arousal/ activation responses when projected to the right hemisphere than the left hemisphere. The present study sought to further investigate these different conceptualizations of the hemispheric processing of emotional stimuli. The purposes of the study were to (a) determine the extent to which laterally presented emotional/nonemotional stimuli might result in differential patterns of benavioral activation (reaction time responses) as well as differential patterns of autonomic arousal (HR, SCR responses); (b) to determine whether there were hemispheric asymmetries in mediating arousal and/or activation responses to these

PAGE 98

90 etnotional/nonemotional stimuli; and (c) to determine whether certain categories of emotion (positive/negative; preparatory/nonpreparatory ) induce asymmetric arousal/activation, depending on the hemisphere to which they are initially presented. Findings from Experiment I, in which RTs were made to midline neutral stimuli that were preceded by lateralized stimuli of different emotional valences, failed to support any of the laterality models of emotion. For example, no overall superiority for lateralized emotional warning stimuli presented to the right versus left hemisphere was found. Likewise, no hemispheric specific emotional valence effects were observed. Similarly, no evidence was present for the view of hemispheric differences in "preparatory" versus "nonpreparatory" emotions. What was found, however, was the following: (a) females showed no laterality effects of any kind; and (b) males, on the other hand, had overall faster RTs to neutral stimuli that were preceded by emotional warning stimuli in the RVF (left hemisphere) versus LVF (right hemisphere). This finding, which suggests that emotional stimuli induce greater behavioral activation when presented to the left hemisphere than to the right hemisphere, is the opposite of that predicted by any model arguing for superiority of the right hemisphere in mediating emotional responsi vity There are several possibilities which might account for these findings. In Experiment I, Ss were required to make a left-right decision based on the emotional/nonemotionai nature of the warning stimulus. That is, they had to respond with one hand to emotional stimuli and with the other hand to neutral stimuli. In other words,

PAGE 99

91 Ss had to make a left-right discrimination judgement during the interstimulus interval between the WS and the imperative stimulus. It is well known that left-right discrimination seems to fail within the domain of left hemisphere functions (Benton, 1968; Gertsmann, ig'iO; Saugeut Benton, & Hecaen, 1971). Consequently, it is possible that this left-right discrimination inherent in the task demands may be related to the finding of faster RTs for male Ss when stimuli were presented to the left hemisphere versus the right hemisphere. The use of a go/no paradigm using a single hand for response would circumvent this possible confounding factor. Secondly, a variety of task strategies and stimulus factors can significantly influence the magnitude or even the direction of perceptual asymmetries (Moskovitch, 1986). Factors such as stimulus duration, spatial frequency, .stimulus clarity, and number and configuration of stimulus features can significantly influence observed lateral asymmetries (Bryden, 1973; Bryden & i^liard, 1976; Moskovitch, 1983; Sergent 1983; Sergent & Bindra, 1981). For example, Patterson and Bradshaw (1975) have reported that decreasing the dimensions along which facial stimuli differed decreased and even reversed the expected LVF superiority on a simultaneous face matching task It may be the case that in the present investigation the stimuli were sufficiently complex to warrant a change in processing strategy which relied perhaps to a greater degree on left hemispheric analysis of details and significant features in distinguishing among the different categories of stimuli. If this is the case, then the

PAGE 100

92 significant processing demands of the warning stimulus may have mitigated any potential activation or preparatory effect that the emotional stimuli may have had because processing of the warning stimulu continued through the interstimulus interval. The use of less complex warning stimuli such as emotional and neutral faces is one way of testing this hypothesis. In addition, it may also be of informational value to use such warning stimuli in a simple reaction time task to look at the general activation effect of emotional and neutral stimuli on simple reaction time to a neutral imperative stimuli. This paradigm would serve to eliminate the processing demands of the warning stimulus which may have interfered with the potential activation effect of emotional warning stimuli on right hemisphere processing. Lastly, it may be the case that emotional warning stimuli used in the present investigation were not arousing enough to provide adequate activation effects in response to the imperative stimulus. Findings the second experiment, in which measures of autonomic arousal were obtained to laterally presented emotional/nonemotional stimuli, were more in line with current views regarding hemispheric differences in processing emotional stimuli. With SCRs there were no significant effects in terms of SCRs to lateralized emotional stimuli. However, a trend was observed (Sex x VF p = .087), whereby male Ss had greater SCRs to happy stimuli when they were presented to tne RVF (left hemisphere) than to the LVF (right hemisphere); greater SCRs occurred when angry stimuli were presented to LVF (right hemisphere) than to the RVF (left hemisphere). Although this pattern

PAGE 101

93 of findings is consistent with both the valence model and the preparatory model of emotional proces3ing, these findings must be viewed with caution since significance was not obtained. The lack of robust findings may be due, in part, to the fact that SCRs for Ss were generally quite small, making detection of differences across visual field, hand, and stimulu type conditions difficult. With regard to HR responses that were also measured in Experiment II, happy and neutral slides induced significantly greater HR deceleratory responses than angry or fearful slides. This finding occurred regardless of which hemisphere initially received the stimulus. In other words, no emotional specific hemispheric arousal effects were present for HR deceleratory responses. Additionally, a trend was observed for greater HR deceler atory responses for stimuli directed to the right hemisphere (LVF) versus those directed to the left hemisphere (RVF). This finding was based primarily on the responses of male Ss as indicated by a significant Sex x VF interaction. Speci f Lcai ly males had significantly larger HR deceleratory responses to LVF stimuli than to RVF stimuli, and this effect was present regardless of the emotional/nonemotional content of tne stimuli. In contrast, females Ss displayed no laterality effect of any kind in their HR deceleratory responses to emotional or neutral stimuli. A similar pattern of findings was also observed for HR acceleratory responses, as reflected in a trend for a Sex x VF interaction (p = .077). Male Ss again tended to produce greater HR acceleratory responses to LVF stimuli than to RVF stimuli.

PAGE 102

94 In contrast to the hemispheric asymmetries in arousal responses observed in males, females did not show significant differences in HR arousal responses to stimuli across left and right visual field. However, females were differentially impacted by the emotional valence of the stimuli. This was revealed as a Sex x Stimulus interaction in analyses of HR deceleratory and HR acceleratory responses. Males did not show this differential pattern of responding to the emotional valence of the stimuli. Specifically, females had significantly greater heart rate deceleratory responses for happy, disgusting, and neutral trials compared to angry and fearful t'^ials. Similarly, females Ss had significantly greater HR acceleratory responses to angry and fearful stimuli r'elative to happy and neutral stimuli. Taken together, findings from both HR deceleratory and acceleratory responses, that wei^e obtained in Experiment II, suggest that greater HR arousal responses occurred when stimuli were presented to the right hemisphere (LVF) than when presented to the left hemisphere (RVF). This laterality effect occurred only for male Ss and was not dependent on the emotional/nonemotional content of the stimuli. That is, emotional and neutral stimuli directed to the right hemisphere resulted in comparable HR arousal effects. This finding is not consistent with the view that the right hemisphere is specifically dominant for mediating arousal responses only to emotional stimuli. Nor is it consistent with valence or preparatory models of emotional processing. Rather, this finding suggests that the right hemisphere of males is dominant for mediating arousal/activation, regardless of

PAGE 103

95 the emotional/nonemotional content of the stimuli and this right hemisphere dominance is not specific to emotional stimuli. Critical Issues Several issues are raised by these findings. The first concerns why there were no differences in arousal responses to emotional versus neutral stimuli. Other investigators have found that emotional stimuli induced greater arousal responses than do neutral stimuli (Hare & 31evings, 1975; Klorman et al 1977; Klorman & Ryan, 1980). One possibility accounting for the lack of differences in arousal responses to emotional versus neutral stimuli is that the emotional stimuli exerted some sort of "priming" effect, such that even the neutral stimuli were treated as "emotional." Kinsbourne (1970; 1973) has suggested that adoption of a cognitive set can serve to asymmetrically arouse or prime the hemispheres for processing. It may have been the case in this study that emotional stimuli exerted a priming effect in which neutral stimuli were treated as emotional stimuli A second issue concerns why the laterality effect in HR responses were observed only in male Ss In a major review of the literature on sex differences, McGione (1980) concludes that overall males possess a significantly greater degree of observed asymmetry relative to female subjects. While this conclusion has been criticized on a number of grounds, the weight of the evidence does suggest that, at least for language functions, the hemispheric representation of males may differ from that of females. That is, on measures of severity of aphasia and

PAGE 104

96 recovery from aphasia following hemispheric lesions, it appears that males are more left hemisphere lateral i zed for language, whereas females have more bilateral representation of speech and language functions (McGlone, 1980). With respect to arousal asymmetries, studies have not systematically addressed sex differences. Prior investigations which have looked at the right hemisphere's role in production of arousal responses in normal, neurologically intact Ss have not investigated the effect of sex on production of laterality effects Ln HR and SCR (Hugdal et al 1983; Hugdahl et al 1982; Walker & Sandman, 1982). Heart rate findings from the present investigation suggest that the greater role of the right hemisphere in production of arousal responses may exist for male Ss to a greater extent than for female Ss; findings which are consonant with aeuropsychological investigations of laterality effects across sex. In addition to studies which have looked at the generation of autonomic arousal (i.e., HR SCR), a related area of research has also investigated cerebral asymmetries in heart beat perception and detection Davidson, Horowitz, Schwartz, and Goodman (1981) measured RT differences between R-wave occurrence and key press latencies. These authors report that finger taps by the left hand (right hemisphere) had shorter mean latencies from heartbeat than did taps from the right hand (left hemisphere), suggesting that the right hemisphere is more sensitive or more efficient than the left hemisphere in response tracking to cardiac afferent feedback.

PAGE 105

97 Similarly, Hantas Katkin, and Reed (1934) using male subjects only demonstrated that right hemisphere preferent individuals (as indexed by conjugate lateral eye movements) performed significantly better than left hemisphere preferent individuals on a task of heart beat detection. Furthermore, these findings were significant both before and after heart rate discrimination training. Montgomery and Jones (1984) reported similar findings in male subjects, with right hemisphere preferent subjects performing significantly better on a heart beat perception task relative to left hemisphere preferent subjects. In addition, they also found higher emotionality scores for right hemisphere preferent individuals relative to left hemisphere preferent individuals. It is of interest to note that these studies which find significant laterality effects on tasks of heart beat perception and detection have used primarily male subjects. This occurrence may, in part account, for significant laterality effects observed on these tasks as male subjects often reveal a significantly greater degree of cerebral lateralization than female subjects (McGlone, 1980). A consistent finding in the cardiac awareness literature is that male subjects are superior to female subjects in detecting heart beats and heart rate across a variety of psychophysiological paradigms (Pennebaker & Hoover, 1984; Whitehead, Drescher, Heiman, & Blackwell, 1977). More recent research (Rouse, Jones, ?t Jones, 1988), however, has suggested that such sex differences in cardiac awareness are accounted for primarily by body fat composition [% fat) and general fitness differences across male and female subjects. By controlling

PAGE 106

98 Tor level of body fat, Rouse et al (1988) did not find the gender effect previously documented in the literature. Furthermore, sex differences were found only when level of body fat differed between males and females In the present investigation, a main effect of Sex was observed in Experiment II for maximum HR deceleratory responses. More specifically, male subjects had significantly greater HR decelerations than female subjects. It may be the case that body fat composition and general fitness also plays some role in magnitude of cardiac decelerations although this relationship is unclear at present. To briefly summarize, findings from Experiment II provide support for the greater role of the right hemisphere in mediating arousal responses, at least for male subjects. That is, males showed gr-eater HR deceleratory and acceleratory responses to stimuli when they were presented to the LVF (right hemisphere) tnan to the RVF (left hemisphere). This concerns the basis for these sex differences. The evolutionary significance of lateralized heart rate responses in male subjects can be viewed from the more general perspective of the evolutionary significance of greater cerebral lateralization in males. Flor-Henry (cited in McGlone, 1930) provides an interesting argument in this regard. He notes that dopaminergic pathways in rodents are known to be asymmetrical and related to directional preferences and turning behaviors witn the direction of these behaviors contralateral to the hemisphere with greatest dopamine concentrations (Glick, Jerussi & Zimmerberg, 1977). Turning behaviors in rodents, cats, and dogs are associated with fighting and

PAGE 107

99 sexual display. Furthermore, he notes that the lateralized control of •3ong in the left brain of canary and chaffinch is related to mate attraction and territoriality, i.e., sex and spatial analysis (Nottebohm, 1977; Webster, 1977). Macn (1959) suggests that this neural asymmetry serves to increase the efficiency of spatial analysis in an environment where no systematic left/right bias exists. Furthermore, he argues the evolutionary advantage of this increased efficiency is related to mate attraction and in species where the male actively seeks the female, this greater lateralization for male organisms is of adaptive significance and survival value. As previously discussed, female subjects did not show significant differences in HR deceleratory responses or HR acceleratory responses across left and right visual fields as found Ln male subjects. However, female subjects were differentially impacted by the emotional valence of the stimuli (revealed as a significant interaction of Sex x Stimulus Type for analysis of both HR deceleratory responses and HR acceleratory responses). Male subjects, in contrast, did not show this differential heart rate responding across stimulus categories. Analysis of deceleratory HR responses for female subjects revealed significantly greater decelerations for happy, disgusting, and neutral trials relative to fearful trials. Similarly, analysis of acceleratory HR responses for female subjects revealed sign! ficantly greater acceleratory responses for angry and fearful slides compared with neutral slides. These results are consistent with those of other

PAGE 108

100 investigators who report HR accelerations in response to fearful stimuli (Hare & Blevings, 1975; Klorman & Ryan, 1980; Vrana et al., 1986). In addition, these results also parallel the findings of Ekman et al (1983) who reported HR increases in response to production of emotional facial expressions of anger, fear, and sadness and HR decelerations in response to disgust, surprise, and happiness. While previous research has suggested that different patterns of autonomic arousal are associated with different types of emotional states, in the present study differences in autonomic responses across different emotional categories were observed only for female subjects. One critical question concerns why this "autonomic patterning" to emotional stimuli should occur in females but not males. It may be the case that female subjects were more amenable to fully cooperating with task demands of emotional imagery. This suggestion is supported by prior investigations which have found significant effects for mood manipulation and emotional imagery instruction for female but not male subjects (Delp & Sackeim, 1987; McKeever & Dixon, 1981). Alternatively, sex differences in autonomic patterning observed in the present study may be related to differences in imagery ability between males and females. Prior investigations have revealed that differences in autonomic control and HR conditioning are related to subject's imagery ability. Carroll, Baker, and Preston (1979) have reported that ability to increase HR through voluntary imaging was significantly correlated with reported vividness of the subject's images. Ikeda and Hirai (1976) reported that ability to control SCR

PAGE 109

101 through biofeedback training was associated with the frequency of subject's images. In addition, Kuzendorf (1982) also found that subject's ability to produce temperature differences across the hands was significantly correlated with prevalence of imagery. Lastly, Arabian and Furedy (1983) report that HR deceleratory conditioning (conditioned response) in good imagers was more similar to HR deceleratory unconditioned responses than HR deceleratory conditioning (conditioned responses) for poor imagers. In the present investigation, imagery ability was not assessed. However, it is possible that the current findings which suggest that female subjects showed differntial autonomic responding across various emotional stimuli might be related to potential differences in imagery ability between female and male subjects. For example, tnere is some evidence in the literature to suggest that female subjects, in general, are better imagers than male subjects (White, Sheehan, & Ashton, 1977). This finding has led some investigators to use only female subjects in investigations of imagery and autonomic responding (Jones i Johnson, 1978; 1980). Regardless of the basis for the sex differences, the differential pattern of HR responses to the various emotional stimuli by female subjects in the present study is similar in pattern to that reported by other investigators (Ekman et al 1983; Hare & Blevings, 1975; Klorman & Ryan, 1980). Findings from the present investigation suggest that HR decelerations were significantly greater for both happy and neutral slides relative to angry and fearful slides.

PAGE 110

102 Similarly, angry and fearful slides elicited significantly greater HR accelerations than neutral slides. A question which arises concerns the basio for the differential HR patterning across different emotion categories. Recently, several authors have systematically examined the relationship between emotional imagery and production of cardiac responses. Jones and Johnson (1978; 1980) have argued that two major factors contribute to the particular patterning of HR responses. These include (a) the emotional content of the image (pleasant vs unpleasant); and (b) the level of activity inherent in the image (high activity vs low activity). By systematically manipulating these two factors, Jones and Johnson (1980) found that high activity images produced significantly greater HR accelerations than low activity images. These findings are congruent with reports from other investigators who have noted that instructional manipulations on imagery tasks can produce significant effects on autonomic responses (Lang, 1977; Melamed, 1969). Similarly, by requesting active engagement in stimulus processing, Bauer and Craighead (1979) also revealed significant effects on psychophysiological responses. Jones and Johnson (1980) also report that negative-low activity images resulted in greater HR accelerations relative to positive-low activity images. Such findings suggest that the emotional valence of the image was also a significant contributing factor to differential HR responses. While the effect of valence of affective stimulus on HR responding had been previously investigated, these authors provide

PAGE 111

103 support for the role of this factor when differences in activity level of imagery are accounted for. It may be the case that a priori differences in activity level across emotional categories could, in part, account for differential patterns of HR acceleration across emotional categories. Lang (1979) has suggested that emotional imagery results in patterns of autonomic activity very similar to those found in the actual emotional situation. In this view, those emotions which inherently contain a greater motor component (i.e., flight or fight type of responding) as in the case of anger or fear, would result in significantly greater HR acceleratory responses than those emotions with less inherent motor demand; predictions which parallel the findings in the present investigation It may also be the case that these changes in HR activity across various emotional categories may serve to differentially prepare the organism for action. In accounting for differences in acceleratory and deceleratory HR responses, Obrist and colleagues (197^) have emphasized the relationship between motor requirements and cardiac activity. In addition, these authors suggest that conditions may exist whereby increases in HR are observed without overt changes in somatic activity. In support of this view, Freyschuss (1970) reported cardiac accelerations under conditions where subjects were instructed to tense or move an arm despite their inability to do so because of experimentally induced paralysis. Based on these findings, it is suggested that cardiac activity is not solely coupled with direct, overt somatic activity but rather that cardiac activity is coupled

PAGE 112

104 with real as well as intended somatic activity (as in the case of imagery) One additional question concerns the fact that findings from Experiment II provided support for the greater role of the right hemisphere in production of arousal responses but results from Experiment I failed to provide support for a greater role of the right hemisphere in production of arousal responses. Differences across the two experiments may help to clarify some of this apparent discrepancy. First, the most plausible explanation for these discrepant findings is that in Experiment I subjects were required to make a left-right decision based on the emotional/nonemotional nature of the warning stimulus. As previously discussed, It is likely that this left-right discrimination (a predominantly left hemisphere ability) inherent in the task demands of Experiment I is related to the finding of faster RTs for male Ss when stimuli were presented to the RVF (left hemisphere). Secondly, exposure durations in the two experiments differed quite significantly. Experiment I utilized relatively short exposure durations while Experiment II utilized quite lengthy exposure durations. The overall complexity of the stimuli together with the relatively short exposure durations used in Experiment I versus Experiment II may have contributed to these divergent findings. Similarly, the task demands of the two experiments were quite different. Requirements inherent in Experiment II in which subjects were asked to generate very personal and likely very meaningful episodes from their own lives may have contributed to the overall

PAGE 113

105 effectiveness of this procedure in demonstrating differential and iateralized heart rate responses in female and male Ss respectively. Conclusions Findings from Experiment I, in which RTs were made to midline neutral stimuli that were preceded by Iateralized stimuli of different emotional valences, failed to support any of the laterality models of emotion. No overall superiority for RTs to LVF (right hemisphere) versus RVF (left hemisphere) trials was found. In addition, no evidence was found for hemispheric specific emotional valence effects. Likewise, no evidence was present for the view of the hemispheric differences in preparatory versus nonpreparatory emotions. Rather, faster RTs were observed for male subjects when WS were presented to RVF (left hemisphere). This finding is not consistent with any of the proposed hemispheric models of emotional processing. The most plausible interpretation of these findings is that subjects were required to make a left-right decision based on the emotiondl/nonemotional nature of the warning stimulus and that this left hemisphere mediated task requirement accounts for the faster RTs of male subjects to RVF (left hemisphere) presentations of WS Findings from Experiment II, in which measures of autonomic arousal (HR and SCR) were obtained to laterally presented emotional/nonemotional stimuli, were more congruent with models of hemispheric differences in processing emotional stimuli. Males showed Iateralized effects of HR arousal responses which supports greater right hemisphere involvement in production of arousal responses.

PAGE 114

106 However, this effect was present regardless of the emotional/ nonemotional content of the stimulus. This finding is not consistent with the view that the right hemisphere is specificaliy dominant for mediating arousal responses only to emotional stimuli. Nor is it consistent with valence or preparatory models of emotional processing. Rather, this finding suggests that the right hemisphere of males is dominant for mediating arousal responses, regardless of the emotional/noneraotional content of the stimulus and this right hemisphere dominance is not specific to emotional stimuli. However, SCRs did suggest some findings which are consistent with emotion specific hemispheric arousal effects. This was revealed in a trend for a Sex x VF interaction (p_ = .08?) effect, in which male subjects had greater SCRs to happy stimuli when they were presented to the RVF (left hemisphere); greater SCRs occurred when angry stimuli were presented to LVF (right hemisphere) than to RVF (left hemisphere). These findings are also consistent with the preparatory model of emotional processing. The question is raised why emotion specific hemispheric arousal effects are found for SCR but not for HR responses. Prior investigations have revealed that SCR are more related to intensity of emotional stimuli whereas HR responses were more related to valence of emotional stimuli (Greenwald, Lang, & Cooke, 1988). It may be that intensity differences existed across the emotional categories and this might account for the discrepancy in findings. However, this is not likely the case as intensity ratings across the different emotional categories were similar. No plausible explanation is available for

PAGE 115

I 107 the lack of emotion specific effects in HR responses together with the possible emotion specific effects in SCRs The other consideration is that the SCR findings represent only a trend and may not truly effect an emotion specific hemispheric effect. The findings of female Ss are not consistent with any hemispheric emotionality models. In contrast to laterality effects observed for male Ss female Ss appeared to show a qualitatively different pattern of HR responding which differentiated their performance from male Ss This was revealed in a significant Sex x Stimulus Type interaction for analysis of HR deceleratory responses and HR acceleratory responses. This pattern of findings demonstrated that female but not male Ss snowed differential deceleratory as well as acceleratory HR responses across the different conditions of stimulus type with significantly greater acceleratory responses to angry and fearful slides compared with neutral slides and significantly greater leceleratory responses to happy, disgusting, and neutral slides relative to fearful slides. These findings also provide further support for the view that different patterns of autonomic arousal may be associated with different types of emotional states. In addition, they also suggest that possible differences in imagery ability across male and female Ss may have, in part, mediated this differential pattern of responding. These findings point to the relative importance of considering both sex and imagery ability of 3s in further investigations of emotional processing and autonomic responding. Sex differences in degree of lateral asymmetry of arousal responses suggest that the

PAGE 116

108 greater role of the right hemisphere in production of arousal responses exists to a significantly greater extent in male Ss versus female Ss; findings which are congruent with findings from neuropsychological investigations which report a greater degree of lateral asymmetries for male Ss on a variety of tasks. Similarly, sex differences in patterning of HR responsivity across different emotional categories suggests that female but not male Ss reveal differences in magnitude of HR acceleratory and HR deceleratory responses across various emotional categories. The extent to which this differential autonomic responding in male versus female Ss is due to differences in imagery ability across these groups is an area of future investigation.

PAGE 117

REFERENCES Ahern, G. L., & Schwartz, G. E. (1979)Differential lateralization for positive versus negative emotions. Neuropsychologia 1 7 693-697. Ainsworth, M. D. S., Blehar M., Waters, E., & Wall, E. (1978). Patterns of attachment Hillsdale, NJ: Erlbaum. Alema, G., Rosadini G., & Rossi, G. F. (1961). Psychic reactions associated with intracarotid amytal injections and relation to brain injury. Excerpta Medica 37, 15^4-155. Arabian, J. M., & Furedy J. J. (1983). Individual differences in imagery ability and pavlovian heart rate decelerative conditioning. Psychophysiology 20 325-331 • Ax, A. F. (1953). The physiological .li ff erentiat ion between fear and anger in humans. Psychosomatic Medicine 15 433-442. Babinski J. (1914). Contribution a I'^tude des troubles mentaux dans 1 hemisplegie organique c^r^brale (anosognosie ) Revue Neurologique 27, 845-848. Bauer, R. M., 4 Craighead, W. E. (1979). Psychophysiological responses to the imagination of fearful and neutral situations: The effects of imagery instruction. Behavior Therapy 10 389-403. Bayley, N. (1968). Behavioral correlates of mental growth: Birth to 36 years. American Psychologist _1_, 1-17. Bayley, N. (1969). Bayley Scales of infant development: Birth to two years New York: Psychological Corporation. Bear, D. M., i Fedio, P. (1977). Quantitative analysis of interictal behavior in temporal lobe epilepsy. Archives of Neurology 34 454-467. Benson, D. F., & Geschwind, N. (1971). Aphasia and related cortical disturbances. In: A. B. Baker & L. H. Baker (Eds.), Clinical Neurology New York: Harper & Row. Benton, A. (1968). Right-left discr^imination. Pediatric Clinics of North America, 15, 747-758. 109

PAGE 118

no Berg, M. R., & Harris, L. J. (1980). The effect of experimenter location and subject anxiety on cerebral activation measured bylateral eye movements. Neuropsychologia 18 89-9^. Berger H. (1933). Uber das electroenkephalogram des menschen. Archiv fur Psychiatrie und Nervenkrankhelten 99 555-57'^. Bihrle, A. M., Brownell, H. H., & Powelson, J. A. (1986). Comprehension of humorous and nonhumorous materials by left and right brain damaged patients. Brain and Cognition 5, 399-411. Borod, J. C, c Caron, H. S. (1980). Facedness and emotion related to lateral dominance, sex and expression type. Neuropsychologia 28, 237-242. Borod, J. C, Caron, H. S., & Koff, J. (1981). Asymmetry in positive and negative facial expressions: Sex differences. Neuropsychologia 1 9 819-824. Borod, J. C, Kent, J., Koff, E., Martin, C, & Alpert, M. (1988a). Facial asymmetry while posing positive and negative emotions: Support for the right hemisphere hypothesis. Neuropsychologia 26, 759-764. Borod, J. C, Koff, E., Perlman, E., Lorch, M., & Nicholas, M. (1986). The expression and perception of facial emotion in brain damaged patients. Neuropsychologia 24 169-130. Borod, J. C, Koff, E., & White, B. (1983). Facial asymmetry in posed and spontaneous expressions of emotion. Brain and Cognition 2, 165-175. Borod, J. C, Vingiano, W., 4 Cytryn, F. (1988b). The effects of emotion and ocular dominance on lateral eye movement. Neuropsychologia 26, 213-220. Bower, G. H. (1981). Mood and memory. American Psychologist 36 129-148. Bowers, D., Bauer, R., Coslett, B., & Heiiman, K. (1985). Processing of faces by patients with unilateral hemispheric lesions. I. Dissociation between judgements of facial affect and facial identity. Brain and Cognition _4, 258-272. Bowers, D., & Heiiman, K. (1984). Dissociation between the processing of affective and non-affective faces: A case study. Journal of Clinical Neuropsychology _6, 367-374, Bowlby, J. (1972). Separation: Volume II. Attachment and loss New York: Basic Books.

PAGE 119

1 1 Briggs, G. G., & Nebes R. D. (1975). Patterns of hand preferences in a student population. Cortex 1 1 230-238. Brody, N., Goodman, S. E., Halm, E., Krinzman, S., & Sebrechts, M. M. (1987). Lateralized affective priming of lateralized affectively valued target words. Neuropsychologia 25 935-9^^6. Brownell, H. H., Potter, H. H., Michelow, D., & Gardner, H. (198^4). Sensitivity to lexical denotation and connotation in braindamaged patients: A double dissociation? Brain and Language 22, 253-265. Bruder, G. E. (1983). Cerebral laterality and psychopathology : A review of dichotic listening studies. Schizophrenia Bulletin _9> 1 43-151 Brumback, R. A., & Staton, R. D. (1981). Depression-induced neurological dysfunction. New England Journal of Medicine 304 642. Bryden, P. (1978). Strategy effects in hemispheric asymmetry. In: G. Underwood (Ed.), Strategies of information processing New York: Academic Press. Bryden, P., & AlLard, F. (1976). Visual hemifield differences depend on typeface. Brain and Language 3_, 191-200. Bryden, P., 4 Ley, R. G. (1983). Right hemisphere involvement in the perception and expression of emotion in normal humans. In: K. M. HeiLman (Ed.), Neuropsychology of human emotion New York: Guillford Press. Bryden, M. P., Ley, R. G., & Sugarman J H (1982). A left ear advantage for identifying the emotional quality of tonal sequences. Neuropsychologia 20 83-87. Buchtel, H. A., Campari, F., DeRisio, C, & Rota, R. (1978). Hemispheric differences in discriminative reaction time to facial expressions. Italian Journal of Psychology _5, 159-169. Buck, R., i Duffy, R. J. (1980). Nonverbal communication of affect in brain damaged patients. Cortex 1 6 351-362. Cannon, W. B. (1927). The James-Lange theory of emotions: A critical examination and an alternative theory. American Journal of Psychology 34, 106-124. Carmon, A., & Nachson, I. (1973). Ear asymmetry in perception of emotional nonverbal stimuli. Acta Psychologica 37, 351-357.

PAGE 120

112 Carroll, D., Baker, J & Preston, M. (1979). Individual differences in visual imaging and the voluntary control of heart rate. British Journal of Psychology 70 39-49. Charlesworth W. B. (1964). Instigation and maintainence of curiosity behavior as a function of surprise versus novel and familiar stimuli. Child Development 35, 1169-1186. Cicone, M., Wapner W., & Gardner, H. (1980). Sensitivity to emotional expression and situations in organic patients. Cortex 26, 145-158. Cimino, C, Verfaellie, M,, Bowers, D., & Heilman, K. (1988). Emotional autobiographical memory in right hemisphere damaged patients Unpublished manuscript. Coulbourn, C. J., & Lishman, W. A. (1974j. Lateralization of function and psychiatric illness. A left hemisphere deficit? In: J. Gruzelie'^ & P. Flor-Henry (Eds.), Hemispheric asymmetries of function and psychopathology Amsterdam: Slseview. Cummings, J. L., & Mendez, M. P. (1984), Secondary mania with focal cerebrovascular lesions. American Journal of Psychiatry 141 108J4-1087. Davidson, R. J., Horowitz, M. E., Schwartz, G. E., 4 Goodman, D. M. (1981). Lateral differences in the latency between finger tapping and the heart beat. Psychophysiology 18 36-41. Davidson, R. J., & Schwartz, G. E. (1976). Patterns of cerebral lateralization during cardiac biofeedback versus the selfregulation of emotion: Sex differences. Psychophysiology 1 3 62-68. Davidson, R. J., Schwartz, G. E., Saron, C, Bennett, J., 4 Goiemena, D. J. (1979). Frontal vs. parietal EEG asymmetry during positive and negative affect. Psychophysiology 1 6 202-203. Davies, J. B. (1978). The psychology of music London: Hutchinson. DeKosky, S., Heilman, K. M., Bowers, D., & Valenstein, E. (1980). Recognition and discrimination of emotional faces and pictures. Brain and Language 9^, 206-214. d'Elia, G., & Perris, C. (1973). Cerebral functional dominance and depression: An analysis of EEG amplitude in depressed patients. Acta Psychiatrica Scandinavica 49, 191-197.

PAGE 121

113 d'Slia, G., & Ferris, C. (1974). Cerebral functional dominance and memory functions: An analysis of EEG integrated amplitude in depressive psychotics. Acta Psychiatrica Scandinavica (Supplement) 255, 143-157. Delp, M. J., & Sackeim, H. A. (1987). Effects of mood on lacrimal flow: Sex differences and asymmetry. Psychophysiology 24 550-556. Denny-Brown, D., Meyer, J. S., & Horenstein, S. (1952). The significance of perceptual rivalry resulting from parietal Lesions. Brain 75, 434-471. DeWitt, G. W. (1978). Laterality, personality and the perception of emotional stimuli. Dissertation Abstracts International 38(8-B) 3850. Dimond, S. J., Farrington, L., & Johnson, P. (1976). Differing emotional response from right and left hemispheres. Nature 261 690-692. Dopson, W. G., Beckwith, B. F., Tucker, D. M., & Bui lard-Bates P. C. (1984). Asymmetry of facial expression in spontaneous emotion. Cortex 20, 243-252. Ehrlichman, H., 4 Weinberger, A. (1978). Lateral eye movraments and hemispheric asymmetry: A critical review. Psychological Bulletin 85, 1080-1101. Ekman, P. (1972). Universals and cultural differences in facial expressions of emotion. In J. K. Cole (Ed.), Nebraska symposium on motivation Lincoln: University of Nebraska Press. Ekman, P., Hager J. C, & Friesen, W. V. (1981). The asymmetry of emotional and deliberate facial actions. Psychophysiology 1 8 101-106. Ekman, P., Levenaon, R. W., & Friesen, W. V. (1983). Autonomic nervous system activity distinguishes among emotions. Science 221 1208-1210. Eliinwood, E. H. (1967). Amphetamine psychosis: 1. Description of the individuals and process. Journal of Nervous and Mental Disease, 144, 273-283. Fehr F. S., & Stern, J. A. (1970). Peripheral psychological variables and emotion: The James-Lange theory revisited. Psychological Bulletin 74 411-424. Flor-Henry, P. (1969). Psychosis and temporal lobe epilepsy: A controlled investigation. Epilepsia, 10, 363-395.

PAGE 122

1U Flor-Henry, P. (1976). Lateraiized temporal limbic dysfunction and psychopathology Annals of the New York Academy of Scieaces 280 777-795. Flor-Henry, P. (1979). On certain aspects of the localization of the cerebral systems regulating and determining emotion. Biological Psychiatry J_M, 677-698. Floi — Henry, P. (1983). Functional hemispheric asymmetry and psychopathology. Integrative Psychiatry 46-52. Flor-Henry, P., & Koles, Z. V. (1980). EEC studies in depression, mania and normals: Evidence for partial shifts of laterality in the affective psychoses. Advances in Biological Psychiatry _it, 21-43. Fox, N. A., & Davidson, R. J. (1984). Hemispheric substrates of affect: A developmental model. In: N. J. Fox & R. J. Davidson (Eds.), The psychobiology of affective development Hillsdale, NJ: Erlbaum. Freeman, R. L., Galaburda, A. M., Cabal, R. D., & Geschwind, N. (1935). The neurology of depression. Archives of Neurology 42, 289-291 Freid, I., Mateu, C, Ojemann, G., Wohns J., & Fedio, P. (1982). Organization of visuospatial function in human cortex. Evidence from electrical stimulation. Brain 1 05 349-371 French, J. E., Hernandez-Peon, R,, & Livingston, R. v1955). Projections from the cortex to cephalic brainstem (reticular formation) in monkey. Brain 18, 74-95. Freyschuss V. (1970). Cardiovascular adjustments to somato-motor activation: The elicitation of increments in heart rate, aortic pressure and venomotor tone with the initiation of muscle contraction. Acta Physiologica Scandinavica 342, I-63. Gage, D. F., & Safer, M. A. (1979). Hemispheric differences in the mood state-dependent effect for recognition of emotional faces. Journal of Experimental Psychology: Learning, Memory and Cognition ri_' 752-763. Gainotti G. (1972). Emotional behavior and hemispheric side of the lesion. Cortex _8, 41-55. Galin, D., Diamond, R., & Braff, D. (1977). Lateralization of conversion symptoms: More frequent on the left. American Journal of Psychiatry, 1 34, 578-530.

PAGE 123

115 Gasparrini, W. G., Satz, P., Heilman, K. M., & Coolidge, F. L. (1978). Hemispheric asymmetries of affective processing as determined by the Minnesota Multiphasic Personality Inventory. Journal of Neurology, Neurosurgery and Psychiatry 41 470-473Gertsmann, J. (1940). Syndrome of finger agnosia, disorientation for right and left, agraphia and acalculia. Archives of Neurology and Psychiatry _44, 398-408. Geschwind, N. (1969). Problems in the anatomical understanding of the aphasia. In A. Benton (Ed.), Contributions to clinical neuropsychology Chicago: Aidine. Geschwind, N. (1975). The apraxias: Neural mechanisms of disorders of learned movement. American Scientist 63 188-195. Click, 3. D., Jerussi, T. P., & Zimmerberg, B, (1977). Behavioral and neuropharmacological correlates of nigrostriatal asymmetry in rat. In: S. Harnad, R. Doty, L. Goldstein, J. Jaynes, & G. Krauthamer (Eds.), Lateralization in the nervous system New York: Academic. Glick, S. D., Meibnach, R. C, Cox, R. D., & Maayani S. ( 1979). Multiple and interrelated functional asymmetries in rat brain. Life Sciences 25, 395-400. Goldstein, K. (1948). Language ani language disturbances New York: Grune & Stratton. Goldstein, K. (1952). The effect of brain damage on the personality. Psychiatry 1 5 245-260. Goldstein, S. G., Filskov, S., Weaver, L. A., i Ives, J. 0. (1977). Neuropsychological effects of electroconvulsive therapy. Journal of Clinical Psychology 33, 798-806. Graham, F. K. (1980). Representing cardiac activity in relation to time. In: I. Martin and P. Venables (Eds.), Techniques in psychophysiology New York: Wiley. Graves, R., Landis, T., & Goodglass H. (1980). Laterality and sex differences for visual recognition of emotional and non-emotional words. Paper presented at the meeting of the Academcy of Aphasia, Cape Cod, MA. Graves, R., Landis, T., & Goodglass, H. (1981). Laterality and sex differences for visual recognition of emotional and non-emotional words. Neuropsychologia, 19, 95-102.

PAGE 124

116 Greenwald, M., Lang, P., & Cooke, E. (1988). Response patterning du ring afrective visual stimulation; Evidence for a dimensional model of emotion Unpublished manuscript. Gur, R. E., & Gur, R. C. (1975). Defense mechanisms, psychosomatic symptomatology and conjugate lateral eye movements. Journal of Consulting and Clinical Psychology 43 416-420. Haggard, M. P., & Parkinson, A. M. (1971). Stimulus and task factors as determinants of ear advantages. Quarterly Journal of Experimental Psychology 23 168-177. Hansch, E. C, & Pirozzolo, F. J. (1980). Task relevant effects on the assessment of cerebral specialization for facial emotion. Brain and Language 10 51-59. Hantas, M. N., Katkin, E. S., & Reed, S. D. (1984). Cerebral lateralization and heartbeat discrimination. Psychophysiology 2A_, 241-278. Hare, R. D., & Blevings, G. (1975). Defensive responses to phobic stimuli. Biological Psychology _3, 1-13. Hecaen, H. (1962). Clinical symptomatology in right and left hemispheric lesions. In: V. B. Mountca3tle (Ed.), Interhemispher ic relations and cerebral aominance Baltimore: Johns Hopkins University Press. Heiiman, K. M., Bowers, D., Speedie, L., & Coslett, B. (1984). Comprehension of affective and nonaffective prosody. Neurology 34, 917-921 Heilman, K. M., Scholes R., & Watson, R. T. (1975). Auditory affective agnosia: Disturbed comprehension of affective speech. Journal of Neurology, Neurosurgery and Psychiatry 38 69-72. Heilman, K. M., Schwartz, H. D., & Watson, R. T. (1978). Hypoarousal in patients with the neglect syndrome and emotional indifference. Neurology 28, 229-232. Heilman, K. M., & Valenstein, E. (1972). Frontal lobe neglect. Neurology, 22, 660-664. Heilman, K. M., 4 Van Den Abel, T. (1979). Right hemispheric dominance for mediating cerebral activation. Neuropsychologia 17, 315-321.

PAGE 125

117 Heilman, K. M., Watson, R. T., & Bowers, D. (1983). Affective disorders associated with hemispheric disease. In: K. M. Heilman (Ed.), Neuropsychology of human emotion New York: Guiliford Press. Heller, W., & Levy, J. (1981). Perception and expression of emotion in right handers and left handers. Neuropsychologia 19 363-372. Holloway, F. A., & Parsons, 0. A. (1969). Unilateral brain damage and bilateral skin conductance levels in humans. Psychophysiology 6, 138-1U8. Hommes, 0. R., & Panhuysen, L. H. H. M. (1971). Depression and cerebral dominance: A study of bilateral intracarotid amytal in eleven depressed patients. Psychiatria, Neurologia, Neurochirurgia 74 259-270. Hugdahl, K., Franzon, M., Anderson, B., & Walldebo, G. (1983). Heart rate responses to lateralized visual stimuli, Pavlovian Journal of Biological Sciences 18, 186-198. Hugdahl, K., Wahlgren, C, & Wass T. (1982). Habituation of the electrodermal orienting reaction is dependent on the cerebral hemisphere initially stimulated. Biological Psychology 15 49-62. Ikeda, Y., & Hirai H. (1976). Voluntary control of electrodermal activity in relation to imagery ana internal perception scores. Psychophysiology 1 3 330-333Iversen, S. D. (1977). Brain dopamine systems and behavior. In: L. L. Iversen, S. D. Iversen, & S. H. Snyder (Eds.), Handbook of psychopharmacology Volume 8. Drugs, neurotransmitters and behavior. New York: Plenum Press. Izard, C. E. (1977). Visual attention in infancy: Processes, methods and clinical applications. In: I. C. Uzgiris & T. Weizmann (Eds.), The structuring of experience New York: Plenum Press. Izard, C. E., Hubner R. R., Risser, D., McGuiness, G. C, ?e Dougherty, L. M. (1980). The young infant's ability to produce discrete emotional expression. Developmental Psychology 1 6 132-140. Johnson, 0., & Crockett, D. (1982). Changes in perceptual asymmetries with clinical improvement of depression and schizophrenia. Journal of Abnormal Psychology, 91, 45-54.

PAGE 126

118 Jones, G. E., 4 Johnson, H. J. (1978). Physiological responding during self-generated imagery of contextually complete stimuli. Psychophysiology 1 5 439-446. Jones, G. E., & Johnson, H. J. (1980). Heart rate and somatic concomitants of mental imagery. Psychophysiology 17 339-347. Karlin, R., Weinapple, M., Rochford, J., & Goldstein, L. (1979). Quantitated EEC features of negative affective states: Report of some hypnotic studies. Research Communications in Psychology, Psychiatry and Behavior _4, 397-413King, F. L., & Kimura, D. (1972). Left ear superiority in dichotic perception of vocal nonverbal sounds. Canadian Journal of Psychology 26, 111-116. Kinsbourne, M. (1970). The cerebral basis of lateral asymmetries in attention. Acta Psychologia 33 163-201. Kinsbourne, M. (1972). Eye and head-turning indicates cerebral lateralization. Science 1 76, 539-541. Kinsbourne, M. (1973). The control of attention by interaction between the cerebral hemispheres. In: S. Kornblum (Ed.), Attention and performance, IV New York: Academic. Kinsbourne, M., & Bemporad, B. (1984). Lateralization of emotion: A model and the evidence. In: N. A. Fox & R. J. Davidson (Eds.), The psychobiology of affective development Hillsdale, NJ: Erlbaum. Kirk, R. E. (1968). Experimental design: Procedures for the behavioral sciences Belmont, CA: Wadsworth Publishing. Klorman, R., & Ryan, R. (1980). Heart rate, contingent negative variation and evoked potentials during anticipation of affective stimulation. Psychophysiology 17 51 3-523. Klorman, R., Weissberg, R., & Weisenfeld, A. (1977). Individual differences in fear and autonomic reactions to affective stimulation. Psychophysiology 4, 45-51 Kolb, B., & Milner, B. (1981). Observations on spontaneous facial expression after focal cerebral excisions and after intracarotid injection of sodium amytal Neuropsychologia 19 505-514. Kolb, B., & Taylor, L. (1981). Affective behavior in patients with localized cortical excisions: Role of lesion site and size. Science, 214, 89-91.

PAGE 127

119 Kronfol, Z., Hamsher, K. De S., Digre, K., & Wazir, R. (1978). Depression and hemisphere functions: Changes associated with unilateral ECT. British Journal of Psychiatry, 1 32, 560-567. Kuzendorf, R. G. (1982). Individual differences in imagery and autonomic control. Journal of Mental Imagery _6, 47-60. Lacey, J. I. (1967). Somatic response patterning and stress: Some revisions of activation theory. In: M. H. Appley & R. Trumball (Eds.), Psychological stress: Issues in research New York: Appleton-Century-Crof ts Lacey, J. I. (1972). Some cardiovascular correlates of sensorimotor behavior: Example of visceral afferent feedback? In: C. H. Hockman (Ed.), Limbic system mechanisms and autonomic function Springfield, IL: Charles C. Thomas. Lacroix, D. M., & Comper P. (1979). Lateralization in the electrodermal system as a function of cognitive/hemispheric manipulations. Psychophysiology 1 6 116-124. Lang, P. J. (1977). Imagery in therapy: An information processing analysis of fear. Behavior Therapy 8^, 862-886. Lang, P. J. (1979). A bio-informational theory of emotional imagery. Psychophysiology 1 6 495-512. Lansing, R. W., Schwartz, E., & Lindsey, D. B. (1959). Reaction time and EEG activation under alerted and nonalerted conditions. Journal of Experiment Psychology 58 1-7. Levy, J. (1969). Possible basis for the evolution of lateral specialization of the human brain. Nature, 224, 614-615. Ley, R. G., & Bryden, M. P. (1979). Hemispheric differences in processing emotions and faces. Brain and Language 7, 127-138. Ley, R. G., & Bryden, M. P. (1982). Dissociation of right and left hemispheric effects for recognizing emotional tone and verbal content. Brain and Cognition 1, 3-9. Lindsley, D. (1970). The role of nonspecific reticulo-thalamocortical systems in emotion. In: P. Black (Ed.), Physiological correlates of emotion New York: Academic Press. Lishman, W. A. (1968). Brain damage in relation to psychiatric disability after head injury. British Journal of Psychiatry 114, 373-410.

PAGE 128

120 Mach, E. (1959). The analysis of sensations, and the relation of the physical to the psychological (S. Waterlow, Ed., and C. M. Williams, Trans.). New York: Dover. (Original work published in 191^) Mahoney, A. M., & Sainsbury, R. S. (1987)Hemispheric asymmetry in the perception of emotional sounds. Brain and Cognition _6, 216-233. McGione, J. (1980). Sex differences in human brain asymmetry: A critical survey. Behavioral and Brain Sciences _3, 215-263. McKeever, W. F., & Dixon, M. S. (1981). Right hemisphere superiority of discriminating memorized from non-memorized faces: Affective imagery, sex and perceived emotional effects. Brain and Language 1_2, 2'<6-260. Melamed, B. G. (1969). The habituation and psychophysiological responses to tone, and to filmed fear stimuli under varying conditions of instructional set Unpublished doctoral dissertation. University of Wisconsin. Montgomery, W. A., & Jones, G. E. (1984), Laterality, emotionality and heartbeat perception. Psychophysiology 21 459-465. Morrow, L., Urtunski P. B., Kim, Y., & Boiler, F. (1981). Arousal responses to emotional stimuli and laterality of lesion. Neuropsychologia 1 9 65-71 Moruzzi, G., & Magoun, H. W. (1949). Brainstem reticular formation and activation of the EEG. EEG and Clinical Neurophysiology 455-475. Moskovitch, M. (1983). Local versus global solutions to problems of hemispheric specialization. Behavioral and Brain Sciences _6, 520-521 Moskovitch, M. (1986). Afferent and efferent models of visual perceptual asymmetries: Theoretical and empirical implications. Neuropsychologia 24 91-114. Moskovitch, M., & Olds, J. (1982). Asymmetries in spontaneous facial expression and their possible relation to hemispheric specialization. Neuropsychologia 20, 71-81. Myslobodsky, M. S., & Horesch, N. (1978). Bilateral electrodermal activity in depressive patients. Biological Psychology 6, 111-120. ~ Myslobodsky, M. S., & Rattok, J. (1977). Bilateral electrodermal activity in waking man. Acta Psychologia, 41, 273-282.

PAGE 129

121 Nebes, R. D. (1974). Hemispheric specialization in comraissurotomized man. Psychological Bulletin 81 1-14. Nottebohm, F. (1977). Asymmetries in neural control of vocalization in the canary. In: S. Harnad, R. Doty, L. Goldstein, J. Jaynes, & G. Krauthamer (Eds.), Lateralization in the nervous system New York: Academic. Obrist, P. A., Howard, J. L., Lawler J. E., Galosy, R. A., Meyers, K. A., & Gaebelein, C. J. (1974). The cardiac-somatic interaction. In: P. A. Obrist, A. H. Black, J. Brener, & L. V. DiCara (Eds.), Cardiovascular psychophysiology: Current issues in response mechanisms, biofeedback and methodology Chicago: Aldine-Atherton Oke, A., Keller, R., Mefford, I., & Adams, R. (1978). Lateralization of norephinephr ine in human thalamus. Science 200 1411-1413. Oke, A., Lewis, R., & Adams, R. N. (1980). Hemispheric asymmetry of norepinephrine distribution in rat thalamus. Brain Research 188, 269-272. Papanicolaou, A. C, Moore, B., Deutsch, G., Levin, H., & Eisenberg, H. (1988). Evidence for right hemisphere involvement in recovery from aphasia. Archives of Neurology 45 1025-1029. Patterson, K., & Bradshaw, J, L. (1975). Differential hemispheric mediation of nonverbal visual stimuli. Journal of Experimental Psychology: Human Perception and Performance ]_, 246-252. Paul, H. (1909). Principien der sprachgeschichte (4th ed ) Germany: Niemeyer Pennebaker, J. W., & Hoover, C. W. (1984). Visceral perception versus visceral detection: Disentangling methods and assumptions. Biofeedback and Self-Regulation 9, 339-352. Perria, P., Rosadini, G., & Rossi, G. F. (1961). Determination of side of cerebral dominance with amobarbital. Archives of Neurology 4, 175-181. Perris, C. (1974). Average evoked responses (AER) in patients with affective disorders. Acta Psychiatrica Scandinavica (Supplement) 225, 89-98. Perris, C. (1975). EEG techniques in the measurement of the severity of depressive syndromes. Neuropsychobiology 1, 16-25.

PAGE 130

122 Rader, N., Bausano, M., & Richards, J. E. (1980). On the nature of visual-cliff avoidance response in human infants. Child Development 51 61-68. Reuter-Lorenz, P. A., & Davidson, R. (1981). Differential contributions of the two cerebral hemispheres for perception of happy and sad faces. Neuropsychologia 19 609-614. Reuter-Lorenz, P. A., Givis, R. P., & Moskovitch, M. (1983). Hemispheric specialization and the perception of emotion: Evidence from right-handers and from inverted and non-inverted left handers Neuropsychologia 21 687-692. Robinson, R. G. (1983). In M. Reivich (Ed.), Cerebrovascular disease, 13th Princeton Conference New York: Raven Press. Robinson, R. G., & Benson, D. F. (1981). Depression in aphasic patients: Frequency, severity and clinical-pathological correlations. Brain and Language _U, 282-291 Robinson, R. G., & Price, T. R. (1982). Post-stroke depressive disorders: A follow-up study of 103 patients. Stroke 1 3 63-641. Rockford, J., Swartzburg, M., Chaudberg, S. M., & Goldstein, L. (1976). Some quantitative EEG correlates of psychopathology Research Communications in Psychology, Psychiatry and Behavior 211-226. Ross, E. D. (1981). The aprosodias: Functional-anatomic organization of the affective components of language in the right hemisphere. Annals of Neurology 38 561-589. Ross, E. D., & Mesulam, M-M. (1979). Dominant language functions of the right hemisphere? Prosody and emotional gesturing. Archives of Neurology 36, 144-148. Rossi, G. F., & Rosadini, G. R. (1967). Experimental analyses of cerebral dominance in man. In: D. H. Millikan & F. L. Darley (Eds.), Brain mechanism underlying speech and language New York: Grune & Stratton. Rouse, C. H., Jones, G. E., & Jones, K. R. (1988). The effect of body composition and gender on cardiac awareness. Psychophysiology 25, 400-407. Rubin, D. A., & Rubin, R. T. (1980). Differences in asymmetry of facial expressions between rightand left-handed children. Neuropsychologia 18, 373-377.

PAGE 131

123 Sackeim, H. A., Greenberg, M. S., Weiman, A. L., Gur, R. C, Hungerbuhler J. P., & Geschwind, N. (1982). Hemispheric asymmetry in the expression of positive and negative emotions. Archives of Neurology 39, 210-218. Sackeim, H. A., Gur, R. C, & Saucy, M. C. (1978). Emotions are expressed more intensely on the left side of the face. Science 202, 434-^36. Safer, M. A. (1981). Sex and hemisphere differences in access to codes for processing emotional expressions and faces. Journal of Experimental Psychology: General 110 86-100. Safer, M. A., & Leventhal, M. (1977). Ear differences in evaluating tones of voices and verbal content. Journal of Experimental Psychology: Human Perception and Performance _3, 75-82. Sauguet, J., Benton, A., & Hecaen, H. (1971). Disturbances of the body schema in relation to language impairment and hemispheric locus of lesion. Journal of Neurology, Neurosurgery and Psychiatry 34, 496-501. Schacter, S. (1957). Pain, fear and anger in hypertensive and normotensives. Psychosomatic Medicine 1 9 17-29. Schacter, S. (1964). The interaction of cognitive and physiological determinants of emotional state. Advances in Experimental Social Psychology J_, 49-80. Schacter, S. (1970). The interaction of cognition and physiological determinants of emotional state. In: J. Berkowitz (Ed.), Advances in experimental social psychology (Vol. 1). New York: Academic Press. Schacter, S., & Singer, J. (1962). Cognitive, social and physiological determinants of emotional state. Psychological Review 69, 379-399. Schneider, S. J. (1983). Multiple measures of hemispheric dysfunction in schizophrenia and depression. Psychological Medicine 1 3 i 287-297. Schwartz, G. E., Ahem, G. L., & Brown, S. L. (1979). Lateralized facial muscle response to positive and negative emotional stimuli. Psychophysiology 16 561-573. Schwartz, G. E., Davidson, R. J., & Maier, F. (1975). Right hemisphere lateralization for emotion in human brain: Interactions with cognition. Science 190, 286-288.

PAGE 132

124 Schwartz, G. E., Weinberger, D. A., & Singer, J. A. (1981). Cardiovascular differentiation of happiness, sadness, anger and fear following imagery and exercise. Psychosomatic Medicine 43 343-364. Sequndo, J. P., Naguet R., & Buser P. (1955). Effects of cortical stimulation on electrocortical activity in monkeys. Journal of Neurophysiology 18 236-245. Sergent, J. (1983). Role of the input in visual hemispheric asymmetries. Psychological Bulletin 93 481-512. Sergent, J., & 3indra, D. (1981). Differential hemispheric processing of faces: Methodological considerations and reinterpretation Psychological Bulletin 89, 541-554. Silberman, E. K., & Weingartner, H. (1986). Hemispheric lateralization of functions related to emotion. Brain and Cognition 5, 322-353Silberman, E. K., Weingartner, H., & Post, R. M. (1983a). Thinking disorder in depression: Logic and strategy in an abstract reasoning task. Archives of General Psychiatry 40 775-730. Silberman, E. K., Weingartner, H,, Stillman, R., Chen, H. J., & Post, R. M. (1983b). Altered lateralization of cognitive processes in depressed women. American Journal of Psychiatry, 1 40 1340-1344. Smokier, I. A., & Shevrin, I. (1979). Cerebral lateralization and personality style. Archives of General Psychiatry 36 949-954. Sokolov, E. N. (1963). Perception and the conditioned reflex Oxford: Pergamon Press. Sroufe, L. A., & Wunsch, J. P. (1972). The development of laughter in the first year of life. Child Development 43, 1325-1344. Steinberg, C. R., & Campos, J, J, (1983). The facial expression of anger in seven-month-old infants. Child Development 54 178-184. Strauss, E. (1983). Perception of emotional words. Neuropsychologia 21_, 99-103. Strauss, E., & Moscovitch, M. (1981). Perception of facial expressions. Brain and Language 1 3 308-332. Suberi M., & McKeever W. F. (1977). Differential right hemisphere memory storage of emotional and non-emotional faces. Neuropsychologia, 15, 757-765.

PAGE 133

125 Taylor, D, C. (1972). Mental status and temporal lobe epilepsy: A correlative account of 100 patients treated surgically. Epilepsia j_3, 727-765. Terzian, H. (1964). Behavioral and EEG effects of intracarotid sodium amytal injections. Acta Neurochirurgic 1 2 230-240. Toone, B. K., Cooke, E., & Lader M. H. (1981). Electrodermal activity in the affective disorders and schizophrenia. Psychological Medicine 1 1 497-508. Tranel, D., Damasio, A., & Damasio, H. (1988). Intact recognition of facial expression, gender and age in patients with impaired recognition of face identity. Neurology 38 690-696. Tucker, D. M. (1981). Lateral brain function, emotion and conceptualization. Psychological Bulletin 89 19-46. Tucker, D. M., Roth, R. S., Arneson, B. A., & Buckingham, V. (1977a). Right hemisphere activation during stress. Neuropsychologia ^5, 697-700. Tucker, D. M., Stenslie, C. E., Roth, R. S., & Shearer, S. L. (1981). Right frontal lobe activation and right hemisphere performance decrement during a depressed mood. Archives of General Psychiatry 38, 169-174. Tucker, D. M., Watson, R. T., & Heilraan, K. H. (1977b). Affective discrimination and evocation in patients with right parietal disease. Neurology, _l_7. 947-950. Verfaellie, M., Bowers, D., HeiLman, K. M. (1987). Hemispheric asymmetries in mediating intention, but not selective attention. Neuropsychologia 26 521-531. Verfaellie, M., & Heilman, K. M. (1987). Response preparation and response inhibition after lesions of the medial frontal lobe. Archives of Neurology 44 1265-1271. Vrana, S. R., Cuthbert B., & Lang, P. (1986). Fear imagery and text processing. Psychophysiology 23 247-253Wagner, H. N., Burns, D. H., Dannals, R. F., Wondg, D. F., Langstrom, B., Duelfer, T., Frost, J, J., Ravert, H. T., Links, J. M., Rosenbloom, S. B., Lucas, S. E., Kramer, A. V., & Kuhlar M. J. (1983). Imaging dopamine receptors in the human brain by positron tomography. Science 221 1264-1266. Walker, B., & Sandman, C. (1982). Visual evoked potentials change as heart rate and carotid pressure change. Psychophysiology 19 520-527.

PAGE 134

126 Watson, R. T., Heilman, K, M., Cauthen, J. C, & King, F. A. (1973). Neglect after cingulectoray Neurology 23 1003-1007. Webster, W. G. (1977). Territoriality and the evolution of brain asymmetry. Annals of the New York Academy of Science 299 213-221 Weerts, T. C, & Roberts, R. (1976), The physiological effects of imagining anger-provoking and fear-provoking scenes. Psychophysiology 1 3 174. Weingartner, H., & Silberman, E. K. (1982). Models of cognitive impairment: Cognitive changes in depression. Psycho pharmacology Bulletin Jl8, 27-^2. Weintraub, S., Mesulam, M., & Kramer, L. (1981). Disturbance in prosody: A right hemisphere contribution to language. Archives of Neurology 38, 7^2-744. Weschler, A. F. (1973). The effect of organic brain disease on recall of emotionally charged versus neutral narrative texts. Neurology 23, 130-135. White, K., Sheehan, P. W., & Ashton, R. (1977). Imagery assessment: A survey of self-report measures. Journal of Mental Imagery j_, 145-170, Whitehead, W. E., Drescher, V. M., Heiman, P,, & Blackwell, B. (1977). Relation of heart rate control to heart beat perception. Biofeedback and Self -Regulation 2, 371-392. Whitlock, F. A. (1982). Symptomatic affective disorders New York: Academic Press. Winer, B. J. (1971). Statistical principles in experimental design (2nd ed.). New York: McGraw-Hill, Woods, D. J. (1977). Conjugate lateral eye movement, repressionsensitization, and emotional style: Sex interactions. Journal of Clinical Psychology 33, 839-841. Yokoyama, K., Jennings, R., Ackles, P., Hood, P., & Boiler, F. (1987). Lack of heart rate changes during an attention demanding task after right hemisphere lesions. Neurology, 37, 748-755.

PAGE 135

BIOGRAPHICAL SKETCH Cynthia Rodrigues Cimino was born May 11, 1958, in Marshfield, Massachusetts. She earned her Bachelor of Science and Master of Science degrees from the University of Florida. She will obtain her Ph.D. in clinical and health psychology from the University of Florida in December, 1988. 1 27

PAGE 136

I certify that I have read this study and that in my opinion it conforms to acceptable standards of scholarly presentation and is fully adequate, in scope and quality, as a dissertation for the degree of Doctor of Philosophy. r/£)ruM-tJ^wn Bowers Associate Professor of Clinical and Health Psychology I certify that I have read this study and that in my opinion it conforms to acceptable standards of scholarly presentation and is fully adequate, in scope and quality, as a dissertation for the degree of Doctor of Philosophy. Hugh C. Q6vis, Jr. Professop' of Clinical and Health Psychology I certify that I have read this study and that in my opinion it conforms to acceptable standards of scholarly presentation and is fully adequate, in scope and quality, as a dissertation for the degree of Doctor of Philosophy. Eileen B. Fennell Professor of Clinical and Health Psychology I certify that I have read this study and that in my opinion it conforms to acceptable standards of scholarly presentation and is fully adequate, in scope and quality, as a dissertation for the degree of Doctor of Philosophy. Associate Professor of Clinical and Health Psychology

PAGE 137

I certify that I have read this study and that in my opinion it conforms to acceptable standards of scholarly presentation and is fully adequate, in scope and quality, as a dissertation for the degree of Doctor of Philosophy. /] Eidviard Valenstein Professor of Neurology I certify that I have read this study and that in my opinion it conforms to acceptable standards of scholarly presentation and is fully adequate, in scope and quality, as a cMssertation for the degree of Doctor of Philosophy. Kenneth M. Heilraan Professor of Neurology This dissertation was submitted to the Graduate Faculty of the Department of Clinical and Health Psychology in the College of Health Related Professions and to the Graduate School and was accepted as partial fulfillment of the requirements for the degree of Doctor of Philosophy December, 1988 Dean, College of Health Related Professions Dean, Graduate School


41
that increased dopaminergic activity is associated with restriction of
behavioral output by the production of motor stereotypies in humans
and other animals (Ellinwood, 1967; Iversen, 1977). Tucker also cites
evidence from psychiatric literature to suggest that negative emotions
of anxiety as well as ritualized, stereotyped behaviors associated
with obsessive-compulsive disorder and, to some extent, left partial
complex seizure disorder represent subcortical release and subsequent
overactivity of this left hemisphere activation system.
In contrast, Tucker proposes a right hemispheric specialization
for phasic arousal responses to perceptual input which exerts its
control through habituation. The presumed neurochemical substrate for
this specialization is the norephinephrine system which some
investigations have suggested is represented to a greater extent in
the right hemisphere (Oke, Keller, Mefford, 4 Adams, 1978; Oke, Lewis,
& Adams, 1980). Lateralized norephinephrine pathways are known to
show a pattern of widespread distribution throughout the brain
providing the necessary substrate for arousal responses and
facilitation of orienting to novelty. In support of this, Tucker
notes greater right hemisphere ability in tasks of "global" versus
"local" processing, requiring integration of perceptual input (Levy,
1969; Nebes, 1974). In addition, he cites evidence from the
psychiatric literature suggesting greater right hemisphere involvement
in hysteric personalities euphoric and often indifferent emotional
responses which appear analogous to the responses of right hemisphere
damaged patients (Galin, Diamond, & Braff, 1977; Gur & Gur, 1975;
Smokier & Shevrin, 1979).


30
A second possibility is that negative emotional faces may be more
configurationally complex (requiring greater right hemispheric
processing), a feature which may be of even greater significance in
the task requirements of the Reuter-Lorenz task. This possibility
does not, however, account for the large number of studies which have
actually analyzed for type of emotion and still failed to find any
laterality effect due to emotional valence (Bowers et al., 1985;
Bryden et al., 1982; Buchtel et al., 1978; Heilman et al., 1984; Ley &
Bryden, 1979).
A third possibility accounting for these discrepant findings is
that studies which do find emotion specific hemispheric effects (i.e.,
left hemisphere-positive and right hemisphere-negative) are those
which deal primarily with mood and/or experiential phenomena. In
contrast, studies which do not find emotion specific hemispheric
effects but instead do find right hemisphere superiority are those
which involve cognitive encoding of emotional stimuli (i.e., "cold,"
cognitive tasks).
In their chapter, Bryden and Ley (1983) conclude that less
evidence exists to strongly support the notion that the left
hemisphere is more involved in the processing of positive affect and
the right hemisphere is more involved in the processing of negative
affect. The available evidence, however, provides strong support for
the role of the right hemisphere in processing both positive and
negative affective material.


RESULTS
58
Experiment I: Reaction Time Task 58
Reaction Time Responses 58
Percent Correct Responses 67
Experiment II 72
Heart Rate Data Reduction 72
Skin Conductance Data Reduction 8M
DISCUSSION 88
Critical Issues 95
Conclusions 105
REFERENCES 109
BIOGRAPHICAL SKETCH 127
v


27
Findings from studies of unilateral carotid injection of sodium
amytal (WADA procedure) also support reported differences in emotional
changes following LHD and RHD. Terzian (1964) and Rossi and Rosadini
(1967) reported depressive-catastrophic reactions following left sided
injections and inappropriate euphoria following right sided
injections. These findings are also supported by other reports
(Alema, Rosadini, & Rossi, 1961; Perria, Rosadini, & Rossi, 1961).
However, Milner (cited in Rossi & Rosadini, 1967) failed to replicate
these findings. In her investigation, only 5% of patients displayed
depressive type responses, while the majority displayed euphoric
reactions. This discrepancy between Milner's study and those of other
investigators may be related to the significantly higher doses of
sodium amytal used in the Milner study (Silberman & Weingartner,
1986).
Investigation of normal, neurologically intact subjects has also
provided some support for the relative superiority of the left
hemisphere in the processing of positive affect and the relative
superiority of the right hemisphere in the processing of negative
affect. Davidson and colleagues (1979) recorded EEG responses while
subjects viewed television programs of varying emotional content and
subsequently indicated their emotional responses. Greater left
hemispheric activity was found in response to positive emotional
content, and greater right hemispheric activation was found in
response to negative emotional content. Interestingly, this
difference was only apparent on more anterior, frontal recordings
while posterior, parietal activity suggested relative right hemisphere


7
described as happy while music written in a minor is more often
described as sad (Davies, 1978). In a DL procedure, subjects were
required to identify the emotional tone of a short seven-note passage
while monitoring either the left or right ear. Subjects were more
accurate when identifying the emotional tone of passages presented to
the left ear relative to those presented to the right ear, again
supporting the notion of the right hemisphere dominance in processing
of emotional stimuli.
Dichotic listening procedures in normal subjects have also
demonstrated left ear advantages in recognition of other nonverbal,
emotional aspects of human speech such as laughing and crying (Garmon
& Nachson, 1973; King & Kimura, 1972). In a study which used a
variant of the monoaural paradigm in which subjects heard spoken
captions and laughter in either the left or right ear, cartoons were
judged as funnier when the laughter was heard by the left ear relative
to the right ear (DeWitt, 1978).
Recently, Mahoney and Sainsbury (1987) investigated hemispheric
asymmetries in perception of human, nonspeech emotional sounds.
During conditions of divided attention, a left ear advantage emerged
during the second block of trials. Under conditions of selective
attention, however, this left ear advantage was seen on the first
block of trials. In addition to providing support for a right
hemisphere superiority in processing of emotional nonspeech sounds,
these findings also suggest that effects of attention influenced the
rate and development of observed laterality effects but not the
direction of these effects.


ReactionTime
63
580 -
560 *
540 -
520 -
500 -
560
550
540
530
520
Figure 4.
Left Visual Field Trials
557.91
Happy Angry Fearful Disgusting Neutral
Emotion
(a)
Right Hand
Left Hand
Right Visual Field Trials
Experiment I--Reaction Time Analysis, Hand x Visual Field x
Stimulus Type Interaction: (a) left visual field trials
and (b) right visual field trials.


115
Gasparrini, W. G., Satz, P., Heilman, K. M., & Coolidge, F. L.
(1978). Hemispheric asymmetries of affective processing as
determined by the Minnesota Multiphasic Personality Inventory.
Journal of Neurology, Neurosurgery and Psychiatry, 41, 470-473.
Gertsmann, J. (1940). Syndrome of finger agnosia, disorientation for
right and left, agraphia and acalculia. Archives of Neurology
and Psychiatry, 44, 398-408.
Geschwind, N. (1969). Problems in the anatomical understanding of the
aphasia. In A. Benton (Ed.), Contributions to clinical
neuropsychology. Chicago: Aldine.
Geschwind, N. (1975). The apraxias: Neural mechanisms of disorders
of learned movement. American Scientist, 63, 188-195.
Glick, S. D., Jerussi, T. P., & Zimmerberg, B. (1977). Behavioral and
neuropharmacological correlates of nigrostriatal asymmetry in
rat. In: S. Harnad, R. Doty, L. Goldstein, J. Jaynes, & G.
Krauthamer (Eds.), Lateralization in the nervous system.
New York: Academic.
Glick, S. D., Meibnach, R. C., Cox, R. D., & Maayani, S. (1979).
Multiple and interrelated functional asymmetries in rat brain.
Life Sciences, 25, 395-400.
Goldstein, K. (1948). Language and language disturbances.
New York: Grue 8c Stratton.
Goldstein, K. (1952). The effect of brain damage on the
personality. Psychiatry, 1 5, 245-260.
Goldstein, S. G., Filskov, S., Weaver, L. A., 4 Ives, J. 0. (1977).
Neuropsychological effects of electroconvulsive therapy. Journal
of Clinical Psychology, 33, 798-806.
Graham, F. K. (1980). Representing cardiac activity in relation to
time. In: I. Martin and P. Venables (Eds.), Techniques in
psychophysiology. New York: Wiley.
Graves, R., Landis, T., & Goodglass, H. (1980). Laterality and sex
differences for visual recognition of emotional and non-emotional
words. Paper presented at the meeting of the Academcy of
Aphasia, Cape Cod, MA.
Graves, R., Landis, T., 8c Goodglass, H. (1981). Laterality and sex
differences for visual recognition of emotional and non-emotional
words. Neuropsychologia, 19, 95-102.


106
However, this effect was present regardless of the emotional/
nonemotional content of the stimulus. This finding is not consistent
with the view that the right hemisphere is specifically dominant for
mediating arousal responses only to emotional stimuli. Nor is it
consistent with valence or preparatory models of emotional
processing. Rather, this finding suggests that the right hemisphere
of males is dominant for mediating arousal responses, regardless of
the emotional/nonemotional content of the stimulus and this right
hemisphere dominance is not specific to emotional stimuli.
However, SCRs did suggest some findings which are consistent with
emotion specific hemispheric arousal effects. This was revealed in a
trend for a Sex x VF interaction (£ = .087) effect, in which male
subjects had greater SCRs to happy stimuli when they were presented to
the RVF (left hemisphere); greater SCRs occurred when angry stimuli
were presented to LVF (right hemisphere) than to RVF (left
hemisphere). These findings are also consistent with the preparatory
model of emotional processing.
The question is raised why emotion specific hemispheric arousal
effects are found for SCR but not for HR responses. Prior
investigations have revealed that SCR are more related to intensity of
emotional stimuli whereas HR responses were more related to valence of
emotional stimuli (Greenwald, Lang, & Cooke, 1988). It may be that
intensity differences existed across the emotional categories and this
might account for the discrepancy in findings. However, this is not
likely the case as intensity ratings across the different emotional
categories were similar. No plausible explanation is available for


xml version 1.0 encoding UTF-8
REPORT xmlns http:www.fcla.edudlsmddaitss xmlns:xsi http:www.w3.org2001XMLSchema-instance xsi:schemaLocation http:www.fcla.edudlsmddaitssdaitssReport.xsd
INGEST IEID EHWO89LVB_MNT20Q INGEST_TIME 2014-12-03T19:02:04Z PACKAGE AA00026458_00001
AGREEMENT_INFO ACCOUNT UF PROJECT UFDC
FILES


10
by RHD patients on facial affect tasks are dissociable from defects in
visuoperceptual processing. They cite two criticisms of the DeKosky
study which directly address this issue. First, they noted that a
small subset of RHD patients in the DeKosky study performed normally
on the neutral visuoperceptual task, yet, were impaired on the facial
affect tasks. This suggested that visuoperceptual deficits alone
cannot account for impaired processing of affective faces in all RHD
patients.
Secondly, they suggested that the use of same actors on affective
faces trials may have allowed subjects to rely on a pure template
matching strategy in which judgements about emotionality could have
been made on the basis of whether the two faces had the same
physiognomic configuration. A defect in this type of perceptual
process could then potentially affect performance on both facial
identity as well as affective facial tasks. As an alternative, Bowers
et al. required subjects to make affective facial judgements across
different actors so that such judgements would take place in an
"associative" context with less reliance on potential defective
perceptual matching.
Results of this study revealed that when patient groups were
statistically equated on visuoperceptual ability (facial identity
task), RHD patients still performed worse than LHD patients and
control subjects on (a) emotional discrimination of different actors,
(b) naming the emotion of a single face, and (c) picking the named
emotion from four pictures of the same actor. These findings provide
strong evidence that differences in LHD and RHD patients' abilities to


94
In contrast to the hemispheric asymmetries in arousal responses
observed in males, females did not show significant differences in HR
arousal responses to stimuli across left and right visual field.
However, females were differentially impacted by the emotional valence
of the stimuli. This was revealed as a Sex x Stimulus interaction in
analyses of HR deceleratory and HR acceleratory responses. Males did
not show this differential pattern of responding to the emotional
valence of the stimuli. Specifically, females had significantly
greater heart rate deceleratory responses for happy, disgusting, and
neutral trials compared to angry and fearful trials. Similarly,
females Ss had significantly greater HR acceleratory responses to
angry and fearful stimuli relative to happy and neutral stimuli.
Taken together, findings from both HR deceleratory and
acceleratory responses, that were obtained in Experiment II, suggest
that greater HR arousal responses occurred when stimuli were presented
to the right hemisphere (LVF) than when presented to the left
hemisphere (RVF). This laterality effect occurred only for male Ss
and was not dependent on the emotional/nonemotional content of the
stimuli. That is, emotional and neutral stimuli directed to the right
hemisphere resulted in comparable HR arousal effects. This finding is
not consistent with the view that the right hemisphere is specifically
dominant for mediating arousal responses only to emotional stimuli.
Nor is it consistent with valence or preparatory models of emotional
processing. Rather, this finding suggests that the right hemisphere
of males is dominant for mediating arousal/activation, regardless of


5
compared to control subjects. In contrast, RHD patients performed
significantly worse on the emotional prosody task relative to LHD
patients, suggesting a greater role for the right hemisphere in the
comprehension of emotional prosody.
In addition to these reports of deficits in the comprehension and
discrimination of affective prosody, Tucker et al. ( 1977b) also found
that RHD patients had difficulty in producing affectively intoned
speech. Patients were asked to say a semantically neutral sentence
using either a happy, sad, angry, or indifferent tone. RHD patients
performed significantly worse than LHD patients suggesting that their
deficits include not only the comprehension and discrimination of
affectively intoned speech but also the expression of affectively
intoned speech.
This finding was later supported by Ross and Mesulam (1979) who
reported two patients who could not express affectively intoned speech
but could comprehend affective speech. In addition, Ross (1981) has
also reported patients who could not comprehend affective intonations
but could repeat affectively intoned speech. Ross has suggested that
the right hemisphere may mediate the comprehension, repetition, and
production of affective speech much in the same way as the left
hemisphere does for propositional speech with anterior lesions
producing primarily production defects and posterior lesions producing
primarily comprehension defects.
With the advent of experimental procedures such as tachistoscopic
presentation and dichotic listening in which stimulus processing is
initially restricted to the left or right hemisphere, investigations


18
right in response to emotional instructions. These findings suggest a
greater role of the right hemisphere in generating emotional imagery.
In addition, Tucker, Roth, Arneson, and Buckingham (1977a) have
reported more left LEM in anxious than nonanxious subjects. Woods
(1977) has also suggested that habitual left eye movers are higher in
intensity and frequency of emotional reactions than right eye movers.
While findings from LEM studies may appear to be conceptually
apparent, interpretations of findings from such investigations must be
cautioned in terms of the questionable reliability and validity of LEM
as indicators of hemispheric activation. Berg and Harris (1980) were
unable to replicate previous findings in LEM studies and concluded
that the validity of the LEM procedure as a measure of hemispheric
activation has yet to be established. Ehrlichman and Weinberger
(1978), in a detailed review of the LEM literature, similarly
concluded that the use of LEM in investigations of hemispheric
functioning was questionable at best.
Recently, lacrimal flow has also been utilized as an index of
hemispheric involvement in production of mood states. Delp and
Sackeim (1987) looked at lacrimal flow following sadness and happiness
mood manipulation in male and female subjects. For female subjects,
the sadness manipulation resulted in greater relative left eye
lacrimal flow, whereas the happiness manipulation resulted in a shift
toward greater relative reduction in left eye flow. Although these
findings may be interpreted as support for greater right hemisphere
involvement in lacrimal flow, this interpretation must be observed
with some caution as the assumed lateralization of specific


28
activation during all periods of felt emotion. These results again
suggest the significance of the anterior-posterior dimension in
processing of emotional information.
Asymmetries of emotional facial expressions have also tended to
support the relative superiority of the left hemisphere in processing
of positive affect and the relative superiority of the right
hemisphere in processing of negative affect. Sackeim et al. (1978)
have reported a tendency for facial expressions to be greater on the
left side of the face. Furthermore, these authors suggest that this
asymmetry was more pronounced for negative than positive facial
expressions. Similarly, Schwartz, Ahern, and Brown (1979) have
investigated facial expressions during spontaneous mood
fluctuations. Right sided contractions were stronger during periods
of nappiness or excitement, while left sided contractions were
stronger during facial expressions of sadness and fear.
Ahern and Schwartz (1979) have reported more right LEM (left
hemisphere activation) when subjects respond to questions that evoked
happiness or excitement. In contrast, more left LEM (right hemisphere
activation) occurred when subjects responded to questions that evoked
sad or fearful affects. As previously discussed, the questionable
validity of LEM as an index of cerebral activation, however, must be
considered in any interpretation of this study.
Studies have also investigated left and right visual field
differences for positive and negative emotions. Using a contact lens
system that restricts visual input to the RVF (left hemisphere) or LVF
(right hemisphere), Dimond, Farrington, and Johnson (1976) reported


84
condition). In the present experiment, Ss were required to also image
a personal episode of the same valence as the stimulus. The time
course of the Ss' imaging could not be controlled and it may be the
case that Ss differed in onset of their imaging and occurrence of
affective aspects of the image.
One observations which is of some interest is the fact that
female Ss do show some indications of "autonomic patterning" in their
HR changes across the 8 post-stimulus seconds. That is, females
appeared to show less deceleration for fearful trials relative to
other affective categories. Furthermore, this pattern appears to be
reflected to a greater extent for LVF (right hemisphere) than RVF
(left hemisphere) presentations.
Skin Conductance Data Reduction
For each trial, SCRs were depicted as the difference between the
average of the skin conductance level during the 2 seconds preceding
stimulus onset (tonic, baseline level) and the maximum skin
conductance level during the 8 post-stimulus seconds (phasic level).
This SCR value for each trial served as the dependent variable in a
repeated measures ANOVA. Sex was the between subjects factor and
Visual Field (left, right), Hand (left, right), and Stimulus Type
(happy, angry, fearful, disgusting, neutral) were the within subjects
factors.
Results of this ANOVA yielded no significant main effects or
interaction effects. Table 7 depicts a summary of this analysis. A
weak trend, however, was observed for a Sex x Visual Field x Stimulus


51
should occur for neutral, disgust, and happy stimuli (assumed to be
nonpreparatory).
Alternatively, it is possible that the right hemisphere may be
dominant for mediating arousal/activation responses to stimuli,
regardless of their emotional-nonemotional content. In this view,
stimuli directed to the right hemisphere should result in greater
arousal/activation responses than stimuli directed to the left
hemisphere. However, any differences in arousal/activation responses
to emotional versus neutral stimuli should be comparable across the
left and right hemispheres. Thus, in Experiment I, WS directed to the
right hemisphere (LVF) should result in faster RTs than WS directed to
the left hemisphere (RVF). Any RT differences between emotional
versus nonemotional WS should be comparable for LVF and RVF
presentations. Likewise, in Experiment II, one would predict greater
autonomic responsivity (HR, SCR) to stimuli directed to the right
versus left hemisphere. Again, however, any differences in arousal
responses to emotional versus neutral stimuli should be comparable
across LVF and RVF presentations of the stimuli.


demands on the organism for survival. Alternatively, it is possible
that the RH may be dominant for mediating arousal/activation responses
to stimuli, regardless of their emotional/nonemotional content.
The focus of the present study was to further examine these
different conceptualizations of the hemispheric processing of
emotional stimuli in neurologically intact male and female subjects.
This was accomplished in two separate experiments: (a) a choice
reaction time task was used to investigate subjects' activation
responses to a centrally presented, neutral stimuli when it was
preceded by neutral or emotional warning stimuli and (b) heart rate
(HR) and skin conductance (SC) measures were used to investigate
subjects' responses to laterally presented neutral and emotional
stimuli. Findings from Experiment I, the reaction time experiment,
failed to support any of the laterality models of emotion. Findings
from Experiment II, using HR and SC measures, were more congruent with
models of hemispheric differences in processing of emotional
stimuli. Males showed lateralized effects of HR, arousal responses
which support greater RH involvement in production of arousal
responses. However, this effect was present regardless of the
emotional/nonemotional content of the stimulus. Skin conductance
responses in male subjects did provide some support for hemispheric
specific emotional valence effects but this did not reach statistical
significance. In contrast, to the hemispheric asymmetries in arousal
responses observed in male subjects, female subjects did not show
significant differences in arousal responses across left and right
visual fields. However, females were differentially impacted by the
vi 1


89
and result in significantly quicker/more accurate detection than
negative emotional stimuli presented initially to the left hemisphere.
A third model, the preparatory model, argues that this positive
negative dichotomy in hemispheric processing of emotions is
artifactual and actually relates to differences in arousal and
activation/preparation for action. According to this model, the right
hemisphere is dominant for mediating arousal/activation and as such,
is more intrinsically involved in processing emotional stimuli that
have greater "preparatory" significance for survival (i.e., fight-
flight emotions). In contrast, the left hemisphere is more involved
in mediating nonpreparatory emotions that place less immediate
attentional demands on the organism for survival.
Lastly, it is possible that the right hemisphere is dominant for
mediating arousal/activation responses to stimuli irrespective of
their emotional/nonemotional content. That is, both emotional and
nonemotional stimuli should result in significantly greater arousal/
activation responses when projected to the right hemisphere than the
left hemisphere.
The present study sought to further investigate these different
conceptualizations of the hemispheric processing of emotional
stimuli. The purposes of the study were to (a) determine the extent
to which laterally presented emotional/nonemotional stimuli might
result in differential patterns of behavioral activation (reaction
time responses) as well as differential patterns of autonomic arousal
(HR, SCR responses); (b) to determine whether there were hemispheric
asymmetries in mediating arousal and/or activation responses to these


Maximum Heart Rale Acceleration Maximum Heart Rale Acceleration
80
Happy Angry Fearful Disgusting Neutral
Emotion
Males
Females
Figure 15. Experiment II--Maximum Acceleratory Heart Rate Analysis,
Stimulus Type x Sex Interaction.
Figure 16. Experiment IIMaximum Acceleratory Heart Rate Analysis,
Visual Field x Sex Interaction.


96
recovery from aphasia following hemispheric lesions, it appears that
males are more left hemisphere lateraiized for language, whereas
females have more bilateral representation of speech and language
functions (McGlone, 1980).
With respect to arousal asymmetries, studies have not
systematically addressed sex differences. Prior investigations which
have looked at the right hemisphere's role in production of arousal
responses in normal, neurologically intact Ss have not investigated
the effect of sex on production of laterality effects in HR and SCR
(Hugdal et al., 1983; Hugdahl et al., 1982; Walker & Sandman, 1982).
Heart rate findings from the present investigation suggest that the
greater role of the right hemisphere in production of arousal
responses may exist for male Ss to a greater extent than for female
Ss; findings which are consonant with neuropsychological
investigations of laterality effects across sex.
In addition to studies which have looked at the generation of
autonomic arousal (i.e., HR, SCR), a related area of research has also
investigated cerebral asymmetries in heart beat perception and
detection. Davidson, Horowitz, Schwartz, and Goodman (1981) measured
RT differences between R-wave occurrence and key press latencies.
These authors report that finger taps by the left hand (right
hemisphere) had shorter mean latencies from heartbeat than did taps
from the right hand (left hemisphere), suggesting that the right
hemisphere is more sensitive or more efficient than the left
hemisphere in response tracking to cardiac afferent feedback.


INTRODUCTION
Neuropsychological approaches to the investigation of emotional
processing have evolved, in large part, from early clinical
observation of brain injured patients and subsequent systematic
investigation of their performance on a variety of emotional tasks.
One of the earliest reports was provided by Babinski (191^) in which
he noted that patients with right hemisphere damage appeared
indifferent or euphoric. Denny-Brown, Meyer, and Horenstein (1952)
also reported evidence of such "indifference" reactions after right
hemisphere lesion and noted its co-occurrence with unilateral neglect
syndrome in which patients failed to orient, report, or respond to the
left side of their body. In 1952, Goldstein published his observation
that "catastrophic" emotional responses often accompanied left
hemisphere damage. These reports were later corroborated by Hecaen
(1962), who al30 noted that catastrophic reactions most often followed
left hemisphere insult whereas indifference reactions were more
frequent following right hemisphere damage.
In 1972, Gainotti reported a large scale study of 160 patients
who had sustained left or right sided lesions. Based on systematic
observation of the frequency and type of symptomatology, Gainotti
reported a consistent relationship between behaviors indicative of a
1


17
Bennett, & Golemena, 1979). Right hemisphere activation has also been
reported during hypnotically induced depression (Tucker, Stenslie,
Roth, & Shearer, 1981), during generation of emotional imagery and
during painful stimulation (Karlin, Weinapple, Rochford, 4 Goldstein,
1979).
In addition to comparisons of left versus right sided activation,
several authors also emphasize the importance of relative differences
in level of activation in anterior versus posterior regions. Tucker
(1981) reported frontal activation but not posterior activation in
depressed mood. Likewise, Davidson et al. (1979) reported that mood
valence varied with right versus left activation in frontal regions,
but that posterior regions showed right hemisphere activation
irrespective of valence.
Lateral eye movements (LEM) as indices of hemispheric activation
have also been used to assess the role of the right hemisphere in
regulation of mood and affect. Prior investigations have revealed a
tendency toward right LEM (left hemisphere activation) with verbal
processing and left LEM (right hemisphere activation) with
visuospatial processing (Kinsbourne, 1972). Schwartz, Davidson, and
Maier (1975) have reported a greater frequency of left LEM in subjects
performing emotional versus neutral mental tasks.
Similarly, Borod, Vingiano, and Cytryn (1988b) measured LEM while
subjects were asked to generate emotional images of positive and
negative valence in auditory, visual, and tactile modalities.
Overall, subjects looked significantly more to the left than to the


ACKNOWLEDGMENTS
I would like to extend my gratitude to those who provided the
encouragement and support that enabled the completion of this
project. First, I am grateful to my dissertation committee. I would
like to thank my chair, Dr. Dawn Bowers, for her time, hard work, and
patience over the years; for her skill at making me really think; and
especially for her belief in my ability. I am also grateful to
Dr. Kenneth Heilman for providing material and moral support and for
sharing his knowledge and never-ending enthusiasm, creativity, and
wonderment. I thank Dr. Rus Bauer for his advice on
psychophysiological technique and statistical method and for his
infrequent but apt clinical interpretations. I would also like to
tnank Dr. Eileen Fennell for ner discerning comments on methodology,
her ground-rooted advice on professional development, and her
kindness. I am also grateful to Dr. Ed Valenstein for his support in
my defense and for his thought-provoking question at the end of my
dissertation copy (humble as always). Finally, I am grateful to
Dr. Hugh Davis for so many things, but especially for his guidance in
development of my clinical abilities, his warmth and playfulness, and
his love of language and verbal tapestries.
I owe a debt of gratitude to Cindy Zimmerman for her skill in
organizing the typing and completion of the manuscript. My
appreciation also goes to Dr. Roger Blashfield for his unconditional


45
able to benefit from preparatory information wnile the patient with
right SMA damage was unable to benefit from preparatory information,
again suggesting a greater role of the right hemisphere in activation
of response.
More recently, Heilman (1988, personal communication) has
suggested that emotion specific hemispheric effects (i.e., left
hemisphere-positive, right hemisphere-negative) reported in the
literature may be artifactual and actually represent hemispheric
differences in arousal and preparation for action. Because the right
hemisphere is dominant for mediating arousal/activation, it may
therefore be more intrinsically involved in processing emotional
stimuli that have greater "preparatory" significance for survival
(i.e., fight-flight emotions such as anger and fever). In contrast,
the left hemisphere may be more involved in mediating nonpreparatory
emotions (i.e., happiness, sadness, disgust) that place less immediate
or "phasic" attentional demands on the organism for survival.
In summary, recent investigations in brain impaired and
neurologically intact subjects suggests a greater role of the right
hemisphere in arousal-activation responses. Furthermore, this finding
is observed across several indices of arousal-activation including
heart rate, skin conductance, and reaction time measures.
Critical Issues
As reviewed in the introduction, there appears to be general
consensus that the two hemispheres in man differ in terms of their
contribution to emotional processing. However, the precise role


67
Percent Correct Responses
A separate ANOVA was conducted which used proportion of correct
responses for each S as the dependent variable. An arcsin square root
transformation was performed to correct for lack of a normal
distribution inherent in proportion data and to meet homogeneity of
variance assumptions for the ANOVA (Kirk, 1968; Winer, 1971).
Results of this analysis utilizing ail 30 Ss are depicted in
Table 3. Findings revealed only a significant main effect of Stimulus
Type [F (4, 112) = 7.60, £ = .0001)], depicted in Figure 7. Duncan's
post-hoc comparisons revealed that percent correct identification of
disgusting (M = .804) trials was significantly worse than all four
remaining categories (happy, = .907; angry, = .904; fearful, M =
.903; neutral, = .938) at £ < .05. In addition, a trend for
Stimulus Type x Visual Field [F (4, 112) = 2.29, £ = .064)] was also
revealed, depicted in Figure 8. This pattern of findings suggest that
happy trials were more accurate in the LVF while neutral trials were
more accurate in the RVF.
A second ANOVA was conducted for percent correct data which
excluded the two female Ss noted above. Results of this analysis are
depicted in Table 4. This analysis revealed only a significant main
effect of Stimulus Type [F (4, 104) = 6.13, £ = .0002)], depicted in
Figure 9. Duncan's post-hoc comparisons revealed that percent correct
identification of disgusting (_M = .810) trials was significantly worse
than all four remaining categories (happy, M = .911; angry, M = .904;
fearful, M = .899; neutral, M = .940) at £ < .05. No other effects
reached significance or trend status.


50
(LVF). Similarly, in Experiment II, one would predict greater
autonomic responsivity (HR acceleratory and deceleratory responses,
SCR) to negative stimuli that are directed to the right hemisphere
(LVF) versus stimuli that are directed to the right hemisphere (LVF)
versus those directed to the left hemisphere (RVF). The opposite
pattern of autonomic arousal should occur for positive emotional
stimuli.
According to the hemispheric preparatory model of emotion, the
right hemisphere is dominant for mediating emotional stimuli that have
a greater preparatory significance for survival (i.e., fight-flight
emotions such as anger and fear). In contrast, the left hemisphere is
dominant for mediating nonpreparatory stimuli that* place less "phasic"
demands on the individual for immediate survival (i.e., happiness,
disgust, neutrality). If this model is correct, then in the first
experiment one would predict that anger and fear WS directed to the
right hemisphere (LVF) should result in faster RTs than when they are
directed to the left hemisphere (RVF). Conversely, happy, disgust,
and neutral WS should result in faster RTs when they are directed to
the left versus right hemisphere. Likewise, if one assumes that
behavioral activation and autonomic responsivity are strongly coupled,
then similar predictions would be made for Experiment II. That is,
one would predict greater autonomic responsivity (HR acceleratory and
deceleratory responses) to anger and fear stimuli (assumed to be
preparatory) when they are directed to the right versus left
hemisphere. The opposite hemispheric pattern of autonomic arousal


48
certain categories of emotion (positive/negative; preparatory/
nonpreparatory) induce asymmetric arousai/activation, depending on the
hemisphere to which they are initially presented.
In order to address these issues, two experiments were
completed. In the first study, laterally presented emotional stimuli
of different valences served as warning stimuli to the subjects who
then made manual RT responses to a neutral midline stimulus. This
warning stimulus paradigm was chosen because it enables one to
determine the extent to which lateralized emotional warning stimuli
serve to behaviorally activate and prepare the individual to respond
to a subsequent stimulus (Lansing, Schwartz, & Lindsey, 1959). In the
second study, laterally presented emotional stimuli were also shown to
subjects, and autonomic indices of arousal (HR, SCR) were measured.
Although it would have been more "ideal" to obtain both autonomic and
RT measures to the lateralized emotional stimuli in the same study,
this wa3 not realistically feasible. The "3low" rise time of the SCR
(2-4 seconds) in conjunction with the relatively short lived
activating effects of warning stimuli (500-2000 msec) precluded such a
direct manipulation. Thus, two separate experiments were conducted.
Four different emotional categories were chosen for the present
investigation. Two categories which have previously been shown to
result predominantly in cardiac deceleratory responses (happy,
disgust) and two categories which have previously been shown to result
predominantly in cardiac acceleratory responses (fear, anger) (Ekman
et al., 1983). Due to the relative paucity of discernable positive
emotions among the wide range of emotional categories (Ekman, 1972),


33
the organism may realign the head and/or body toward the source of
stimulation. At the neurophysiological level, several changes occur
which include a transient increase in skin conductance, pupil
dilation, heart rate deceleration, pauses of respiration, and EEG
desynchronization. The presumed functional value of these collective
components of the OR is to make the organism more receptive to
incoming stimuli as well as to prepare the organism for action.
A second component of Sokolov's model is that of the defensive
response (DR), which is likely to be of equal, if not greater,
potential significance in the processing of emotional stimuli. In
Sokolov's view, when high intensity or aversive stimuli are presented,
the orienting response is soon replaced by the defensive response.
This response is characterized by greater increases in sympathetic
activity across response systems including heart rate acceleration and
cephalic vasoconstriction. The functional value of this response at
the behavioral level is avoidance of the stimulus.
It is interesting to note that both orienting and defensive
responses are conceptualized as arousal responses, yet each results in
characteristically distinct patterns of responding. This occurrence
presents difficulty for the view of arousal as a unidimensional
phenomenon. Subsequent investigators have looked at these seemingly
paradoxical heart rate responses and attempted to correlate them with
psychological processes.
Lacey (1967) reconceptualized this "directional fractionation" of
cardiac activity in terms of the conditions under which stimulation
occurred and their effects on the organism's processing of stimuli.


99
sexual display. Furthermore, he notes that the lateralized control of
song in the left brain of canary and chaffinch is related to mate
attraction and territoriality, i.e., sex and spatial analysis
(Nottebohm, 1977; Webster, 1977).
Mach (1959) suggests that this neural asymmetry serves to
increase the efficiency of spatial analysis in an environment where no
systematic left/right bias exists. Furthermore, he argues the
evolutionary advantage of this increased efficiency is related to mate
attraction and in species where the male actively seeks the female,
this greater lateralization for male organisms is of adaptive
significance and survival value.
As previously discussed, female subjects did not show significant
differences in HR deceleratory responses or HR acceleratory responses
across left and right visual fields as found in male subjects.
However, female subjects were differentially impacted by the emotional
valence of the stimuli (revealed as a significant interaction of Sex x
Stimulus Type for analysis of both HR deceleratory responses and HR
acceleratory responses). Male subjects, in contrast, did not show
this differential heart rate responding across stimulus categories.
Analysis of deceleratory HR responses for female subjects revealed
significantly greater decelerations for happy, disgusting, and neutral
trials relative to fearful trials. Similarly, analysis of
acceleratory HR responses for female subjects revealed significantly
greater acceleratory responses for angry and fearful slides compared
with neutral slides. These results are consistent with those of other


101
through biofeedback training was associated with the frequency of
subject's images. In addition, Kuzendorf (1982) also found that
subject's ability to produce temperature differences across the hands
was significantly correlated with prevalence of imagery. Lastly,
Arabian and Furedy (1983) report that HR deceleratory conditioning
(conditioned response) in good imagers was more similar to HR
deceleratory unconditioned responses than HR deceleratory conditioning
(conditioned responses) for poor imagers.
In the present investigation, imagery ability was not assessed.
However, it is possible that the current findings which suggest that
female subjects showed differntial autonomic responding across various
emotional stimuli might be related to potential differences in imagery
ability between female and male subjects. For example, there is some
evidence in the literature to suggest that female subjects, in
general, are better imagers than male subjects (White, Sheehan, &
Ashton, 1977). This finding has led some investigators to use only
female subjects in investigations of imagery and autonomic responding
(Jones & Johnson, 1978; 1980).
Regardless of the basis for the sex differences, the differential
pattern of HR responses to the various emotional stimuli by female
subjects in the present study is similar in pattern to that reported
by other investigators (Ekman et al., 1983; Hare & Blevings, 1975;
Klorman & Ryan, 1980). Findings from the present investigation
suggest that HR decelerations were significantly greater for both
happy and neutral slides relative to angry and fearful slides.


108
greater role of the right hemisphere in production of arousal
responses exists to a significantly greater extent in male Ss versus
female Ss; findings which are congruent with findings from
neuropsychological investigations which report a greater degree of
lateral asymmetries for male Ss on a variety of tasks. Similarly, sex
differences in patterning of HR responsivity across different
emotional categories suggests that female but not male Ss reveal
differences in magnitude of HR acceleratory and HR deceleratory
responses across various emotional categories. The extent to which
this differential autonomic responding in male versus female Ss is due
to differences in imagery ability across these groups is an area of
future investigation.


13
advantage for processing of facial affect exists above and beyond the
right hemisphere's advantage for processing of facial identity.
Evidence about possible differences due to sex of subjects, however,
remains equivocal.
Related Research
Several studies have implicated the right hemisphere in memory
for emotionally charged materials. Weschler (1973) reported one of
the few studies of emotional memory in brain impaired subjects. Right
hemisphere and LHD patients were presented with two types of stories--
one emotional and the other nonemotional. When asked for subsequent
recall, RHD subjects made significantly more errors in recalling
emotional stories relative to LHD patients.
Cimino, Verfaellie, Bowers, and Heilman (1988) investigated
whether RHD patients have difficulty remembering past affective
episodes by asking them to recall prior emotional and neutral
experiences. Findings revealed that RHD patients produced
significantly less emotional reports than control subjects as judged
by independent raters. However, their own emotionality ratings were
no different from those of control subjects suggesting some
discordance between their actual production of emotional memories
versus their own perceived emotionality of such memories.
Unfortunately, most patients with LHD are aphasic and could not be
used in this study. Therefore, this report cannot conclude that this
defect is specific to RHD.
An investigation in normal subjects (Gage & Safer, 1979) looked
at hemispheric differences in mood-state dependent effects for


42
Tucker suggests that the two hemisphere's differing modes of
processing may be the primary factor in lateralized valence effects
reported in the literature. He hypothesizes that what we experience
as emotions arise from operation of these arousal and attentional
modulatory processes.
Heilman's Model
From investigations of indifference reaction associated with
right hemisphere damage and unilateral neglect syndrome, Heilman and
colleagues (1983) have suggested a model of emotional processing based
on hemispheric differences in arousal-activation responses. Heilman
has suggested that right hemisphere damaged patient's difficulties in
emotional expression may be a result of (a) deficits in arousal-
activation and (b) an ability to develop an appropriate cognitive
state due to basic deficits in comprehension of prosodic elements of
speech and affective facial expressions. Patients with indifference
reaction often have the unilateral neglect syndrome in which they may
fail to orient, report, or respond to stimuli in the contralateral
side of space (Denny-Brown et al 1952; Gainotti, 1972; Heilman &
Valenstein, 1972). Heilman et al. have suggested that unilateral
neglect is a defect in attenuation-arousal-activation due to
disruption of a corticolimbic-reticular loop (Heilman & Van Den Abel,
1979). Based on the fact that neglect occurs most often following
right hemisphere damage, he has proposed that the right hemisphere may
be dominant for mediating attention-arousal-activation responses.
To investigate arousal responses in brain impaired patients,
Heilman et al. (1978) stimulated the forearm ipsilateral to the side


19
neuroanatomical pathways regulating lacrimal flow have not been
clearly established.
Measures of facial asymmetries observed during emotional
expression have also provided support for the hypothesis of right
hemisphere dominance in regulating affect and mood states.
Musculature of the lower part of the face is contralaterally
innervated and asymmetries observed with respect to facial expression
may be used to infer relative hemispheric involvement in production of
emotional expression.
Studies with normal subjects have revealed that the left hemiface
moves more extensively during posed facial expression (Borod & Caron,
1980; Borod, Caron, i Koff, 1981; Borod, Kent, Koff, Martin, & Alpert,
1988a; Borod, Koff, & White, 1983; Moskovitcn & Olds, 1982). One
criticism of such studies, however, is that they have used posed
facial expressions which may not actually reflect the underlying
affect or mood of the subject. In response to this criticism, several
authors have regarded spontaneous facial expression as a more valid
index of the subject's affective state. Ekman, Hager, and Friesen
(1981) failed to find such asymmetries of emotional expression during
spontaneous facial expressions. In contrast, other investigators
(Borod et al., 1983; Moskovitch & Olds, 1982) have observed greater
left sided (right hemisphere) involvement for both spontaneous and
posed facial expressions.
Several studies have also used composite photographs in which the
mirror image of the left or right half of the face is combined with
the original ipsilateral image. This process results in a complete


93
of findings is consistent with both the valence model and the
preparatory model of emotional processing, these findings must be
viewed with caution since significance was not obtained. The lack of
robust findings may be due, in part, to the fact that SCRs for Ss were
generally quite small, making detection of differences across visual
field, hand, and stimulu type conditions difficult.
With regard to HR responses that were also measured in Experiment
II, happy and neutral slides induced significantly greater HR
deceleratory responses than angry or fearful slides. This finding
occurred regardless of which hemisphere initially received the
stimulus. In other words, no emotional specific hemispheric arousal
effects were present for HR deceleratory responses. Additionally, a
trend was observed for greater HR deceleratory responses for stimuli
directed to the right hemisphere (LVF) versus those directed to the
left hemisphere (RVF). This finding was based primarily on the
responses of male Ss, as indicated by a significant Sex x VF
interaction. Specifically, males had significantly larger HR
deceleratory responses to LVF stimuli than to RVF stimuli, and this
effect was present regardless of the emotional/nonemotional content of
the stimuli. In contrast, females Ss displayed no laterality effect
of any kind in their HR deceleratory responses to emotional or neutral
stimuli. A similar pattern of findings was also observed for HR
acceleratory responses, as reflected in a trend for a Sex x VF
interaction (£ = .077). Male Ss again tended to produce greater HR
acceleratory responses to LVF stimuli than to RVF stimuli.


43
of lesion in RHD and LHD patients while recording galvanic skin
response from the same side. The authors noted that RHD patients had
significantly smaller GSR arousal responses than LHD patients or
nonbrain damaged control subjects. Similarly, Morrow, Urtunski, Kim,
and Boiler (1981) presented LHD and RHD patients with neutral and
emotional stimuli. Right hemisphere patients showed decreased
galvanic skin responses to both neutral as well as emotional stimuli
relative to LHD patients.
More recently, Yokoyama, Jennings, Ackles, Hood, and Boiler
(1987) have looked at heart rate and reaction time responses in
patients with right unilateral hemispheric lesions. These authors
found that RHD patients had significantly slower reaction times and
decreased heart rate responses (both deceleratory as well as
acceleratory) relative to LHD patients. These findings indicate that
the greater role of the right hemisphere in attention may be reflected
in both reaction time as well as anticipatory heart rate changes.
Investigations in neuroiogically intact subjects corroborate
these findings. Hugdahl, Franzon, Anderson, and Walldebo (1983)
report greater anticipatory heart rate accelerations for emotional
stimuli presented to the LVF (right hemisphere) compared with RVF
(left hemisphere) trials. Similarly Walker and Sandman (1982) also
report greater right hemisphere activity (as measured by the P100
component of the average evoked potential) when the heart was
spontaneously accelerated. In addition, Hugdahl, Wahlgren, and Wass
(1982) found delayed habituation of the electrodermal orienting


15
stimuli (faces/words) on subsequent judgements of the affective value
of laterally presented emotional and nonemotional verbal target
stimuli. They found that affective primes presented to the RVF (left
hemisphere) resulted in decreased accuracy judgements of the target
stimuli that were also presented to that hemisphere. In contrast,
affective primes presented to the LVF (right hemisphere) resulted in
increased accuracy judgements regarding the affective value of verbal
stimuli presented to the right hemisphere.
Taken together, findings with both aphasic and neurologically
intact individuals suggest a right hemisphere advantage in processing
emotional verbal stimuLi. However, this view is not entirely clearcut
in that other studies have failed to replicate the "right hemisphere"
laterality effect for identifying emotional versus nonemotional verbal
stimuli (Strauss, 1983).
Another area of investigation concerning the role of the right
hemisphere in processing of emotional stimuLi is that of humor
appreciation. Brownell et al. (1984) have recently reported that RHD
patients have significant difficulty in understanding narrative humor
as portrayed in short story jokes. Similarly, Bihrle et al. (1986)
have reported that RHD patients performed significantly worse than LHD
patients on a nonverbal cartoon completion task.
In summary, a large body of literature suggests that the right
hemisphere plays a greater role in the comprehension and expression of
emotional information. These findings have been consistently
demonstrated in both neurologic as well as normal populations across a
wide range of tasks including affective facial judgements and


122
Rader, N., Bausano, M.f & Richards, J. E. (1980). On the nature of
visual-cliff avoidance response in human infants. Child
Development, 51, 61-68.
Reuter-Lorenz, P. A., & Davidson, R. (1981). Differential
contributions of the two cerebral hemispheres for perception of
happy and sad faces. Neuropsychologia, 19, 609-614.
Reuter-Lorenz, P. A., Givis, R. P., & Moskovitch, M. (1983).
Hemispheric specialization and the perception of emotion:
Evidence from right-handers and from inverted and non-inverted
left handers. Neuropsychologia, 21, 687-692.
Robinson, R. G. (1983). In M. Reivich (Ed.), Cerebrovascular disease,
13th Princeton Conference. New York: Raven Press.
Robinson, R. G., & Benson, D. F. (1981). Depression in aphasic
patients: Frequency, severity and clinical-pathological
correlations. Brain and Language, 14, 282-291.
Robinson, R. G., & Price, T. R. (1982). Post-stroke depressive
disorders: A foilow-up study of 103 patients. Stroke, 13.
63-641.
Rockford, J., Swartzburg, M., Chaudberg, S. M., & Goldstein, L.
(1976). Some quantitative EEG correlates of psychopathology.
Research Communications in Psychology, Psychiatry and Behavior,
T, 211-226.
Ross, E. D. (1981). The aprosodias: Functional-anatomic organization
of the affective components of language in the right
hemisphere. Annals of Neurology, 38, 561-589.
Ross, E. D., & Mesulam, M-M. (1979). Dominant language functions of
the right hemisphere? Prosody and emotional gesturing. Archives
of Neurology, 36, 144-148.
Rossi, G. F., & Rosadini, G. R. (1967). Experimental analyses of
cerebral dominance in man. In: D. H. Millikan & F. L. Darley
(Eds.), Brain mechanism underlying speech and language.
New York: Grue & Stratton.
Rouse, C. H., Jones, G. E., & Jones, K. R. (1988). The effect of body
composition and gender on cardiac awareness. Psychophysiology,
25, 400-407.
Rubin, D. A., & Rubin, R. T. (1980). Differences in asymmetry of
facial expressions between right- and left-handed children.
Neuropsychologia, 18, 373-377.


116
Greenwald, M., Lang, P., & Cooke, E. (1988). Response patterning
during affective visual stimulation: Evidence for a dimensional
model of emotion. Unpublished manuscript.
Gur, R. E., & Gur, R. C. (1975). Defense mechanisms, psychosomatic
symptomatology and conjugate lateral eye movements. Journal of
Consulting and Clinical Psychology, _4_3 416-420.
Haggard, M. P., & Parkinson, A. M. (1971). Stimulus and task factors
as determinants of ear advantages. Quarterly Journal of
Experimental Psychology, 23. 168-177.
Hansch, E. C., & Pirozzolo, F. J. (1980). Task relevant effects on
the assessment of cerebral specialization for facial emotion.
Brain and Language, 10, 51-59.
Hantas, M. N., Katkin, E. S., & Reed, S. D. (1984). Cerebral
lateralization and heartbeat discrimination. Psychophysiology,
_21_, 241-278.
Hare, R. D., & Blevings, G. (1975). Defensive responses to phobic
stimuli. Biological Psychology, _3, 1-13*
Hecaen, H. (1962). Clinical symptomatology in right and left
hemispheric lesions. In: V. B. Mountcastle (Ed.),
Interhemispheric relations and cerebral aominance. Baltimore:
Johns Hopkins University Press.
Heilman, K. M., Bowers, D., Speedie, L., & Coslett, B. (1984).
Comprehension of affective and nonaffective prosody. Neurology,
34, 917-921.
Heilman, K. M., Scholes, R., & Watson, R. T. (1975). Auditory
affective agnosia: Disturbed comprehension of affective
speech. Journal of Neurology, Neurosurgery and Psychiatry, 38,
69-72.
Heilman, K. M., Schwartz, H. D., & Watson, R. T. (1978). Hypoarousal
in patients with the neglect syndrome and emotional
indifference. Neurology, 28, 229-232.
Heilman, K. M., 4 Valenstein, E. (1972). Frontal lobe neglect.
Neurology, 22, 660-664.
Heilman, K. M., 4 Van Den Abel, T. (1979). Right hemispheric
dominance for mediating cerebral activation. Neuropsychologia,
17, 315-321.


114
Flor-Henry, P. (1976). Lateralized temporal limbic dysfunction and
psychopathology. Annals of the New York Academy of Sciences,
280, 777-795.
Flor-Henry, P. (1979). On certain aspects of the localization of the
cerebral systems regulating and determining emotion. Biological
Psychiatry, 14, 677-698.
Flor-Henry, P. (1983). Functional hemispheric asymmetry and
psychopathology. Integrative Psychiatry, J_, 46-52.
Flor-Henry, P., & Koles, Z. V. (1980). EEC studies in depression,
mania and normals: Evidence for partial shifts of laterality in
the affective psychoses. Advances in Biological Psychiatry, 4,
21-43.
Fox, N. A., & Davidson, R. J. (1984). Hemispheric substrates of
affect: A developmental model. In: N. J. Fox & R. J. Davidson
(Eds.), The psychobiology of affective development. Hillsdale,
NJ: Erlbaum.
Freeman, R. L., Galaburda, A. M., Cabal, R. D., & Geschwind, N.
(1935). The neurology of depression. Archives of Neurology, 42,
289-291.
Freid, I., Mateu, C., Ojemann, G., Wohns, J., b Fedio, P. (1982).
Organization of visuospatial function in human cortex. Evidence
from electrical stimulation. Brain, 105, 349-371.
French, J. E., Hernandez-Peon, R., & Livingston, R. (1955).
Projections from the cortex to cephalic brainstem (reticular
formation) in monkey. Brain 1 8, 74-95.
Freyschuss, V. (1970). Cardiovascular adjustments to somato-motor
activation: The elicitation of increments in heart rate, aortic
pressure and venomotor tone with the initiation of muscle
contraction. Acta Physiologica Scandinavica, 342, 1-63.
Gage, D. F., & Safer, M. A. (1979). Hemispheric differences in the
mood state-dependent effect for recognition of emotional faces.
Journal of Experimental Psychology: Learning, Memory and
Cognition, 11, 752-763.
Gainotti G. (1972). Emotional behavior and hemispheric side of the
lesion. Cortex, _8, 41-55.
Galin, D., Diamond, R., b Braff, D. (1977). Lateralization of
conversion symptoms: More frequent on the left. American
Journal of Psychiatry, 134, 578-580.


110
Berg, M. R., & Harris, L. J. (1980). The effect of experimenter
location and subject anxiety on cerebral activation measured by
lateral eye movements. Neuropsychologia, 18, 89-9^.
Berger, H. (1933)- Uber das electroenkephalogram des menschen.
Archiv fur Psychiatrie und Nervenkrankheiten, 99, 555-574.
Bihrle, A. M., Brownell, H. H., & Powelson, J. A. (1986).
Comprehension of humorous and nonhumorous materials by left and
right brain damaged patients. Brain and Cognition, 5, 399-411.
Borod, J. C., 4 Caron, H. S. (1980). Facedness and emotion related to
lateral dominance, sex and expression type. Neuropsychologia,
J_8, 237-242.
Borod, J. C., Caron, H. S., 4 Koff, J. (1981). Asymmetry in positive
and negative facial expressions: Sex differences.
Neuropsychologia, 19, 819-824.
Borod, J. C., Kent, J., Koff, E., Martin, C., & Alpert, M. (1988a).
Facial asymmetry while posing positive and negative emotions:
Support for the right hemisphere hypothesis. Neuropsychologia,
26, 759-764.
Borod, J. C., Koff, E., Perlman, E., Lorch, M., & Nicholas, M.
(1986). The expression and perception of facial emotion in brain
iamaged patients. Neuropsychologia, 24, 169-180.
Borod, J. C., Koff, E., & White, B. (1983). Facial asymmetry in posed
and spontaneous expressions of emotion. Brain and Cognition, 2,
165-175.
Borod, J. C., Vingiano, W., & Cytryn, F. (1988b). The effects of
emotion and ocular dominance on lateral eye movement.
Neuropsychologia, 26, 213-220.
Bower, G. H. (1981). Mood and memory. American Psychologist, 38,
129-148.
Bowers, D., Bauer, R., Coslett, B., & Heilman, K. (1985). Processing
of faces by patients with unilateral hemispheric lesions. I.
Dissociation between judgements of facial affect and facial
identity. Brain and Cognition, _4, 258-272.
Bowers, D., & Heilman, K. (1984). Dissociation between the processing
of affective and non-affective faces: A case study. Journal of
Clinical Neuropsychology, _6, 367-374.
Bowlby, J. (1972). Separation: Volume II. Attachment and loss.
New York: Basic Books.


73
Maximum deceleratory responses
A repeated measures ANOVA was performed with Sex (male, female)
as a between subjects factor and Visual Field (left, right), and
Stimulus Type (happy, angry, fearful, disgusting, neutral) as the
witnin subjects factors. A summary of the results of this ANOVA are
depicted in Table 5. Results of this analysis revealed a main effect
of Sex [F (1, 28) = 5.16, p = .0310)]. Males (M = -9.153) showed
significantly greater maximum deceleratory HR responses than females
(M = -7.756). These findings are depicted in Figure 10. Results also
revealed a trend for a main effect of Visual Field [F_ ( 1, 28) = 3-33*
p = .0789)] with a pattern of greater deceleratory HR responses to LVF
trials (M. = -8.7OO) relative to RVF trials (_M = 8.209) (see Figure
11).
This analysis also yielded a main effect of Stimulus Type [F
(4, 112) = 4.96, p = .0010)]. Duncan's post-hoc comparisons revealed
happy slides (£ = -9.016) elicited significantly greater HR
decelerations than angry (N1 = -7.969) or fearful (M = -7.663) slides
at £ < .05. Neutral slides (M. = -8.984) also elicited significantly
greater HR decelerations than angry or fearful slides at p < .05.
These relationships are depicted in Figure 12.
Findings also revealed a significant Visual Field x Sex
interaction [F_ ( 1, 28) = 7.86, £ = .0091 )]. Simple effects testing
revealed that males showed significantly greater deceleratory HR
changes in LVF (M = -9.776) relative to RVF (M = -8.530) trials,
depicted in Figure 13- Results also revealed a significant Sex x


Reaction Time r Reaction Time
66
600 -
550 -
500 -
450 -
400 -
552.16

556.34
485.49
477.10

Left
Right
Visual Field
Females
Males
;ure 5.
Experiment I--Reaction Time Analysis (minus outliers),
Visual Field x Sex Interaction.
Happy Angry Fearful Disgusting Neutral
Emotion
Females
Males
Figure 6. Experiment IReaction Time Analysis (minus outliers), Sex
x Stimulus Type Interaction.


26
reported a consistent relationship between (a) depressive-catastrophic
reaction and left hemisphere damage and (b) indifference/minimization
of deficits and right hemisphere damage. Recently, Heilman and
colleagues (Heilman et al., 1 975; Heilman, Schwartz, 4 Watson, 1978;
Heilman, Watson, & Bowers, 1983) have noted indifference reactions
occur with striking frequency in RHD patients with the neglect
syndrome suggesting that the two syndromes may be associated in some
manner. This also suggests that right hemisphere changes associated
with inappropriate euphoria and indifference may represent either (a)
two points on a continuum or (b) two distinct emotional reactions
following brain injury, one of them sharing a common mechanism with
the unilateral neglect syndrome.
Using the Depression scale of the Minnesota Multiphasic
Personality Inventory as an index of depressive symptomatology,
Gasparrini, Satz, Heilman, and Coolidge (1978) reported significantly
elevated scores for LHD but not for RHD patients. More recently,
Robinson and colleagues (Robinson, 1983; Robinson 4 Price, 1982) have
also found that LHD patients were more likely to become clinically
depressed and that RHD patients were more likely to be inappropriately
euphoric. In addition, these results were not correlated with overall
cognitive impairment, suggesting that the patient's emotional
reactions to their deficits cannot solely account for these
findings. Location of damage within the hemisphere has been found to
be important in that these emotional reactions are more frequently
associated with damage to anterior, frontal regions (Kolb 4 Milner,
1981; Robinson 4 Benson, 1981).


3
processing of emotional stimuli Two major areas of research have
addressed this specific question: those which examine the processing
of the prosodic elements of speech and those which examine the
processing of affective faces. Related areas of investigation will
also be discussed.
Right Hemisphere Superiority for Recognizing
Emotional Aspects of Stimuli
Emotional Prosody Studies
It is well known that in right handers, the left hemisphere is
more adept than the right hemisphere in decoding the linguistic
content (semantic and phonemic elements) of speech (Benson &
Geschwind, 1971). However, speech may carry at least two levels of
information content: the linguistic content which conveys what is
said and the prosodic content which conveys the way in which it is
said. Prosodic elements which are defined as pitch, tempo, and
rhythm, carry information about the emotional as well as nonemotional
content of prosodic speech (Paul, 1909). Nonemotional prosody is
important for conveying whether a sentence is a question, a statement,
or a command. Emotional prosody is critical for conveying affective
information.
In 1975, Heilman, Scholes, and Watson studied the ability of
right temporo-parietal and left temporo-parietal damaged patients to
identify affective prosody. Patients were presented with semantically
neutral sentences which were read in one of four emotional tones
happy, sad, angry, or indifferent. In this study, the subject's task


86
Type interaction [F (4, 112) = 2.10, £ = .0875)], depicted in Figure
19. This pattern of findings revealed for male Ss only a Visual Field
x Stimulus Type interaction which approached significance. This
pattern suggested greater SCR in LVF (M = .1302) relative to RVF (fi =
.0651) for angry slides. In addition, happy 3lides elicited greater
SCR in RVF (M = .1268) compared to LVF (M = .0736) trials.


PSYCHOPHYSIOLOGICAL AND REACTION TIME RESPONSES
TO LATERALLY PRESENTED EMOTIONAL STIMULI
BY
CYNTHIA RODRIGUES CIMINO
A DISSERTATION PRESENTED TO THE GRADUATE SCHOOL
OF THE UNIVERSITY OF FLORIDA IN PARTIAL FULFILLMENT
OF THE REQUIREMENTS FOR THE DEGREE OF
DOCTOR OF PHILOSOPHY
UNIVERSITY OF FLORIDA
1988


I certify that I have read this study and that in my opinion it
conforms to acceptable standards of scholarly presentation and is
fully adequate, in scope and quality, as a dissertation for the degree
of Doctor of Philosophy.
YJUM
Dawn Bowers
Associate Professor of Clinical and
Health Psychology
I certify that I have read this study and that in my opinion it
conforms to acceptable standards of scholarly presentation and is
fully adequate, in scope and quality, as a dissertation for the degree
of Doctor of Philosophy.
Hugh C. Qvis, Jr,
Professor of Clinical and Health
Psychology
I certify that I have read this study and that in my opinion it
conforms to acceptable standards of scholarly presentation and is
fully adequate, in scope and quality, as a dissertation for the degree
of Doctor of Philosophy.
^jJUtvu C5
Eileen B. Fennell
Professor of Clinical and Health
Psychology
I certify that I have read this study and that in my opinion it
conforms to acceptable standards of scholarly presentation and is
fully adequate, in scope and quality, as a dissertation for the degree
of Doctor of Philosophy.
M-
ussell M. Bauer
Associate Professor of Clinical and
Health Psychology


70
Table 4
Summary ofAnalysis of Variance: Experiment I, Percent Correct (minus
outliers)
Source
df
ss
F
P
Sex
1 ,
26
.9794
2.42
-
Visual Field
1,
26
.0009
.01
-
Visual Field x Sex
1 ,
26
.0552
.87
-
Hand
1,
26
.0498
.76
-
Hand x Sex
1 ,
26
.0091
.14
-
Stimulus Type
4,
104
2.0531
6.13
**
Stimulus Type x Sex
4,
104
.2958
.88
-
Visual Field x Hand
1 ,
26
.0205
.27
-
Visual Field x Hand x Sex
1 ,
26
.0180
.24
-
Visual Field x Stimulus Type
104
.3694
1.86
-
Visual Field x Stimulus Type x
Sex
4,
104
.2289
1 .15
-
Hand x Stimulus Type
4,
104
. 3254
1 .16
-
Hand x Stimulus Type x Sex
4,
104
.1308
.47
-
Visual Field x Hand x Stimulus
Type
4,
104
.0452
.28
-
Visual Field x Hand x Stimulus
x Sex
Type
4,
104
. 0866
.54
-
** p < .01 .


I certify that I have read this study and that in my opinion it
conforms to acceptable standards of scholarly presentation and is
fully adequate, in scope and quality, as a dissertation for the degree
of Doctor of Philosophy.
I certify that I have read this study and that in my opinion it
conforms to acceptable standards of scholarly presentation and is
fully adequate, in scope and quality, as a dissertation for the degree
of Doctor of Philosophy. x /
Professor of Neurology
This dissertation was submitted to the Graduate Faculty of the
Department of Clinical and Health Psychology in the College of Health
Related Professions and to the Graduate School and was accepted as
partial fulfillment of the requirements for the degree of Doctor of
Philosophy.
December, 1988
Dean, College of Health Related
Professions
Dean, Graduate School


Maximum Heart Rate Deceleration
75
Figure 10. Experiment IIMaximum Deceleratory Heart Rate Analysis,
Sex Main Effect.
Figure 11. Experiment II--Maximum Deceleratory Heart Rate Analysis,
Visual Field Trend.


79
Table 6
Summary of Analysis of Variance: Experiment II, Heart Rate
Acceleration
Source
df
SS
F
P
Sex
1 ,
28
28.2236
.65
-
Visual Field
1.
28
.521 4
.13
-
Visual Field x Sex
1 ,
28
13-2770
3-36
-
Stimulus Type
1,
1 12
28.6964
1.74
-
Stimulus Type x Sex
112
57.5180
3.48
**
Visual Field x Stimulus Type
4,
1 12
13.3162
.73
-
Visual Field x Stimulus Type x Sex
4,
112
9.4844
.52
-
* *
p < .01.


124
Schwartz, G. E., Weinberger, D. A., & Singer, J. A. (1981).
Cardiovascular differentiation of happiness, sadness, anger and
fear following imagery and exercise. Psychosomatic Medicine, 43,
343-364.
Sequndo, J. P., Naguet, R., & Buser, P. ( 1955). Effects of cortical
stimulation on electrocortical activity in monkeys. Journal of
Neurophysiology, 18, 236-245.
Sergent, J. (1983). Role of the input in visual hemispheric
asymmetries. Psychological Bulletin, 93 481-512.
Sergent, J., & 3indra, D. (1981). Differential hemispheric processing
of faces: Methodological considerations and reinterpretation.
Psychological Bulletin, 89, 541-554.
Silberman, E. K., & Weingartner, H. (1986). Hemispheric
lateralization of functions related to emotion. Brain and
Cognition, 5, 322-353.
Silberman, E. K., Weingartner, H., & Post, R. M. (1983a). Thinking
disorder in depression: Logic and strategy in an abstract
reasoning task. Archives of General Psychiatry, 40, 775-780.
Silberman, E. K., Weingartner, H., Stillman, R., Chen, H. J., & Post,
R. M. (1983b). Altered lateralization of cognitive processes in
depressed women. American Journal of Psychiatry, 140, 1340-1344.
Smokier, I. A., & Shevrin, I. (1979). Cerebral lateralization and
personality style. Archives of General Psychiatry, 36, 949954.
Sokolov, E. N. (1963). Perception and the conditioned reflex.
Oxford: Pergamon Press.
Sroufe, L. A., & Wunsch, J. P. (1972). The development of laughter in
the first year of life. Child Development, 43, 1325-1344.
Steinberg, C. R., & Campos, J. J. (1983). The facial expression of
anger in seven-month-old infants. Child Development, 54,
178-184.
Strauss, E. (1983). Perception of emotional words. Neuropsychologia,
21_, 99-103.
Strauss, E., & Moscovitch, M. (1981). Perception of facial
expressions. Brain and Language, 13, 308-332.
Suberi M., & McKeever, W. F. (1977). Differential right hemisphere
memory storage of emotional and non-emotional faces.
Neuropsychologia, 15, 757-765.


TABLE OF CONTENTS
Page
ACKNOWLEDGMENTS
ABSTRACT vi
INTRODUCTION 1
Right Hemisphere Superiority for Recognizing Emotional
Aspects of Stimuli 3
Emotional Prosody Studies 3
Affective Faces Studies 8
Related Research 13
Right Hemisphere Dominance in Regulation of Mood and Affect....16
Emotional Activation Studies 16
Mood and Affect Studies 20
Left Hemisphere Superiority for Positive Affect; Right
Hemisphere Superiority for Negative Affect 25
Mechanisms of Emotional Processing: The Role of Arousal 31
Neuropsychological Models of Emotional Processing 36
The Model of Fox and Davidson 37
Kinsbourne's Model 38
Tucker's Model 40
Heilman's Model 42
Critical Issues 45
Hypotheses and Predictions 49
METHOD 52
Subjects 52
Experiment I: Reaction Time Task 52
Stimuli 52
Apparatus 53
Procedure 54
Experiment II: Psychophysiological Responses to Laterally
Presented Emotional Material 55
Stimuli 55
Apparatus 55
Procedure 57
IV


BIOGRAPHICAL SKETCH
Cynthia Rodrigues Cimino was born May 11, 1958, in Marshfield,
Massachusetts. She earned her Bachelor of Science and Master of
Science degrees from the University of Florida. She will obtain her
Ph.D. in clinical and health psychology from the University of Florida
in December, 1988.


24
relative impairment on right hemisphere tasks as compared to left
hemisphere tasks.
Several studies have also reported the occurrence of "reversed
lateralization" or "functional delateralization" in depressed patients
on tasks of verbal and nonverbal processing. Bruder (1983) published
a review of dichotic listening studies and concluded that depressed
patients demonstrated decreased lateralization on both verbal as well
as nonverbal dichotic listening tasks. Several authors, however, have
suggested that decreased lateralization is present primarily on
nonverbal tasks (Coulbourn & Lishman, 1974; Johnson & Crockett,
1982). Evidence of reversed lateralization has also been reported by
some authors. Silberman, Weingartner, Stillman, Chen, & Post (1983b)
have reported a left visual field superiority on a verbal task in a
sample of depressed females. Similarly, research in depressed
patients has also found (Flor-Henry, 1979; Flor-Henry & Koles, 1980)
increased parietal activity during rest, with left temporal activation
during spatial task3 and right parietal activation during verbal
tasks. These findings are at variance with predicted asymmetries in
normal populations. Hommes and Panhuysen (1971), using a small sample
of depressed patients, have reported that right sided sodium amytal
injections resulted in a degree of aphasia for all subjects.
Furthermore, this finding was significantly correlated with the
severity of depressive symptomatology.
In summary, a large body of the neuropsychology literature on
emotional processing has suggested a right hemisphere dominance in the
comprehension and expression of emotional information. These findings


102
Similarly, angry and fearful slides elicited significantly greater HR
accelerations than neutral slides.
A question which arises concerns the basis for the differential
HR patterning across different emotion categories. Recently, several
authors have systematically examined the relationship between
emotional imagery and production of cardiac responses. Jones and
Johnson (1978; 1980) have argued that two major factors contribute to
the particular patterning of HR responses. These include (a) the
emotional content of the image (pleasant vs unpleasant); and (b) the
level of activity inherent in the image (high activity vs low
activity). By systematically manipulating these two factors, Jones
and Johnson (1980) found that high activity images produced
significantly greater HR accelerations than low activity images.
These findings are congruent with reports from other investigators who
have noted that instructional manipulations on imagery tasks can
produce significant effects on autonomic responses (Lang, 1977;
Melamed, 1969). Similarly, by requesting active engagement in
stimulus processing, Bauer and Craighead (1979) also revealed
significant effects on psychophysiological responses.
Jones and Johnson (1980) also report that negative-low activity
images resulted in greater HR accelerations relative to positive-low
activity images. Such findings suggest that the emotional valence of
the image was also a significant contributing factor to differential
HR responses. While the effect of valence of affective stimulus on HR
responding had been previously investigated, these authors provide


103
support for the role of this factor when differences in activity level
of imagery are accounted for.
It may be the case that a priori differences in activity level
across emotional categories could, in part, account for differential
patterns of HR acceleration across emotional categories. Lang (1979)
has suggested that emotional imagery results in patterns of autonomic
activity very similar to those found in the actual emotional
situation. In this view, those emotions which inherently contain a
greater motor component (i.e., flight or fight type of responding) as
in the case of anger or fear, would result in significantly greater HR
acceleratory responses than those emotions with less inherent motor
demand; predictions which parallel the findings in the present
investigation.
It may also oe the case that these changes in HR activity across
various emotional categories may serve to differentially prepare the
organism for action. In accounting for differences in acceleratory
and deceleratory HR responses, Obrist and colleagues (197^) 'nave
emphasized the relationship between motor requirements and cardiac
activity. In addition, these authors suggest that conditions may
exist whereby increases in HR are observed without overt changes in
somatic activity. In support of this view, Freyschuss (1970) reported
cardiac accelerations under conditions where subjects were instructed
to tense or move an arm despite their inability to do so because of
experimentally induced paralysis. Based on these findings, it is
suggested that cardiac activity is not solely coupled with direct,
overt somatic activity but rather that cardiac activity is coupled


DISCUSSION
Several models have been proposed to account for lateral
symmetries observed on tasks of emotional processing. One model, the
right hemisphere model, suggests that the right hemisphere is globally
more involved in all aspects of emotional processing including the
cognitive encoding/decoding, arousal-activation, and behavioral
responses to emotional stimuli. In this view, emotional stimuli
presented initially to the right hemisphere via the left sensory
channel (left visual field, left ear) elicit significantly greater
arousal responses and result in significantly quicker/more accurate
detection than emotional stimuli presented initially to the left
hemisphere via the right sensory channel (right visual field, right
ear) .
A second model, the hemispheric valence model, proposes that the
left hemisphere is more adept at processing positive emotions and the
right hemisphere is more adept at processing negative emotions.
Within this framework, positive emotional stimuli initially presented
to the left hemisphere would elicit significantly greater arousal
responses and result in significantly quicker/more accurate detection
than positive emotional stimuli presented initially to the right
hemisphere. Likewise, negative emotional stimuli presented to the
right hemisphere would elicit significantly greater arousal responses
88


46
played by each remains unclear. Some investigators have argued that
the right hemisphere is globally involved in ail aspects of emotional
processing including the cognitive encoding/decoding of emotional
stimuli, arousal-activation responses to emotional stimuli and
behavioral responses to these stimuli (Heilman et al., 1983; bey &
Bryden, 1979). Other investigators have argued that the two
hemispheres differ in terms of the type of emotions that are
preferentially mediated by each (Fox St Davidson, 1984; Kinsbourne Sc
Bemporad, 1984; Tucker, 1981). The most popular version of this view
i3 that the left hemisphere is dominant for positive (approach)
emotions, whereas the right hemisphere is dominant for negative
(avoidance) emotions.
Still others have argued that this positive-negative dichotomy in
hemispheric processing of emotions is artifactual and actually relates
to hemispheric differences in arousal and preparation for action
(e.g., activation) (Heilman, 1988, personal communication). In this
view, the right hemisphere is dominant for mediating
arousal/activation and as such, is more intrinsically involved in
processing emotional stimuli that have greater "preparatory"
significance for survival (i.e., fight-flight emotions such as anger
and fear). In contrast, the left hemisphere is more involved in
mediating nonpreparatory emotions (i.e., happiness, sadness, disgust)
that place less immediate or "phasic" attentional demands on the
organism for survival.
In order to distinguish among these models, it would be necessary
to determine whether different categories of emotional stimuli result


the overall mean RT of females for each of the 20 Visual Field x Hand
x Stimulus Type conditions, these two Ss possessed 8 (40$ of total)
and 10 (50$ of total) mean reaction times which fell two standard
deviations above the overall mean performance of female Ss. Of the
remaining 13 female Ss, no S possessed a single mean reaction time
greater than 2 standard deviations above the overall mean of female
Ss.
For this reason, a second analysis of the RT data was conducted
in which the data from the two female Ss noted above was excluded. A
summary of the results of this analysis are depicted in Table 2.
Findings revealed no significant main effects. The main effect of Sex
observed on the preceding analysis was no longer significant
suggesting that this effect may have been significantly influenced by
the markedly slowed performance of the two female Ss noted above.
Findings, however, did reveal a significant Visual Field x Sex
interaction [F (1, 26) = 5.20, £ < .0311)] as in the preceding
analysis, depicted in Figure 5. Simple effects testing revealed a
pattern of findings similar to those noted in the first analysis with
no significant difference between LVF and RVF performance for female
Ss. Males, however, performed significantly faster to RVF stimuli
relative to LVF stimuli. A trend for the Sex x Stimulus Type
interaction was also noted [F (4, 104) = .0722, £ = .0722)] which had
previously attained significance in the first analysis. This
interaction is depicted in Figure 6. The Hand x Visual Field x
Stimulus Type interaction significant, in the first analysis, failed
to reach significance in the present analysis.


8
In summary, investigations in both normal and brain impaired
subjects have provided evidence that the right hemisphere is
preferentially involved in the comprehension and expression of
emotional prosodic elements of speech and other nonverbal
vocalizations. Similarly, a large body of literature has also
investigated the role of the right hemisphere in the processing of
affective faces.
Affective Faces Studies
In 1980, DeKosky, Heilman, Bowers, and Valenstein reported a
study which investigated RHD and LHD patients' ability to make neutral
facial discriminations as well as affective facial discriminations.
They found that RHD patients performed more poorly than LHD patients
on both facial affect judgements as well as neutral facial
discrimination. In fact, when the two groups were statistically
equated for performance on the neutral discrimination task,
differences between RHD and LHD patients on the affective facial
discrimination task disappeared. These findings suggested that RHD
patients' poor performance on facial affect judgements can be solely
accounted for by their poor performance in facial discrimination
ability. This has led to the question of whether processing the
emotionality of a face, in fact, involves a "stimulus-content
dimension" in its own right or whether such processing merely involves
an increase in the configurational complexity of the stimulus and
consequently increases the demand on right-hemisphere mediated
visuospatial skills.


12
these findings, Suberi and McKeever argued that the LVF effect for
processing nonemotional faces is significantly enhanced by
presentation of emotional faces.
McKeever and Dixon (1981) used emotional imagery and neutral
faces to investigate right hemisphere effects in processing of
affective material. They instructed subjects to imagine that
something very sad happened to a number of predetermined target
faces. In a subsequent target/nontarget discrimination task with
lateralized presentations, they report that the use of emotional
imagery significantly enhanced LVF (right hemisphere) performance.
This effect, however, was demonstrated in female subjects only.
Safer (1981) reported a study in which subjects memorized faces
by either empathizing with their emotional expressions or by labeling
the emotional expressions. Results demonstrated that subjects who
used empathy recognized more faces presented to the LVF than RVF. No
laterality effect was demonstrated for those who labeled faces. This
laterality effect for the empathy condition, however, was found for
male subjects only. Similarly, Buchtel, Campari, DeRisio, and Rota
(1978) reported faster responding to both positive and negative
stimuli presented in LVF relative to neutral targets. Hansch and
Pirozzolo (1980) and Strauss and Moskovitch (1981) also reported a LVF
effect for neutral and emotional faces.
In summary, investigations in both normal and brain impaired
subjects have supported the notion that the right hemisphere is
preferentially involved in the processing of affective faces. In
addition, several studies have also suggested that right hemisphere


59
Table 1
Summary of Analysis of Variance: Experiment I, Reaction Time
Source
df
SS
F
P
Sex
1 ,
28
6.0277
4.44
#
Visual Field
1,
28
.0016
.25
-
Visual Field x Sex
1,
28
.0267
4.06
*
Hand
1 ,
28
.0375
OO
-
Hand x Sex
1 ,
28
.0032
.07
-
Stimulus Type
4,
112
.0874
1.38
-
Stimulus Type x Sex
4,
112
.1674
2.64
#
Visual Field x Hand
1.
28
.0000
.01
-
Visual Field x Hand x Sex
1 ,
28
.0008
.05
-
Visual Field x Stimulus Type
4,
112
.0334
35
-
Visual Field x Stimulus Type x Sex
4,
11 2
.0966
1 .02
-
Hand x Stimulus Type
4,
1 12
.1497
L*J
CD
-
Hand x Stimulus Type x Sex
4,
112
.0386
36
-
Visual Field x Hand x Stimulus Type
4,
1 12
.1995
2.49
#
Visual Field x Hand x Stimulus Type
x Sex
4,
112
.0357
.45
"
*p < .05.


make affective judgements cannot be accounted for solely by
differences in the visuoperceptual processes underlying facial
identity discrimination.
Recent support for the relative dissociation of facial identity
judgements from facial affect judgements has also been provided by
Tranel, Damasio, and Damasio (1988). Tranel et al. described four
patients with bilateral lesions of occipitotemporal or temporal
regions whose performance on facial affect tasks were significantly
better than their performance on facial identity tasks.
Research in normal subjects has also investigated the role of the
right hemisphere in the processing of affective faces. Ley and Bryden
(1979) tachistoscopically presented faces to the left visual field
(LVF) and right visual field (RVF), and subjects made either facial
identity judgements or facial affective judgements. They found a LVF
superiority for both tasks. However, when performance on the facial
affect task was reanalyzed using performance on the facial identity
task as a covariate, the LVF superiority for making affective facial
judgements remained. These findings, which are similar to those of
Bowers et al. (1985), suggest that the right hemisphere superiority
for processing facial affect exists above and beyond the superiority
for processing facial identity.
in 1977, Suoeri and McKeever reported a reaction time (RT) task
in which subjects responded to previously memorized emotional or
nonemotional faces presented in LVF or RVF. Subjects who memorized
emotional faces showed significantly faster reaction times to LVF
targets than subjects who memorized nonemotional faces. Based on


BPM from Baseline BPM from Baseline
33
Females
Post-Stimulus Seconds
Females
Post-Stimulus Seconds
Females -o- LVF-Fearful
RVF-Fearful
Post-Stimulus Seconds

c


ra
CO
E
o
2
0.
CQ
Females
Post-Stimulus Seconds

C
UJ
ra
CO
E
o
i
0.
CO
o -
-1 -
-2-
-3-
-4 -
-5 -
Females
-Q-
LVF-Neutral
RVF-Neutral
TD (T
I i i 1 I 1 I I I I 1
D1 D2 D3 D4 D5 D6 D7 D8
Post-Stimulus Seconds
Figure 18. Experiment lI--Second x Second Post-Stimulus Heart Rate
Changes in Female Subjects for Each Visual Field x
Stimulus Type Condition.


6
of emotional processing in normal, neurologically intact subjects have
also examined the right hemisphere's role in the processing of
emotional prosody. Using a dichotic listening (DL) procedure, Haggard
and Parkinson (1971) paired speech babble with short sentences spoken
in one of four emotional tones. They found that accuracy in
identifying the emotional tone was significantly better on left ear
trials, suggesting a right hemisphere superiority.
Similarly, Safer and Leventhal (1977) used monoaural presentation
of sentences with positive, negative, and neutral content spoken in a
positive, negative, or neutral tone. Results demonstrated that
subjects who listened to sentences in their left ear tended to use the
intonation in making their judgements whereas subjects who listened to
sentences in their right ear tended to use the content of the sentence
in making judgement. Interpretation of these findings, however, has
been questioned because of the use of a between groups comparison for
each ear presentation. The findings are suggestive, nevertheless.
Ley and Bryden (1982), using a DL procedure, had subjects report
on the emotional tone and content of a sentence arriving at either the
left or right ear (specified on each trial). Subjects were more
accurate in judging the emotional tone of the sentence when monitoring
the left ear and more accurate in judging the content of the sentences
when monitoring the right ear.
In a subsequent experiment, Bryden, Ley, and Sugarman (1982)
investigated hemispheric differences in ability to judge the emotional
tone of musical stimuli They did this by taking advantage of the
fact that in Western culture music written in a major key is generally


90
emotionai/nonemotional stimuli; and (c) to determine whether certain
categories of emotion (positive/negative; preparatory/nonpreparatory)
induce asymmetric arousal/activation, depending on the hemisphere to
which they are initially presented.
Findings from Experiment I, in which RTs were made to midline
neutral stimuli that were preceded by lateralized stimuli of different
emotional valences, failed to support any of the laterality models of
emotion. For example, no overall superiority for lateralized
emotional warning stimuli presented to the right versus left
hemisphere was found. Likewise, no hemispheric specific emotional
valence effects were observed. Similarly, no evidence was present for
the view of hemispheric differences in "preparatory" versus
"nonpreparatory" emotions. What was found, however, was the
following: (a) females showed no laterality effects of any kind; and
(b) males, on the other hand, had overall faster RTs to neutral
stimuli that were preceded by emotional warning stimuli in the RVF
(left hemisphere) versus LVF (right hemisphere). This finding, which
suggests that emotional stimuli induce greater behavioral activation
when presented to the left hemisphere than to the right hemisphere, is
the opposite of that predicted by any model arguing for superiority of
the right hemisphere in mediating emotional responsivity.
There are several possibilities which might account for these
findings. In Experiment I, Ss were required to make a left-right
decision based on the emotionai/nonemotional nature of the warning
stimulus. That is, they had to respond with one hand to emotional
stimuli and with the other hand to neutral stimuli. In other words,


119
Kronfol, Z., Hamsher, K. De S., Digre, K., & Wazir, R. (1978).
Depression and hemisphere functions: Changes associated with
unilateral ECT. British Journal of Psychiatry, 132, 560-567.
Kuzendorf, R. G. (1982). Individual differences in imagery and
autonomic control. Journal of Mental Imagery, _6, 47-60.
Lacey, J. I. (1967). Somatic response patterning and stress: Some
revisions of activation theory. In: M. H. Appley & R. Trumball
(Eds.), Psychological stress: Issues in research. New York:
Appleton-Century-Crofts.
Lacey, J. I. (1972). Some cardiovascular correlates of sensorimotor
behavior: Example of visceral afferent feedback? In: C. H.
Hockman (Ed.), Limbic system mechanisms and autonomic function.
Springfield, IL: Charles C. Thomas.
Lacroix, D. M., & Comper, P. (1979). Lateralization in the
electrodermal system as a function of cognitive/hemispheric
manipulations. Psychophysiology, 16, 116-124.
Lang, P. J. (1977). Imagery in therapy: An information processing
analysis of fear. Behavior Therapy, j), 862-886.
Lang, P. J. (1979). A bio-informational theory of emotional
imagery. Psychophysiology, 16, 495-512.
Lansing, R. W., Schwartz, E., & Lindsey, D. B. (1959). Reaction time
and EEG activation under alerted and nonalerted conditions.
Journal of Experiment Psychology, 58, 1-7.
Levy, J. (1969). Possible basis for the evolution of lateral
specialization of the human brain. Nature, 224, 614-615.
Ley, R. G., & Bryden, M. P. (1979). Hemispheric differences in
processing emotions and faces. Brain and Language, 7, 127-138.
Ley, R. G., & Bryden, M. P. (1982). Dissociation of right and left
hemispheric effects for recognizing emotional tone and verbal
content. Brain and Cognition, 1, 3-9.
Lindsley, D. (1970). The role of nonspecific reticulo-thalamo-
cortical systems in emotion. In: P. Black (Ed.), Physiological
correlates of emotion. New York: Academic Press.
Lishman, W. A. (1968). Brain damage in relation to psychiatric
disability after head injury. British Journal of Psychiatry,
114, 373-410.


21
epileptic patients with unilateral foci, head injury patients, and
patients with unilateral subcortical infarction or surgical removal.
In a factor analytic study which assessed interictal personality
changes associated with right temporal lobe epilepsy, Bear and Fedio
(1977) reported that right foci were more associated with affective
changes while left foci were more associated with cognitive changes.
Similarly, Taylor (1972) described a predominance of right sided foci
in a sample of epileptics with associated symptoms of depression,
anxiety, and phooias. In 1969, Flor-Henry reported that of patients
with unilateral epileptic foci, manic-depressive psychosis was found
twice as many times in individuals with right sided foci than in those
with left sided foci.
Lishman (1968), in his review of 144 cases of head injury, found
that affective disturbances were more common following right
hemisphere injury while cognitive/intellectual changes were more often
seen following left sided injury. These findings parallel those of
Bear and Fedio (1977) which looked at changes associated with left and
right epileptic foci. Recently, mania has also been reported
following right thalamectomy (Whitlock, 1982) and right thalamic
infarct (Cummings & Mendez, 1984).
Investigators have also looked at signs of hemispheric
dysfunction in patients with primary mood disturbance as a means of
investigating lateralized hemispheric functioning in relation to
disorders of mood and affect. Several investigators, utilizing EEC as
a measure of hemispheric activity, have reported low left versus right
hemisphere activity in depressed patients with differences occurring


57
Procedure
Prior to the experiment, Ss were requested to list specific
episodes from their own lives which they considered happy, angry,
fearful, disgusting, or neutral. Subjects were requested to list five
episodes for each category for a total of 25.
The experiment took place in a quiet, dimly lit room. Subjects
were seated in a comfortable, recliner chair positioned 152 cm from
the projection screen. After electrode placement and a 20-minute
adaptation period, Ss were instructed to refrain from any unnecessary
movement during the experiment.
Subjects were presented with a neutral or emotional slide in left
or right visual field for 8 seconds. Subjects were instructed to
maintain fixation on the centrally positioned circle throughout slide
presentation. To decrease Ss' habituation and encourage continued
processing of the stimuli during this time, Ss were also instructed to
recall one of the five episodes of the same valence as the slide
presented. Approximately 10 seconds after slide offset, Ss were asked
to indicate the valence of the slide and responses were recorded
manually on a separate sheet. Inter-trial interval varied randomly
from 23 to 45 seconds to minimize occurrence of anticipatory HR and
SCR. Subjects received a total of 120 trials with visual field of
presentation randomized and counterbalanced across stimulus type.
Subjects also received 10 practice trials for maintained fixation
prior to the experiment utilizing neutral stimuli only.


117
Heilman, K. M., Watson, R. T., 4 3owers, D. (1983). Affective
disorders associated with hemispheric disease. In: K. M.
Heilman (Ed.), Neuropsychology of human emotion. New York:
Guillford Press.
Heller, W., 4 Levy, J. (1981). Perception and expression of emotion
in right handers and left handers. Neuropsychologia, 19,
363-372.
Holloway, F. A., & Parsons, 0. A. (1969). Unilateral brain damage and
bilateral 3kin conductance levels in humans. Psychophysiology,
6, 138-148.
Hommes, 0. R., & Panhuysen, L. H. H. M. (1971). Depression and
cerebral dominance: A study of bilateral intracarotid amytal in
eleven depressed patients. Psychiatria, Neurologa,
Neurochirurgia, 74, 259-270.
Hugdahl, K., Franzon, M., Anderson, B., 4 Walldebo, G. (1983). Heart
rate responses to lateralized visual stimuli. Pavlovian Journal
of Biological Sciences, 18, 186-198.
Hugdahl, K., Wahlgren, C., 4 Wass, T. (1982). Habituation of the
electrodermal orienting reaction is dependent on the cerebral
hemisphere initially stimulated. Biological Psychology, 15,
49-62.
Ikeda, Y., 4 Hirai, H. (1976). Voluntary control of electrodermal
activity in relation to imagery ana internal perception scores.
Psychophysiology, 13, 330-333.
Iversen, S. D. (1977). Brain dopamine systems and behavior. In:
L. L. Iversen, S. D. Iversen, 4 S. H. Snyder (Eds.), Handbook of
psychopharmacology, Volume 8. Drugs, neurotransmitters and
behavior. New York: Plenum Press.
Izard, C. E. (1977). Visual attention in infancy: Processes, methods
and clinical applications. In: I. C. Uzgiris 4 T. Weizmann
(Eds.), The structuring of experience. New York: Plenum Press.
Izard, C. E., Hubner, R. R., Risser, D., McGuiness, G. C., 4
Dougherty, L. M. (1980). The young infant's ability to produce
discrete emotional expression. Developmental Psychology, 16,
132-140.
Johnson, 0., 4 Crockett, D. (1982). Changes in perceptual asymmetries
with clinical improvement of depression and schizophrenia.
Journal of Abnormal Psychology, 91, 45-54.


85
Table 7
Summary of Analysis of Variance: Experiment II, Skin Conductance
Responses
Source
df
SS
F
P
Sex
1,
28
.6410
2.60
-
Visual Field
1,
28
.001 1
.08
-
Visual Field x Sex
1,
28
.0004
.03
-
Hand
1,
28
.0056
.13
-
Hand x Sex
1 ,
28
.0399
.94
-
Stimulus Type
112
.0167
36
-
Stimulus Type x Sex
11 2
.0628
1 .35
-
Visual Field x Hand
1,
28
.0007
.19
-
Visual Field x Hand x Sex
1,
28
.0018
.43
-
Visual Field x Stimulus Type
4,
1 12
.0664
1.61
-
Visual Field x Stimulus Type x
Sex
4,
1 1 2
.0868
2.10
-
Hand x Stimulus Type
4,
1 12
.0186
1.31
-
Hand x Stimulus Type x Sex
4,
1 1 2
.0097
.69
-
Visual Field x Hand x Stimulus
Type
4,
112
.0249
1.59
-
Visual Field x Hand x Stimulus
x Sex
Type
4,
1 12
.01 12
.71
-


23
Findings from investigations of galvanic skin responses (GSR)
have provided only partial support for the presence of lateralized
dysfunction in depressed patients. Schneider (1983) reported lower
right handed GSRs in a depressed sample. In addition, Myslobodsky and
Horesch (1978) have noted higher left handed GSRs in depressed
subjects. Toone, Cooke, and Lader (1981), however, were unable to
replicate these findings. Such discrepancies may be accounted for by
conflicting reports which suggest that GSR responses may be controlled
ipsilaterally, contralaterally, or bilaterally (Holloway & Parsons,
1969; Lacroix & Comper, 1979; Myslobodsky & Rattok, 1977).
The performance of depressed patients on tests likely to require
greater right hemisphere processing has also been investigated.
Several studies sugggest that depressed subjects perform more poorly
on vi3uospatial tasks than verbal tasks (Flor-Henry, 1976; 1983;
Goldstein, Filskov, Weaver, & Ives, 1977; Kronfol, Hamsher, Digre, &
Wazir, 1978). Silberman, Weingartner, and Post (1983a) have also
suggested that the pattern of errors in depressed subjects closely
resembles that of right temporal lobectomized patients and that the
degree of impairment is correlated with the overall severity of
depression. However, as cautioned by Weingartner and Silberman
(1982), the impaired verbal learning and memory performance that is
frequently observed in depressed patients also points to left
hemisphere dysfunction. Because of decrements observed on left as
well as right hemisphere tasks, Weingartner and Siloerman (1982)
suggested that it is best to describe the deficits observed as a


Abstract of Dissertation Presented to the Graduate School
of the University of Florida in Partial Fulfillment of the
Requirements for the Degree of Doctor of Philosophy
PSYCHOPHYSIOLOGICAL AND REACTION TIME RESPONSES
TO LATERALLY PRESENTED EMOTIONAL STIMULI
BY
CYNTHIA RODRIGUES CIMINO
December, 1988
Chair: Dawn Bowers
Major Department: Clinical and Health Psychology
Neuropsychological investigations of brain injured and
neurologically intact subjects have suggested that the two hemispheres
differ in terms of their contribution to emotional processing.
Several different models have been proposed to account for these
observed hemispheric asymmetries. One model, the right hemisphere
emotion model, suggests that the right hemisphere (RH) is globally
involved in all aspects of emotional processing. A second model, the
hemispheric valence model, suggests that the left hemisphere (LH) is
dominant for processing positive emotions, whereas the RH is dominant
for processing negative emotions. A third model, the preparatory
model, suggests the RH is dominant for mediating arousal/activation
and, as such, is more intrinscally involved in processing emotional
stimuli that have greater "preparatory" significance for survival
(i.e., fight-flight emotions such as anger and fear). In contrast,
the LH is more involved in mediating nonpreparatory emotions (i.e.,
happiness, sadness, disgust) that place less immediate or "phasic"
vi


53
of each of four different valences (happy, angry, fearful, and
disgusting). The emotional scenes were selected from a variety of
materials including magazines and photography books. The scenes used
in this experiment did not include familiar landmarks or personalities
in an effort to avoid possible confounding effects of familiarity on
RT and HR responses to these stimuli. The stimuli were rated for type
and intensity of affect by 20 (10 male, 10 female) University of
Florida students who did not participate in the present experiments.
All stimuli averaged at least 91% agreement. Mean intensity ratings
(on a scale of 1 to 5) for the five categories were happy, 3.4; angry,
3.6; fearful, 3.9; disgusting, 3.8; and neutral, .3.
Apparatus
Slides were projected onto a 40x35-cm Kodak milk-glass, rear view
projection screen. A 5-mm diameter red light-emitting diode (LED) was
placed at the center of the screen to serve as a central fixation
point and imperative stimulus in the RT task. Two spring loaded keys
were placed 30 cm to the left and right of body midline. The timing,
presentation of stimuli, and recording of Ss' responses from release
of spring loaded keys were accomplished by an I3M-PC microcomputer
interfaced with BRS logic. Slides were projected at 5 degrees of
visual angle lateral to the central fixation LED. Attached to the
projector lens were Uniblitz 325B high speed shutters which allowed
maximum rise and fall time of slide presentation. The room was dimly
lit to avoid the effects of visual startle during stimulus
presentation.


118
Jones, G. E., & Johnson, H. J. (1978). Physiological responding
during self-generated imagery of contextually complete stimuli.
Psychophysiology, 15, 439-446.
Jones, G. E., & Johnson, H. J. (1980). Heart rate and somatic
concomitants of mental imagery. Psychophysiology, 17, 339~347.
Karlin, R., Weinapple, M., Rochford, J., & Goldstein, L. (1979).
Quantitated EEG features of negative affective states: Report of
some hypnotic studies. Research Communications in Psychology,
Psychiatry and Behavior, _4, 397-413.
King, F. L., & Kimura, D. (1972). Left ear superiority in dichotic
perception of vocal nonverbal sounds. Canadian Journal of
Psychology, 26, 111-116.
Kinsbourne, M. (1970). The cerebral basis of lateral asymmetries in
attention. Acta Psychologia, 33. 163-201.
Kinsbourne, M. (1972). Eye and head-turning indicates cerebral
lateralization. Science, 176, 539-541.
Kinsbourne, M. (1973). The control of attention by interaction
between the cerebral hemispheres. In: S. Kornblum (Ed.),
Attention and performance, IV. New York: Academic.
Kinsbourne, M., & Bemporad, B. (1984). Lateralization of emotion: A
model and the evidence. In: N. A. Fox & R. J. Davidson (Eds.),
The psychobiology of affective development. Hillsdale, NJ:
Erlbaum.
Kirk, R. E. (1968). Experimental design: Procedures for the
behavioral sciences. Belmont, CA: Wadsworth Publishing.
Klorman, R., & Ryan, R. (1980). Heart rate, contingent negative
variation and evoked potentials during anticipation of affective
stimulation. Psychophysiology, 17, 513-523.
Klorman, R., Weissberg, R., & Weisenfeld, A. (1977). Individual
differences in fear and autonomic reactions to affective
stimulation. Psychophysiology, 4, 45-51.
Kolb, B., & Milner, B. (1981). Observations on spontaneous facial
expression after focal cerebral excisions and after intracarotid
injection of sodium amytal. Neuropsychologia, 19, 505-514.
Kolb, B., & Taylor, L. (1981). Affective behavior in patients with
localized cortical excisions: Role of lesion site and size.
Science, 214, 89-91.


2
depressive-catastrophic reaction and left hemisphere damage (LHD).
Behaviors of emotional indifference and minimization of deficits were
associated with right hemisphere damage (RHD). Similarly, Terzian
(1964) and Rossi and Rosadini (1967) reported findings from unilateral
carotid injection of sodium amytal and found depressive-catastrophic
reaction following left sided injection and inappropriate euphoria
following right sided injection. Milner (cited in Rossi & Rosadini,
1967) attempted to replicate these findings without success. In 1982,
Sackeim and colleagues reported 119 cases of pathological laughing and
crying in response to unilateral lesions. Results were congruent with
Gainotti's findings with laughing outbursts more frequent following
RHD and crying more often following LHD.
Goldstein (1948) as well as Gainotti (1972) have interpreted such
depressive-catastrophic reactions following LHD as a "normal" response
to a significant loss of physical as well as psychological function.
In contrast, emotional changes seen with RHD were interpreted by
Gainotti (1972) as an abnormal response associated with anosogno3ia or
denial of illness.
More recent interpretations of such findings, however, have
suggested that the indifference reaction may be due to the RHD
patient's inability to accurately comprehend and/or express affect.
This led to the hypothesis that the right hemisphere, in the intact
state, is superior for the perception and/or expression of emotional
material.
In the years which followed, more systematic investigation was
applied to the understanding of the right hemisphere's role in the


92
significant processing demands of the warning stimulus may have
mitigated any potential activation or preparatory effect that the
emotional stimuli may have had because processing of the warning
stimulu continued through the interstimulus interval. The use of less
complex warning stimuli such as emotional and neutral faces is one way
of testing this hypothesis.
In addition, it may al30 be of informational value to use such
warning stimuli in a simple reaction time task to look at the general
activation effect of emotional and neutral stimuli on simple reaction
time to a neutral imperative stimuli. This paradigm would serve to
eliminate the processing demands of the warning stimulus which may
have interfered with the potential activation effect of emotional
warning stimuli on right hemisphere processing.
Lastly, it may be the case that emotional warning stimuli used in
the present investigation were not arousing enough to provide adequate
activation effects in response to the imperative stimulus.
Findings the second experiment, in which measures of autonomic
arousal were obtained to laterally presented emotional/nonemotional
stimuli, were more in line with current views regarding hemispheric
differences in processing emotionaL stimuli. With SCRs, there were no
significant effects in terms of SCRs to lateralized emotional
stimuLi. However, a trend was observed (Sex x VF, p = .087), whereby
male Ss had greater SCRs to happy stimuli when they were presented to
tne RVF (left hemisphere) than to the LVF (right hemisphere); greater
SCRs occurred when angry stimuli were presented to LVF (right
hemisphere) than to the RVF (left hemisphere). Although this pattern


100
investigators who report HR accelerations in response to fearful
stimuli (Hare & Blevings, 1975; Klorman & Ryan, 1980; Vrana et al.,
1986). In addition, these results also parallel the findings of Ekman
et al. (1983) who reported HR increases in response to production of
emotional facial expressions of anger, fear, and sadness and HR
decelerations in response to disgust, surprise, and happiness.
While previous research has suggested that different patterns of
autonomic arousal are associated with different types of emotional
states, in the present study differences in autonomic responses across
different emotional categories were observed only for female
subjects. One critical question concerns why this "autonomic
patterning" to emotional stimuli should occur in females but not
males. It may be the case that female subjects were more amenable to
fully cooperating with task demands of emotional imagery. This
suggestion is supported by prior investigations which have found
significant effects for mood manipulation and emotional imagery
instruction for female but not male subjects (Delp & Sackeim, 1987;
McKeever & Dixon, 1981).
Alternatively, sex differences in autonomic patterning observed
in the present study may be related to differences in imagery ability
between males and females. Prior investigations have revealed that
differences in autonomic control and HR conditioning are related to
subject's imagery ability. Carroll, Baker, and Preston (1979) have
reported that ability to increase HR through voluntary imaging was
significantly correlated with reported vividness of the subject's
images. Ikeda and Hirai (1976) reported that ability to control SCR


123
Sackeim, H. A., Greenberg, M. S., Weiman, A. L., Gur, R. C.,
Hungerbuhler, J. P., & Geschwind, N. (1982). Hemispheric
asymmetry in the expression of positive and negative emotions.
Archives of Neurology, 39, 210-218.
Sackeim, H. A., Gur, R. C., & Saucy, M. C. (1978). Emotions are
expressed more intensely on the left side of the face. Science,
202, 434-436.
Safer, M. A. (1981). Sex and hemisphere differences in access to
codes for processing emotional expressions and faces. Journal of
Experimental Psychology: General, 110, 86-100.
Safer, M. A., & Leventhal, M. (1977). Ear differences in evaluating
tones of voices and verbal content. Journal of Experimental
Psychology: Human Perception and Performance, _3, 75-82.
Sauguet, J., Benton, A., & Hecaen, H. (1971). Disturbances of the
body schema in relation to language impairment and hemispheric
locus of lesion. Journal of Neurology, Neurosurgery and
Psychiatry, 34, 496-501.
Schacter, S. (1957). Pain, fear and anger in hypertensive and
normotensives. Psychosomatic Medicine, 19, 17-29.
Schacter, S. (1964). The interaction of cognitive and physiological
determinants of emotional state. Advances in Experimental Social
Psychology, J_, 49-80.
Schacter, S. (1970). The interaction of cognition and physiological
determinants of emotional state. In: J. Berkowitz (Ed.),
Advances in experimental social psychology (Vol. 1). New York:
Academic Press.
Schacter, S., & Singer, J. (1962). Cognitive, social and
physiological determinants of emotional state. Psychological
Review, 69, 379-399.
Schneider, S. J. (1983). Multiple measures of hemispheric dysfunction
in schizophrenia and depression. Psychological Medicine, 13,
287-297.
Schwartz, G. E., Ahern, G. L., & Brown, S. L. (1979). Lateralized
facial muscle response to positive and negative emotional
stimuli. Psychophysiology 16, 561 -57 3 -
Schwartz, G. E., Davidson, R. J., & Maier, F. (1975). Right
hemisphere lateralization for emotion in human brain:
Interactions with cognition. Science, 190, 286-288.


40
and on the status of the organism's attempt to exert control over its
environment or itself.
Kinsbourne provides a very interesting model of emotional
processing which is based, to some extent, on the approach-avoidance
model of Fox and Davidson and further elaborated to include an
anterior-posterior dimension. However, less evidence exists to
support or refute his claims.
Tuckers Model
Tucker (1981) has also suggested a neuropsychological model of
emotional processing based on lateralized neuroanatomical systems.
These systems control (a) tonic activation and motor readiness and (b)
phasic arousal responses to perceptual input. In contrast to notions
of reciprocal inhibition of the hemispheres, Tucker invokes mechanisms
of subcortical release of lateralized arousal systems to account for
left-negative versus right-positive valence findings.
According to Tucker, the left hemisphere is specialized for
activation and complex motor operations. Support for this is provided
by the preponderance of right hand motor dominance in the general
population as well as observed deficits in both right and left hand
production of learned, skilled motor movements (apraxias) following
left hemisphere lesions (Geschwind, 1975). The presumed neurochemical
substrate for this specialization is the dopamine system which several
investigations suggest is predominantly a left lateralized system
(Click, Meibnach, Cox, & Maayani, 1979; Wagner et al., 1983). Tucker
argues that activation operates in a tonic fashion to increase
informational redundancy. This view is supported by the observation


74
Table 5
Summary of Analysis of Variance: Experiment II, Heart Rate
Deceleration
Source
df
SS
F
P
Sex
1 ,
28
146.4789
5.16
*
Visual Field
1.
28
18.0850
3-33
-
Visual Field x Sex
1 ,
28
42.7167
7.86
*#
Stimulus Type
4,
1 12
89.6031
4.96
#*
Stimulus Type x Sex
4,
112
60.4664
3-35
**
Visual Field x Stimulus Type
4,
1 12
7.7819
.47
-
Visual Field x Stimulus Type x Sex
4,
112
12.6660
.77
-
*p < .05. ** p < .01.


105
effectiveness of this procedure in demonstrating differential and
lateralized heart rate responses in female and male Ss, respectively.
Conclusions
Findings from Experiment I, in which RTs were made to midline
neutral stimuli that were preceded by lateralized stimuli of different
emotional valences, failed to support any of the laterality models of
emotion. No overall superiority for RTs to LVF (right hemisphere)
versus RVF (left hemisphere) trials was found. In addition, no
evidence was found for hemispheric specific emotional valence
effects. Likewise, no evidence was present for the view of the
hemispheric differences in preparatory versus nonpreparatory
emotions. Rather, faster RTs were observed for male subjects when WS
were presented to RVF (left hemisphere). This finding is not
consistent with any of the proposed hemispheric models of emotional
processing. The most plausible interpretation of these findings is
that subjects were required to make a left-right decision based on the
emotional/nonemotional nature of the warning stimulus and that this
left hemisphere mediated task requirement accounts for the faster RTs
of male subjects to RVF (left hemisphere) presentations of WS.
Findings from Experiment II, in which measures of autonomic
arousal (HR and SCR) were obtained to laterally presented
emotional/nonemotional stimuli, were more congruent with models of
hemispheric differences in processing emotional stimuli. Males showed
lateralized effects of HR arousal responses which supports greater
right hemisphere involvement in production of arousal responses.


60
£
c
o
o
03
cc
Happy Angry Fearful Disgusting Neutral
Emotion
Figure 1. Experiment I--Reaction Time Analysis, Sex Main Effect,
Figure 2. Experiment I--Reaction Time Analysis, Sex x Visual Field
Interaction.


38
more interest expressions and greater relative left-sided EEG activity
following sucrose administrations compared with citric acid
administrations in infants and argued that these findings provide
support for left hemisphere superiority in processing of positive
affect. They further proposed that these emotions are under
unilateral hemispheric control since little functional interconnection
between the hemispheres exists at birth.
Through the course of development, changes in interhemispheric
communication are proposed as the necessary substrate for emergence of
fear and sadness in the emotional repertoire. In support of this, the
authors cite evidence that the onset of locomotion, a behavior
associated with commissural transfer, is tightly coupled to the
emergence of fear (Bayley, 1963; 1969; Rader, Bausano, & Richards,
1980). In addition, the authors argue that the expression of sadness
is often associated with alternating sequences of approach and
avoidance, again implicating a critical role for interhemispheric
communication (Ainsworth, Blehar, Waters, & Wall, 1978; Izard,
1977). The capacity to inhibit negative affective responses, which
emerge during the second year, are also presumed to be linked to the
functional integrity of the commissural system. In addition, these
authors propose that the left hemisphere normally exerts an inhibitory
influence on the right hemisphere through transcallosal connections,
resulting in attenuation of negative affect in the normal state.
Kinsbournes Model
Kinsbourne and Bemporad (1984) proposed a separate model also
based on the behavioral dichotomy that he terms action-approach and


112
Carroll, D., Baker, J., & Preston, M. (1979). Individual differences
in visual imaging and the voluntary control of heart rate.
British Journal of Psychology, 70, 39-49.
Charlesworth, W. 3. (1964). Instigation and maintainence of curiosity
behavior as a function of surprise versus novel and familiar
stimuli. Child Development, 35, 1169-1186.
Cicone, M., Wapner, W., & Gardner, H. (1980). Sensitivity to
emotional expression and situations in organic patients. Cortex,
J_6, 1 45-158.
Cimino, C., Verfaellie, M., Bowers, D., & Heilman, K. (1988).
Emotional autobiographical memory in right hemisphere damaged
patients. Unpublished manuscript.
Coulbourn, C. J., & Lishman, W. A. (1974;. Lateralization of function
and psychiatric illness. A left hemisphere deficit? In: J.
Gruzelier & P. Flor-Henry (Eds.), Hemispheric asymmetries of
function and psychopathology. Amsterdam: Elseview.
Cummings, J. L., & Mendez, M. P. (1984). Secondary mania with focal
cerebrovascular lesions. American Journal of Psychiatry, 141,
1084-1087.
Davidson, R. J., Horowitz, M. E., Schwartz, G. E., & Goodman, D. M.
(1981). Lateral differences in the latency between finger
tapping and the heart beat. Psychophysiology, 18, 36-41.
Davidson, R. J., & Schwartz, G. E. (1976). Patterns of cerebral
lateralization during cardiac biofeedback versus the self
regulation of emotion: Sex differences. Psychophysiology, 13.
62-68.
Davidson, R. J., Schwartz, G. E., Saron, C., Bennett, J., & Golemena,
D. J. (1979). Frontal vs. parietal EEG asymmetry during positive
and negative affect. Psychophysiology, 16, 202-203.
Davies, J. B. (1978). The psychology of music. London: Hutchinson.
DeKosky, S., Heilman, K. M., Bowers, D., & Valenstein, E. (1980).
Recognition and discrimination of emotional faces and pictures.
Brain and Language, 206-214.
d'Elia, G., & Perris, C. (1973). Cerebral functional dominance and
depression: An analysis of EEG amplitude in depressed
patients. Acta Psychiatrica Scandinavica, 49, 191-197.


54
Eye movements were continuously monitored by electro-oculography
(EOG) to ensure maintained fixation and lateralized presentation of
slides. The EOG signal was detected by Beckman Ag/AgCl miniature
electrodes attached at the temporal canthus of the left and right
eye. The EOG signal was filtered and fed into a DC amplifier and
recorded on a Grass Model 78B polygraph. Event marker input from
IBM-PC microcomputer was also fed into the polygraph recording to
identify occurrence of slide presentation and to facilitate subsequent
identification of eye movements during this interval.
Procedure
Subjects were seated 91 cm from the projection screen with left
and right hands placed on left and right sided keys, respectively.
The task was a choice reaction time task. Subjects viewed a laterally
presented warning stimulus (neutral or emotional scene) of 500 msec.
This was followed 500, 1000, or 1500 msec later by a centrally
presented neutral, imperative stimulus (red LED) with interstimulus
interval randomly varied across trials. Half of the Ss were
instructed to release the left key following onset of the LED if the
preceding warning stimulus was a neutral scene or the right key if the
warning stimulus was an emotional scene. The remaining Ss were
instructed to release the left key following onset of the LED if the
preceding warning stimulus was an emotional scene and the right key if
the preceding warning stimulus was a neutral scene. Half-way through
the experiment, hand order was reversed for all subjects. Subjects
received a total of 192 trials. Response hand and visual field of


support and friendship and for his honest and direct means of pushing
me to grow, personally and professionally. I thank Mieke Verfaellie
and Karen Froming for their solid support and friendship over the
years. With love and deepest appreciation, I also thank my husband,
Pat, for his continued encouragement and enduring love and support.


95
the emotional/nonemotional content of the stimuli and this right
hemisphere dominance is not specific to emotional stimuli.
Critical Issues
Several issues are raised by these findings. The first concerns
why there were no differences in arousal responses to emotional versus
neutral stimuli. Other investigators have found that emotional
stimuli induced greater arousal responses than do neutral stimuli
(Hare & 31evings, 1975; Klorman et al., 1977; Klorman & Ryan, 1980).
One possibility accounting for the lack of differences in arousal
responses to emotional versus neutral stimuli is that the emotional
stimuli exerted some sort of "priming" effect, such that even the
neutral stimuli were treated as "emotional." Kinsbourne (1970; 1973)
has suggested that adoption of a cognitive set can serve to
asymmetrically arouse or prime the hemispheres for processing. It may
have been the case in this study that emotional stimuli exerted a
priming effect in which neutral stimuli were treated as emotional
stimuli.
A second issue concerns why the laterality effect in HR responses
were observed only in male Ss. In a major review of the literature on
sex differences, McGlone (1980) concludes that overall males possess a
significantly greater degree of observed asymmetry relative to female
subjects. While this conclusion has been criticized on a number of
grounds, the weight of the evidence does suggest that, at least for
language functions, the hemispheric representation of males may differ
from that of females. That is, on measures of severity of aphasia and


Briggs, G. G., & Nebes, R. D. (1975). Patterns of hand preferences in
a student population. Cortex, 11, 230-238.
Brody, N., Goodman, S. E., Halm, E., Krinzman, S., 4 Sebrechts, M. M.
(1987). Lateralized affective priming of lateralized affectively
valued target words. Neuropsychologia, 25, 935-946.
Brownell, H. H., Potter, H. H., Michelow, D., 4 Gardner, H. (1984).
Sensitivity to lexical denotation and connotation in brain
damaged patients: A double dissociation? Brain and Language,
22, 253-265.
Bruder, G. E. (1983). Cerebral laterality and psychopathology: A
review of dichotic listening studies. Schizophrenia Bulletin, _9,
143-151.
Brumback, R. A., 4 Staton, R. D. (1981). Depression-induced
neurological dysfunction. New England Journal of Medicine, 304,
642.
Bryden, P. (1978). Strategy effects in hemispheric asymmetry. In:
G. Underwood (Ed.), Strategies of information processing.
New York: Academic Press.
Bryden, P., 4 AlLard, F. (1976). Visual hemifield differences depend
on typeface. Brain and Language, _3, 191-200.
Bryden, P., 4 Ley, R. G. (1983). Right hemisphere involvement in the
perception and expression of emotion in normal humans. In:
K. M. HeiLman (Ed.), Neuropsychology of human emotion.
New York: Guillford Press.
Bryden, M. P., Ley, R. G., 4 Sugarman, J. H. (1982). A left ear
advantage for identifying the emotional quality of tonal
sequences. Neuropsychologia, 20, 8387.
Buchtei, H. A., Campari, F., DeRisio, C., 4 Rota, R. (1978).
Hemispheric differences in discriminative reaction time to facial
expressions. Italian Journal of Psychology, jj, 159-169.
Buck, R., 4 Duffy, R. J. (1980). Nonverbal communication of affect in
brain damaged patients. Cortex, 16, 351-362.
Cannon, W. B. (1927). The James-Lange theory of emotions: A critical
examination and an alternative theory. American Journal of
Psychology, 34, 106-124.
Carmon, A., 4 Nachson, I. (1973). Ear asymmetry in perception of
emotional nonverbal stimuli. Acta Psychologica, 37, 351-357.


107
the lack of emotion specific effects in HR responses together with the
possible emotion specific effects in SCRs. The other consideration is
that the SCR findings represent only a trend and may not truly effect
an emotion specific hemispheric effect.
The findings of female Ss are not consistent with any hemispheric
emotionality models. In contrast to laterality effects observed for
male Ss, female Ss appeared to show a qualitatively different pattern
of HR responding which differentiated their performance from male
Ss. This was revealed in a significant Sex x Stimulus Type
interaction for analysis of HR deceleratory responses and HR
acceleratory responses. This pattern of findings demonstrated that
female but not male Ss snowed differential deceleratory as well as
acceleratory HR responses across the different conditions of stimulus
type with significantly greater acceleratory responses to angry and
fearful slides compared iith neutral slides and significantly greater
deceleratory responses to happy, disgusting, and neutral slides
relative to fearful slides.
These findings also provide further support for the view that
different patterns of autonomic arousal may be associated with
different types of emotional states. In addition, they also suggest
that possible differences in imagery ability across male and female Ss
may have, in part, mediated this differential pattern of responding.
These findings point to the relative importance of considering
both sex and imagery ability of Ss in further investigations of
emotional processing and autonomic responding. Sex differences in
degree of lateral asymmetry of arousal responses suggest that the


44
response to visual stimuli initially projected to the right
hemisphere.
Heilman and Van Den Abel (1979) have also suggested a right
hemisphere superiority for activation. Using a neutral warning
stimulus paradigm, these authors reported that warning stimuli
projected to the right hemisphere reduced reaction times of the right
hand more than warning stimuli projected to the left hemisphere. In
addition, warning stimuli projected to the right hemisphere reduced
reaction times of the right hand more than warning stimuli projected
to the left hemisphere reduced reaction times of the left hand. Based
on these findings, it can be seen that warning stimuli projected to
the right hemisphere reduced reaction times of both hands to a greater
extent than left hemisphere presentations, suggesting that the right
hemisphere was better able to activate responses in both hands
relative to the left hemisphere.
In addition, Verfaellie, Bowers, and Heilman (1987) reported a
study of neurologically intact subjects which provides some support
for the right hemisphere dominance of activation. By manipulating
preliminary intentional warning cues (which hand to use in
responding), they found faster left hand versus right hand responses
suggesting that the left hand (right hemisphere) was better able to
benefit from this preparatory information than the right hand (left
hemisphere). In support of this, Verfaellie and Heilman (1987) also
report the performance of two patients with right and left
supplementary motor area (SMA) damage on this paradigm. They report
that the patient with left SMA damage (intact right hemisphere) was


BPM Irom Baseline BPM from Baseline
82
Males
Post-Stimulus Seconds
Males
Post-Stimulus Seconds
Males
Post-Stimulus Seconds

c
Q
U1
01
CQ
E
o
2
CL
CD
Males
C
*3
a
CQ
e
o
2
Q-
CQ
Males
Post-Stimulus Seconds
Figure 17. Experiment II--Second x Second Post-Stimulus Heart Rate
Changes in Male Subjects for Each Visual Field x Stimulus
Type Condition.


Skin Conductance Response
87
0.14-
0.12-
0.10-
0.08 -
0.06-
Figure 19.
Trend
Emotion
Experiment II--Skin Conductance Response Analysis, Sex x
Visual Field x Stimulus Type Trend.


77
Stimulus Type interaction [F (4, 112) = 3-35, £ = .0125)]. Simple
effects testing revealed no significant effects across Stimulus Type
for males [£ (4, 56) = 1.86, £ = .130)]. However, for female subjects
this Stimulus Type effect was significant [F (4, 56) = 5-32, £ =
.001)]. Duncan's post-hoc comparisons revealed that deceleratory HR
responses to happy = -8.306), disgusting (>4 = -8.3^5) and neutral
(M = -8.638) were all significantly greater than deceleratory HR
responses to angry (_M = -6.120) slides at £ < .05. These
relationships are depicted in Figure 14.
Maximum acceleratory responses
A repeated measures ANOVA was performed with Sex (male, female)
as the between subjects factor and Visual Field (left, right) and
Stimulus Type (happy, angry, fearful, disgusting, neutral) as the
within subjects factors. A summary of the results of this analysis
are depicted in Table 6. Results revealed a significant interaction
of Stimulus Type x Sex [F (4, 112) = 3-48, £ = .0102)], depicted in
Figure 15. Simple effects testing revealed no significant effect of
Stimulus Type for males [F_ (4, 56) = 1.86, £ = .130)]. For female Ss,
however, this effect was significant [F (1, 14) = 5.32, £ = .001)].
Duncan's post-hoc comparisons revealed that angry (14 = 3-^73) and
fearful (_M = 3-888) slides elicited significantly greater HR
accelerations than neutral (_M = 2.158) slides at p < .05.
This analysis also revealed a trend for an interaction between
Visual Field x Sex [F (1, 28) = 3-36, £ = .0776)], with a pattern of
greater acceleratory HR responses for males in LVF (M = 2.531) trials


56
sternum. Electrode sites were prepared by mild abrasion of the skin
with Hewlett Packard Redux paste. Electrodes were fastened by the use
of adhesive collars and were filled with Hewlett Packard Jel Redux
cream as the electrolyte. The electrocardiogram (ECG) was amplified
by a Colbourn S75-03 high gain bioamplifier. This signal was input to
a Coibourn S75-38 Ban-Pass Biofilter with subsequent detection of the
R-wave component which interrupted the computer to provide inteibeat
intervals. A-D conversion was accomplished by Colbourn R65-17 Data
Translation Board and signal was downloaded to an IBM-PC interfaced
with data acquisition modules.
Skin conductance was recorded with Met-Associates electrodes
placed on the thenar and hypothenar eminences of the left and right
hands. Electrode sites were wiped clean with distilled water.
Electrodes were attached by the use of adhesive collars filled with KY
jelly. The analog SC signal was fed into a Colbourn Model S71-22 Skin
Conductance Module. This signal was digitized by a Colbourn R65-17
Data Translation Board and the signal was downloaded to an IBM-PC
microcomputer interfaced with the data acquisition modules.
Respiration depth was also recorded to detect and subsequently
exclude those trials in which unusually large inhalations or
exhalations may have confounded HR or SCR. Respiration depth was
measured by means of a Colbourn Model T41 -91 Aneroid Chest Bellows,
and processed by a Colbourn S72-25 Module. A-D conversion and
computer interface utilized the same equipment as HR and SCR measures.


97
Similarly, Hantas, Katkin, and Reed (1984) using male subjects
only demonstrated that right hemisphere preferent individuals (as
indexed by conjugate lateral eye movements) performed significantly
better than left hemisphere preferent individuals on a task of heart
beat detection. Furthermore, these findings were significant both
before and after heart rate discrimination training. Montgomery and
Jones (1984) reported similar findings in male subjects, with right
hemisphere preferent subjects performing significantly better on a
heart beat perception task relative to left hemisphere preferent
subjects. In addition, they also found higher emotionality scores for
right hemisphere preferent individuals relative to left hemisphere
preferent individuals.
It is of interest to note that these studies which find
significant laterality effects on tasks of heart beat perception and
detection have used primarily male subjects. This occurrence may, in
part account, for significant laterality effects observed on these
tasks as male subjects often reveal a significantly greater degree of
cerebral lateralization than female subjects (McGlone, 1980).
A consistent finding in the cardiac awareness literature is that
male subjects are superior to female subjects in detecting heart beats
and heart rate across a variety of psychophysiological paradigms
(Pennebaker & Hoover, 1984; Whitehead, Drescher, Heiman, & Blackwell,
1977). More recent research (Rouse, Jones, 4 Jones, 1988), however,
has suggested that such sex differences in cardiac awareness are
accounted for primarily by body fat composition {% fat) and general
fitness differences across male and female subjects. By controlling


125
Taylor, D. C. (1972). Mental status and temporal lobe epilepsy: A
correlative account of 100 patients treated surgically.
Epilepsia, 13, 727-765.
Terzian, H. (1964). Behavioral and EEC effects of intracarotid sodium
amytal injections. Acta Neurochirurgic, 12, 230-240.
Toone, B. K., Cooke, E., & Lader, M. H. (1981). Electrodermal
activity in the affective disorders and schizophrenia.
Psychological Medicine, 11, 497-508.
Tranel, D., Damasio, A., & Damasio, H. (1988). Intact recognition of
facial expression, gender and age in patients with impaired
recognition of face identity. Neurology, 38, 690-696.
Tucker, D. M. (1981). Lateral brain function, emotion and
conceptualization. Psychological Bulletin, 89, 19-46.
Tucker, D. M., Roth, R. S., Arneson, B. A., & Buckingham, V.
(1977a). Right hemisphere activation during stress.
Neuropsychologia, 15, 697-700.
Tucker, D. M., Stenslie, C. E., Roth, R. S., & Shearer, S. L.
(1981). Right frontal lobe activation and right hemisphere
performance decrement during a depressed mood. Archives of
General Psychiatry, 38, 169-174.
Tucker, D. M., Watson, R. T., & Heilman, K. H. (1977b). Affective
discrimination and evocation in patients with right parietal
disease. Neurology, 17, 947-950.
Verfaellie, M., Bowers, D., Heilman, K. M. (1987). Hemispheric
asymmetries in mediating intention, but not selective
attention. Neuropsychologia, 26, 521-531.
Verfaellie, M., & Heilman, K. M. (1987). Response preparation and
response inhibition after lesions of the medial frontal lobe.
Archives of Neurology, 44, 1265-1271.
Vrana, S. R., Cuthbert, B., & Lang, P. (1986). Fear imagery and text
processing. Psychophysiology, 23 247-253.
Wagner, H. N., Burns, D. H., Dannal3, R. F., Wondg, D. F., Langstrom,
B., Duelfer, T., Frost, J. J., Ravert, H. T., Links, J. M.,
Rosenbloom, S. B., Lucas, S. E., Kramer, A. V., & Kuhlar, M. J.
(1983). Imaging dopamine receptors in the human brain by
positron tomography. Science, 221, 1264-1266.
Walker, B., & Sandman, C. (1982). Visual evoked potentials change as
heart rate and carotid pressure change. Psychophysiology, 19,
520-527.


recognition of emotional faces. The mood-state dependent effect
states that stimuli will be better recalled when subject's mood at
encoding and retrieval are congruent than when they are disparate
(Bower, 1981). Using a mood induction procedure, Gage and Safer
demonstrated that recognition of faces initially presented in a
discrepant mood was significantly worse when presented to the LVF
(right hemisphere) than when presented to the RVF (left hemisphere).
Based on these observations, the authors suggest that the right
hemisphere stores the subject's mood as an integral part of the memory
representation to a greater extent than the left hemisphere.
Although it is well known that the left hemisphere is specialized
for mediating speech and linguistic stimuli in the vast majority of
right handed adults, the right hemisphere also appears to have some
reduced language capacity (Geschwind, 1969; Papanicolaou, Moore,
Deutsch, Levin, & Eisenberg, 1988). Several recent studies have
suggested that the right hemisphere may better process emotional
versus nonemotional verbal stimuli, at least at the single word
level. For example, Graves and coworkers (Graves, Landis, 4
Goodglass, 1980) found that aphasic patients with alexia due to left
hemisphere lesions could read emotional words better than nonemotional
words. In a subsequent study, these same investigators (Graves,
Landis, & Goodglass, 1981) found that neurologically intact males
subjects better recognized emotional words than nonemotional words
when these stimuli were presented to the LVF (right hemisphere).
Other researchers (Brody, Goodman, Holm, Krinzman, & Sebrechts,
1987) have looked at the effects of lateralized affective priming


68
Table 3
Summary of Analysis of Variance: Experiment I, Percent Correct
Source
df
SS
F
P
Sex
1,
28
.8665
2.28
-
Visual Field
1,
28
.0181
.22
-
Visual Field x Sex
1,
28
.1683
2.01
-
Hand
1,
28
.0757
1.23
-
Hand x Sex
1,
28
.0203
33
-
Stimulus Type
4,
112
2.5844
7.60
#*
Stimulus Type x Sex
4,
112
.2799
.82
Visual Field x Hand
1,
28
.0009
.01
-
Visual Field x Hand x Sex
1 ,
28
.0004
.01
-
Visual Field x Stimulus Type
4,
112
.4369
2.29
-
Visual Field x Stimulus Type x Sex
4,
112
.1908
1 .00
-
Hand x Stimulus Type
4,
1 12
.4237
1.44
-
Hand x Stimulus Type x Sex
4,
112
.1018
.35
-
Visual Field x Hand x Stimulus Type
4,
1 12
.0793
.45
-
Visual Field x Hand x Stimulus Type
x Sex
4,
112
.1774
1.02
"
#*
p < .01.


36
Levenson, and Friesen (1983) have also reported heart rate increases
in response to production of emotional facial expressions of anger,
fear, and sadness and heart rate decreases in response to disgust,
surprise, and happiness.
In summary, there is some evidence to suggest that autonomic
arousal is not as uniform as once suggested. In fact, several studies
support the notion that different patterns of autonomic arousal may be
associated with different types of emotional states. In addition,
recent conceptualizations of heart rate arousal responses have
suggested that such changes may be linked to overt and/or covert
motoric responses. This view is also consistent with the bio-
informational theory proposed by Lang (1979). Lang proposes that
emotional imagery results in patterns of autonomic activity very
similar to those found in the actual emotional situation. This raises
the possibility that certain emotions, by virtue of their strong motor
components, may be more associated with heart rate acceleration while
others with less motor demands may be more associated with heart rate
deceleration.
Neuropsychological Models of Emotional Processing
Several different neuropsychological models of emotional
processing have been proposed to account for findings of
investigations in brain impaired and neurologically intact subjects.
The models of Fox and Davidson (1984), Kinsbourne (Kinsbourne &
Bemporad, 1984), Tucker (1981), and Heilman (Heilman et al. 1983;
Heilman, 1988, personal communication) will be briefly reviewed.


9
More recent investigations have challenged this latter
interpretation. Support for such a dissociation has recently been
provided in a study by Kolb and Taylor (1981). Their findings
revealed that patients with parietal excision were impaired on both
matching of facial affect and matching of facial identity. However,
these authors also report that patients with damage restricted to
right temporal and right frontal regions were more impaired on
processing of facial affect relative to processing of facial
identity. Similarly, a study by Freid, Mateu, Ojemann, Wohns, and
Fedio (1982) reported that neutral facial matching was disrupted by
stimulation of right parietal-occipital regions, whereas naming the
affect depicted on pictures of faces was disrupted by stimulation of
right posterior middle temporal gyrus.
In contrast, Cicone, Wapner, and Gardner (1980) found no
relationship between RHD patients' performance on an emotional
perception task and a facial identity task, suggesting that deficits
associated with RHD cannot be accounted for by facial recognition
deficits alone.
More recently, Bowers and Heilman (1984) reported a case which
demonstrated a dissociation between processing of affective and
nonaffective faces. Although this RHD patient was able to perform
well on neutral facial tasks and on same-different emotional faces
tasks, he was impairea on naming and comprehension of verbal labels
for facial expressions.
In a subsequent investigation, Bowers, Bauer, Cosiett, and
Heilman (1985) again addressed the question of whether defects shown


Percent Correct
71
Figure 9- Experiment I--Percent Correct Analysis, Stimulus Type Main
Effect.


20
facial photo composed entirely of the left or right half of the
face. These studies report that left sided composites of posed facial
expressions are judged as more intense than right sided composites
(Heller & Levy, 1981; Rubin & Rubin, 1980; Sackeim, Gur, & Saucy,
1978). This effect has also been reported for spontaneous expressions
as well (Dopson, Beckwith, Tucker, & Bullard-Bates, 1984).
Clinical reports have indicated that spontaneous emotional facial
expressions are less likely to occur in RHD patients compared to LHD
patients (Borod, Koff, Perliman, Lorch, & Nicholas, 1986; Buck &
Duffy, 1980; Ross & Mesulam, 1979). Kolb and Milner (1981), however,
were unable to find differences between RHD and LHD patients in
spontaneous facial expressions and movements (only some of which were
emotional). They did find, however, differences with respect to the
anterior versus posterior distribution of the lesion, with anterior
lesions resulting in less spontaneous movements than more posterior
lesions. Borod et al. (1986) recently reported the only systematic
investigation, to date, of posed and spontaneous facial expressions in
brain impaired patients. They found that RHD patients produced fewer
posed as well as spontaneous emotional facial expressions than did LHD
patients.
Mood and Affect Studies
A second group of studies has also examined disorders of mood and
affect associated with right hemisphere function. These
investigations have studied mood and affective changes in patients
with known hemispheric pathology. Patient groups have included


55
presentation were randomized and counterbalanced across stimulus
type. Subjects were given 10 practice trials prior to the experiment.
Experiment II: Psychophysiological Responses to
Laterally Presented Emotional Material
Stimuli
The stimuli were identical to those used in Experiment I. They
included 24 neutral and 96 emotional scenes [24 of each of four
differing valences (happy, angry, fearful, disgusting)].
Apparatus
Slides were projected onto a 40x35-cm Kodak milk-glass, rear view
projection screen. A 5-mm diameter adhesive circle was placed at the
center of the screen to serve as a central fixation point. Slides
were projected at 5 degrees of visual angle lateral to the central
fixation point. Lafayette Model #43016 shutters were attached to the
projector lens to maximum the rise and fall time of slide
presentation. The timing and presentation of stimuli were
accomplished by IBM-PC microcomputer interfaced with BRS logic. Eye
movements were continuously monitored in the same fashion as
Experiment I. The room was again dimly lit to avoid the effects of
visual startle during stimulus presentation.
Psychophysiological measures (HR, SCR, and respiration depth)
were recorded for the 3 seconds prior to stimulus onset and 8 seconds
of stimulus presentation. Heart rate responses were recorded by two
Beckman Ag/AgCl electrodes attached to the right and left lateral
margins of the chest. A third electrode was attached to the


113
d'Elia, G., & Perris, C. (1974). Cerebral functional dominance and
memory functions: An analysis of EEG integrated amplitude in
depressive psychotics. Acta Psychiatrica Scandinavica
(Supplement), 255, 143-157.
Delp, M. J., & Sackeim, H. A. (1987). Effects of mood on lacrimal
flow: Sex differences and asymmetry. Psychophysiology, 24,
550-556.
Denny-Brown, D., Meyer, J. S., & Horenstein, S. (1952). The
significance of perceptual rivalry resulting from parietal
Lesions. Brain, 75, 434-471 .
DeWitt, G. W. (1978). Laterality, personality and the perception of
emotional stimuli. Dissertation Abstracts International,
38(8-B), 3850.
Dimond, S. J., Farrington, L., 4 Johnson, P. (1976). Differing
emotional response from right and left hemispheres. Nature, 261,
690-692.
Dopson, W. G., Beckwith, B. F., Tucker, D. M., & Bullard-Bates, P. C.
(1984). Asymmetry of facial expression in spontaneous emotion.
Cortex, 20, 243-252.
Ehrlichman, H., 4 Weinberger, A. (1978). Lateral eye movements and
hemispheric asymmetry: A critical review. Psychological
Bulletin, 85, 1080-1101.
Ekman, P. (1972). Universal and cultural differences in facial
expressions of emotion. In J. K. Cole (Ed.), Nebraska symposium
on motivation. Lincoln: University of Nebraska Press.
Ekman, P., Hager, J. C., & Friesen, W. V. (1981). The asymmetry of
emotional and deliberate facial actions. Psychophysiology, 18,
101-106.
Ekman, P., Levenson, R. W., & Friesen, W. V. (1983). Autonomic
nervous system activity distinguishes among emotions. Science,
221, 1208-1210.
Ellinwood, E. H. (1967). Amphetamine psychosis: 1. Description of
the individuals and process. Journal of Nervous and Mental
Disease, 144, 273-283.
Fehr, F. S., & Stern, J. A. (1970). Peripheral psychological
variables and emotion: The James-Lange theory revisited.
Psychological Bulletin, 74, 411-424.
Flor-Henry, P. (1969). Psychosis and temporal lobe epilepsy: A
controlled investigation. Epilepsia, 10, 363-395.


29
that unpleasant films were rated as more unpleasant when presented to
the LVF than when presented to the RVF. More recently, Reuter-Lorenz,
Givis, and Moskovitch (1983) have also reported shorter reaction times
to RVF presentations of happy faces and LVF presentations of sad
faces. These findings are congruent with proposed left hemisphere
positive affect and right hemisphere-negative affect distinctions.
Although numerous studies are consistent with the hypothesis that
the right hemisphere preferentially mediates negative emotions and the
left hemisphere mediates positive emotions, an equal number of studies
find no support for this hemispheric valence hypothesis. Rather both
positive and negative stimuli seem to be preferentially mediated by
the right hemisphere. To account for these discrepant views on the
hemispheric processing of positive versus negative stimuli, Bryden and
Ley (1983) have argued that methodological differences across studies
might contribute to the discrepant findings. For example, Reuter-
Lorenz and Davidson (1981) report faster reaction times for LVF
presentations of sad faces and RVF presentations of happy faces. In
this study, subjects were required to identify which of two laterally
presented faces (one neutral and one emotional) showed an affective
expression. In investigations which show a significant overall right
hemisphere effect, the task is quite different. In these studies, the
task is usually to determine whether a laterally presented face is the
same or different than a centrally presented face. It is possible
that these differences in task requirements may, in some way, account
for the findings.


98
for level of body fat, Rouse et al. (1988) did not find the gender
effect previously documented in the literature. Furthermore, sex
differences were found only when level of body fat differed between
males and females.
In the present investigation, a main effect of Sex was observed
in Experiment II for maximum HR deceleratory responses. More
specifically, male subjects had significantly greater HR decelerations
than female subjects. It may be the case that body fat composition
and general fitness also plays some role in magnitude of cardiac
decelerations although this relationship is unclear at present.
To briefly summarize, findings from Experiment II provide support
for the greater role of the right hemisphere in mediating arousal
responses, at least for male subjects. That is, males showed greater
HR deceleratory and acceleratory responses to stimuli when they were
presented to the LVF (right hemisphere) tnan to the RVF (left
hemisphere). This concerns the basis for these sex differences.
The evolutionary significance of lateralized heart rate responses
in male subjects can be viewed from the more general perspective of
the evolutionary significance of greater cerebral lateralization in
males. Flor-Henry (cited in McGlone, 1930) provides an interesting
argument in this regard. He notes that dopaminergic pathways in
rodents are known to be asymmetrical and related to directional
preferences and turning behaviors with the direction of these
behaviors contralateral to the hemisphere with greatest dopamine
concentrations (Glick, Jerussi, & Zimmer berg, 1977). Turning
behaviors in rodents, cats, and dogs are associated with fighting and


Reaction Time
62
(a)
(b)
Figure 3- Experiment I--Reaction Time Analysis, Sex x Stimulus Type
Interaction: (a) males only and (b) males and females.


4
was to identify the emotional tone of the speaker. Results
demonstrated that the RHD performed significantly worse on this task
than the LHD group.
In a subsequent study, Tucker, Watson, and Heilman (1977b)
investigated patient's ability to discriminate between different
affectively toned sentences. In this task, subjects were not required
to identify the affective tone but to discriminate between sentences
presented with the same or different affective tone. RHD patients,
again, performed more poorly than LHD patients, providing further
evidence for the right hemisphere's greater role in the processing of
emotional aspects of speech.
Weintraub, Mesulam, and Kramer (1981) have suggested that RHD
patients have difficulty in both emotional and nonemotional aspects of
prosodic speech. In their study, RHD patients had significantly
greater difficulty in distinguishing whether filtered sentences were
questions, commands, or statements. Based on their findings,
Weintraub et al. suggested that the poor performance of RHD patients
on emotional prosody tasks may be accounted for by a more general
deficit in the processing of prosodic information. No LHD group was
reported in this investigation.
In a subsequent study which addressed this question, HeiLman,
Bowers, Speedie, and Coslett (1984) investigated the ability of RHD
and LHD patients to comprehend filtered sentences which contained
either emotional (happy, sad, angry) or nonemotional (declarative,
imperative, or interrogative) prosody. Results demonstrated that both
RHD and LHD patients were impaired on the nonemotional prosody task


49
it was not possible to equate occurrence of positive versus negative
emotions.
Hypotheses and Predictions
According to the right hemisphere emotion model, the right
hemisphere plays a greater role than the left hemisphere in mediating
arousal/activation responses to emotional materials. If this model is
correct, then one would predict faster RT responses to a midline
neutral stimulus when it is preceded by an emotional warning stimulus
(WS) directed to the right hemisphere (LVF) than by an emotional WS
directed to the left hemisphere (RVF). Additionally, RTs should also
be faster when emotional WS versus nonemotional WS are directed to the
right hemisphere (LVF). These predictions were examined in Experiment
I. Similarly, one would also predict that autonomic responsivity to
emotional versus nonemotional stimuli should be greater when they are
directed to the right hemisphere (LVF) versus the left hemisphere
(RVF). These predictions were examined in Experiment II.
According to hemispheric valence models of emotional processing,
negative emotional stimuli are preferentially mediated by the right
hemisphere, and positive emotional stimuli are mediated by the left
hemisphere. If this hypothesis is correct, then one would predict
that negative emotional WS directed to the right hemisphere (LVF)
should result in faster RTS to a neutral midline stimulus than when
the negative WS is directed to the left hemisphere (RVF). Conversely,
positive emotional WS directed to the left hemisphere (RVF) snould
result in faster RTs than positive WS directed to the right hemisphere


65
Table 2
Summary of Analysis of Variance
i: Experiment
I,
Reaction
Time (minus
outliers)
Source
df
SS
F
£
Sex
1 ,
26
2.3201
2.17
-
Visual Field
1.
26
.0002
.03
-
Visual Field x Sex
1 .
26
.0332
5.20
#
Hand
1,
26
.0441
1.15
-
Hand x Sex
1 ,
26
.0060
.16
-
Stimulus Type
4,
104
.1194
1.96
-
Stimulus Type x Sex
4,
104
.1350
2.22
-
Visual Field x Hand
1.
26
.0019
.12
-
Visual Field x Hand x Sex
1 ,
26
.0000
.00
-
Visual Field x Stimulus Type
4,
1 04
.0137
.20
-
Visual Field x Stimulus Type x
Sex
4,
104
.0699
.76
-
Hand x Stimulus Type
4.
104
.1714
1.59
-
Hand x Stimulus Type x Sex
4,
104
.0357
33
-
Visual Field x Hand x Stimulus
Type
4,
104
.1514
1.86
-
Visual Field x Hand x Stimulus
Type
x Sex
4,
104
.0362
. 45
p < .05.


with real as well as intended somatic activity (as in the case of
imagery).
One additional question concerns the fact that findings from
Experiment II provided support for the greater role of the right
hemisphere in production of arousal responses but results from
Experiment I failed to provide support for a greater role of the right
hemisphere in production of arousal responses. Differences across the
two experiments may help to clarify some of this apparent
discrepancy. First, the most plausible explanation for these
discrepant findings is that in Experiment I subjects were required to
make a left-right decision based on the emotional/nonemotional nature
of the warning stimulus. As previously discussed, it is likely that
this left-right discrimination (a predominantly left hemisphere
ability) inherent in the task demands of Experiment I is related to
the finding of faster RTs for male Ss when stimuli were presented to
the RVF (left hemisphere).
Secondly, exposure durations in the two experiments differed
quite significantly. Experiment I utilized relatively short exposure
durations while Experiment II utilized quite lengthy exposure
durations. The overall complexity of the stimuli together with the
relatively short exposure durations used in Experiment I versus
Experiment II may have contributed to these divergent findings.
Similarly, the task demands of the two experiments were quite
different. Requirements inherent in Experiment II in which subjects
were asked to generate very personal and likely very meaningful
episodes from their own lives may have contributed to the overall


126
Watson, R. T.f Heilman, K. M., Cauthen, J. C., & King, F. A. (1973).
Neglect after cingulectomy. Neurology, 23, 10031007.
Webster, W. G. (1977). Territoriality and the evolution of brain
asymmetry. Annals of the New York Academy of Science, 299,
213-221 .
Weerts, T. C., & Roberts, R. (1976). The physiological effects of
imagining anger-provoking and fear-provoking scenes.
Psychophysiology, 13, 174.
Weingartner, H., & Silberman, E. K. (1982). Models of cognitive
impairment: Cognitive changes in depression. Psychopharmacology
Bulletin, J_8, 27-42.
Weintraub, S., Mesulam, M., & Kramer, L. (1981). Disturbance in
prosody: A right hemisphere contribution to language. Archives
of Neurology, 38, 742-744.
Weschler, A. F. (1973). The effect of organic brain disease on recall
of emotionally charged versus neutral narrative texts.
Neurology, 23, 130-135.
White, K., Sheehan, P. W., & Ashton, R. (1977). Imagery assessment:
A survey of self-report measures. Journal of Mental Imagery, J_,
145-170.
Whitehead, W. E., Drescher, V. M., Heiman, P., 4 Blackwell, B.
(1977). Relation of heart rate control to heart beat
perception. Biofeedback and Seif-Regulation, 2, 371-392.
Whitlock, F. A. (1982). Symptomatic affective disorders. New York:
Academic Press.
Winer, B. J. (1971). Statistical principles in experimental design
(2nd ed.). New York: McGraw-Hill.
Woods, D. J. (1977). Conjugate lateral eye movement, repression-
sensitization, and emotional style: Sex interactions. Journal
of Clinical Psychology, 33, 839-841.
Yokoyama, K., Jennings, R., Ackles, P., Hood, P., & Boiler, F.
(1987). Lack of heart rate changes during an attention demanding
task after right hemisphere lesions. Neurology, 37, 748-755.


32
assumption that arousal and cognition may play an important role in
emotion.
Early investigations in neurophysiology laid much of the
groundwork for our current knowledge of physiological arousal. In
1933 Berger reported that the electroencephalographic (EEG) pattern
during behavioral arousal displayed decreases in amplitude and
increases in frequency. This "electroencephalographic
desynchronization" observed during periods of behavioral arousal was
also later reported to occur during emotional states (Lindsley, 1970).
Studies have also identified critical neuroanatomic structures
involved in the elicitation of arousal responses. Stimulation in
nonspecific thalamic nuclei or the mesencephalic reticular formation
(MRF) result in behavioral manifestations of arousal as well as EEG
desynchronization (Moruzzi & Magoun, 19^9). Similarly, stimulation of
frontal or temporoparietal cortex activates the MRF (French,
Hernandez-Peon, & Livingston, 1955) and elicits an arousal response
(Sequndo, Nasuet, & Buser, 1955). Another pathway by which cortical
stimulation can produce arousal is via limbic system projections to
cortex and MRF (Heilman & Valenstein, 1972; Watson, Heilman, Cauthen,
& King, 1973).
This conceptualization of reciprocal connections between MRF
system and cortical regions is central to a model of arousal proposed
by Sokolov (1963). Sokolov described a specific pattern of
physiological changes which occurred in response to novel or
"significant" stimuli. This specific pattern of physiological changes
was termed the orientating response (OR). At the behavioral level,


81
relative to RVF (M = 2.027) trials. This pattern of findings is
depicted in Figure 16.
Second by second responses
Second by second post-stimulus HR changes for male and female Ss
are displayed for each Visual Field x Stimulus Type condition in
Figures 17 and 18, respectively. These values represent the average
change from baseline for each of the eight po3t-stimulus seconds.
Post-stimulus HR changes from baseline for both male and female Ss
suggest that, overall, Ss showed primarily HR deceleratory changes
throughout the presentation of the stimuli. This may be accounted
for, in part, by task factors which required perceptual intake
throughout the presentation of the stimulus. Perceptual intake has
been associated with HR deceleration as suggested by Lacey (1967).
It is also of note that post-stimulus HR changes from baseline do
not reflect occurrence of acceleratory HR responses (i .e., positive HR
change from baseline). It may be that the emotional stimuli
themselves were not inherently "strong enough to elicit acceleratory
HR responses. However, findings from the analysis of maximum
acceleratory HR responses do indicate that, at some point during the 8
post-stimulus seconds, Ss are experiencing HR acceleration. The most
plausible explanation for the lack of post-stimulus HR accelerations
in graphic representation of the 8 post-stimulus seconds is that these
acceleratory responses are occurring at different points in time
across different trials and possibly across different S3. That is,
accelerations may be "averaged out" by decelerations occurring at the
same time on separate but like trials (i.e., same VF x Stimulus Type


emotional valence of the stimuli. Female subjects showed autonomic
patterning effects across emotional categories, a finding which may be
accounted for by imagery differences across male and female
subjects. Taken together, these findings point to the relative
importance of considering both sex and imagery ability of subjects in
future investigations of emotional processing and autonomic
responding.
vi ii


PSYCHOPHYSIOLOGICAL AND REACTION TIME RESPONSES
TO LATERALLY PRESENTED EMOTIONAL STIMULI
BY
CYNTHIA RODRIGUES CIMINO
A DISSERTATION PRESENTED TO THE GRADUATE SCHOOL
OF THE UNIVERSITY OF FLORIDA IN PARTIAL FULFILLMENT
OF THE REQUIREMENTS FOR THE DEGREE OF
DOCTOR OF PHILOSOPHY
UNIVERSITY OF FLORIDA
1988

ACKNOWLEDGMENTS
I would like to extend my gratitude to those who provided the
encouragement and support that enabled the completion of this
project. First, I am grateful to my dissertation committee. I would
like to thank my chair, Dr. Dawn Bowers, for her time, hard work, and
patience over the years; for her skill at making me really think; and
especially for her belief in my ability. I am also grateful to
Dr. Kenneth Heilman for providing material and moral support and for
sharing his knowledge and never-ending enthusiasm, creativity, and
wonderment. I thank Dr. Rus Bauer for his advice on
psychophysiological technique and statistical method and for his
infrequent but apt clinical interpretations. I would also like to
tnank Dr. Eileen Fennell for ner discerning comments on methodology,
her ground-rooted advice on professional development, and her
kindness. I am also grateful to Dr. Ed Valenstein for his support in
my defense and for his thought-provoking question at the end of my
dissertation copy (humble as always). Finally, I am grateful to
Dr. Hugh Davis for so many things, but especially for his guidance in
development of my clinical abilities, his warmth and playfulness, and
his love of language and verbal tapestries.
I owe a debt of gratitude to Cindy Zimmerman for her skill in
organizing the typing and completion of the manuscript. My
appreciation also goes to Dr. Roger Blashfield for his unconditional

support and friendship and for his honest and direct means of pushing
me to grow, personally and professionally. I thank Mieke Verfaellie
and Karen Froming for their solid support and friendship over the
years. With love and deepest appreciation, I also thank my husband,
Pat, for his continued encouragement and enduring love and support.

TABLE OF CONTENTS
Page
ACKNOWLEDGMENTS
ABSTRACT vi
INTRODUCTION 1
Right Hemisphere Superiority for Recognizing Emotional
Aspects of Stimuli 3
Emotional Prosody Studies 3
Affective Faces Studies 8
Related Research 13
Right Hemisphere Dominance in Regulation of Mood and Affect....16
Emotional Activation Studies 16
Mood and Affect Studies 20
Left Hemisphere Superiority for Positive Affect; Right
Hemisphere Superiority for Negative Affect 25
Mechanisms of Emotional Processing: The Role of Arousal 31
Neuropsychological Models of Emotional Processing 36
The Model of Fox and Davidson 37
Kinsbourne's Model 38
Tucker's Model 40
Heilman's Model 42
Critical Issues 45
Hypotheses and Predictions 49
METHOD 52
Subjects 52
Experiment I: Reaction Time Task 52
Stimuli 52
Apparatus 53
Procedure 54
Experiment II: Psychophysiological Responses to Laterally
Presented Emotional Material 55
Stimuli 55
Apparatus 55
Procedure 57
IV

RESULTS
58
Experiment I: Reaction Time Task 58
Reaction Time Responses 58
Percent Correct Responses 67
Experiment II 72
Heart Rate Data Reduction 72
Skin Conductance Data Reduction 8M
DISCUSSION 88
Critical Issues 95
Conclusions 105
REFERENCES 109
BIOGRAPHICAL SKETCH 127
v

Abstract of Dissertation Presented to the Graduate School
of the University of Florida in Partial Fulfillment of the
Requirements for the Degree of Doctor of Philosophy
PSYCHOPHYSIOLOGICAL AND REACTION TIME RESPONSES
TO LATERALLY PRESENTED EMOTIONAL STIMULI
BY
CYNTHIA RODRIGUES CIMINO
December, 1988
Chair: Dawn Bowers
Major Department: Clinical and Health Psychology
Neuropsychological investigations of brain injured and
neurologically intact subjects have suggested that the two hemispheres
differ in terms of their contribution to emotional processing.
Several different models have been proposed to account for these
observed hemispheric asymmetries. One model, the right hemisphere
emotion model, suggests that the right hemisphere (RH) is globally
involved in all aspects of emotional processing. A second model, the
hemispheric valence model, suggests that the left hemisphere (LH) is
dominant for processing positive emotions, whereas the RH is dominant
for processing negative emotions. A third model, the preparatory
model, suggests the RH is dominant for mediating arousal/activation
and, as such, is more intrinscally involved in processing emotional
stimuli that have greater "preparatory" significance for survival
(i.e., fight-flight emotions such as anger and fear). In contrast,
the LH is more involved in mediating nonpreparatory emotions (i.e.,
happiness, sadness, disgust) that place less immediate or "phasic"
vi

demands on the organism for survival. Alternatively, it is possible
that the RH may be dominant for mediating arousal/activation responses
to stimuli, regardless of their emotional/nonemotional content.
The focus of the present study was to further examine these
different conceptualizations of the hemispheric processing of
emotional stimuli in neurologically intact male and female subjects.
This was accomplished in two separate experiments: (a) a choice
reaction time task was used to investigate subjects' activation
responses to a centrally presented, neutral stimuli when it was
preceded by neutral or emotional warning stimuli and (b) heart rate
(HR) and skin conductance (SC) measures were used to investigate
subjects' responses to laterally presented neutral and emotional
stimuli. Findings from Experiment I, the reaction time experiment,
failed to support any of the laterality models of emotion. Findings
from Experiment II, using HR and SC measures, were more congruent with
models of hemispheric differences in processing of emotional
stimuli. Males showed lateralized effects of HR, arousal responses
which support greater RH involvement in production of arousal
responses. However, this effect was present regardless of the
emotional/nonemotional content of the stimulus. Skin conductance
responses in male subjects did provide some support for hemispheric
specific emotional valence effects but this did not reach statistical
significance. In contrast, to the hemispheric asymmetries in arousal
responses observed in male subjects, female subjects did not show
significant differences in arousal responses across left and right
visual fields. However, females were differentially impacted by the
vi 1

emotional valence of the stimuli. Female subjects showed autonomic
patterning effects across emotional categories, a finding which may be
accounted for by imagery differences across male and female
subjects. Taken together, these findings point to the relative
importance of considering both sex and imagery ability of subjects in
future investigations of emotional processing and autonomic
responding.
vi ii

INTRODUCTION
Neuropsychological approaches to the investigation of emotional
processing have evolved, in large part, from early clinical
observation of brain injured patients and subsequent systematic
investigation of their performance on a variety of emotional tasks.
One of the earliest reports was provided by Babinski (191^) in which
he noted that patients with right hemisphere damage appeared
indifferent or euphoric. Denny-Brown, Meyer, and Horenstein (1952)
also reported evidence of such "indifference" reactions after right
hemisphere lesion and noted its co-occurrence with unilateral neglect
syndrome in which patients failed to orient, report, or respond to the
left side of their body. In 1952, Goldstein published his observation
that "catastrophic" emotional responses often accompanied left
hemisphere damage. These reports were later corroborated by Hecaen
(1962), who al30 noted that catastrophic reactions most often followed
left hemisphere insult whereas indifference reactions were more
frequent following right hemisphere damage.
In 1972, Gainotti reported a large scale study of 160 patients
who had sustained left or right sided lesions. Based on systematic
observation of the frequency and type of symptomatology, Gainotti
reported a consistent relationship between behaviors indicative of a
1

2
depressive-catastrophic reaction and left hemisphere damage (LHD).
Behaviors of emotional indifference and minimization of deficits were
associated with right hemisphere damage (RHD). Similarly, Terzian
(1964) and Rossi and Rosadini (1967) reported findings from unilateral
carotid injection of sodium amytal and found depressive-catastrophic
reaction following left sided injection and inappropriate euphoria
following right sided injection. Milner (cited in Rossi & Rosadini,
1967) attempted to replicate these findings without success. In 1982,
Sackeim and colleagues reported 119 cases of pathological laughing and
crying in response to unilateral lesions. Results were congruent with
Gainotti's findings with laughing outbursts more frequent following
RHD and crying more often following LHD.
Goldstein (1948) as well as Gainotti (1972) have interpreted such
depressive-catastrophic reactions following LHD as a "normal" response
to a significant loss of physical as well as psychological function.
In contrast, emotional changes seen with RHD were interpreted by
Gainotti (1972) as an abnormal response associated with anosogno3ia or
denial of illness.
More recent interpretations of such findings, however, have
suggested that the indifference reaction may be due to the RHD
patient's inability to accurately comprehend and/or express affect.
This led to the hypothesis that the right hemisphere, in the intact
state, is superior for the perception and/or expression of emotional
material.
In the years which followed, more systematic investigation was
applied to the understanding of the right hemisphere's role in the

3
processing of emotional stimuli Two major areas of research have
addressed this specific question: those which examine the processing
of the prosodic elements of speech and those which examine the
processing of affective faces. Related areas of investigation will
also be discussed.
Right Hemisphere Superiority for Recognizing
Emotional Aspects of Stimuli
Emotional Prosody Studies
It is well known that in right handers, the left hemisphere is
more adept than the right hemisphere in decoding the linguistic
content (semantic and phonemic elements) of speech (Benson &
Geschwind, 1971). However, speech may carry at least two levels of
information content: the linguistic content which conveys what is
said and the prosodic content which conveys the way in which it is
said. Prosodic elements which are defined as pitch, tempo, and
rhythm, carry information about the emotional as well as nonemotional
content of prosodic speech (Paul, 1909). Nonemotional prosody is
important for conveying whether a sentence is a question, a statement,
or a command. Emotional prosody is critical for conveying affective
information.
In 1975, Heilman, Scholes, and Watson studied the ability of
right temporo-parietal and left temporo-parietal damaged patients to
identify affective prosody. Patients were presented with semantically
neutral sentences which were read in one of four emotional tones
happy, sad, angry, or indifferent. In this study, the subject's task

4
was to identify the emotional tone of the speaker. Results
demonstrated that the RHD performed significantly worse on this task
than the LHD group.
In a subsequent study, Tucker, Watson, and Heilman (1977b)
investigated patient's ability to discriminate between different
affectively toned sentences. In this task, subjects were not required
to identify the affective tone but to discriminate between sentences
presented with the same or different affective tone. RHD patients,
again, performed more poorly than LHD patients, providing further
evidence for the right hemisphere's greater role in the processing of
emotional aspects of speech.
Weintraub, Mesulam, and Kramer (1981) have suggested that RHD
patients have difficulty in both emotional and nonemotional aspects of
prosodic speech. In their study, RHD patients had significantly
greater difficulty in distinguishing whether filtered sentences were
questions, commands, or statements. Based on their findings,
Weintraub et al. suggested that the poor performance of RHD patients
on emotional prosody tasks may be accounted for by a more general
deficit in the processing of prosodic information. No LHD group was
reported in this investigation.
In a subsequent study which addressed this question, HeiLman,
Bowers, Speedie, and Coslett (1984) investigated the ability of RHD
and LHD patients to comprehend filtered sentences which contained
either emotional (happy, sad, angry) or nonemotional (declarative,
imperative, or interrogative) prosody. Results demonstrated that both
RHD and LHD patients were impaired on the nonemotional prosody task

5
compared to control subjects. In contrast, RHD patients performed
significantly worse on the emotional prosody task relative to LHD
patients, suggesting a greater role for the right hemisphere in the
comprehension of emotional prosody.
In addition to these reports of deficits in the comprehension and
discrimination of affective prosody, Tucker et al. ( 1977b) also found
that RHD patients had difficulty in producing affectively intoned
speech. Patients were asked to say a semantically neutral sentence
using either a happy, sad, angry, or indifferent tone. RHD patients
performed significantly worse than LHD patients suggesting that their
deficits include not only the comprehension and discrimination of
affectively intoned speech but also the expression of affectively
intoned speech.
This finding was later supported by Ross and Mesulam (1979) who
reported two patients who could not express affectively intoned speech
but could comprehend affective speech. In addition, Ross (1981) has
also reported patients who could not comprehend affective intonations
but could repeat affectively intoned speech. Ross has suggested that
the right hemisphere may mediate the comprehension, repetition, and
production of affective speech much in the same way as the left
hemisphere does for propositional speech with anterior lesions
producing primarily production defects and posterior lesions producing
primarily comprehension defects.
With the advent of experimental procedures such as tachistoscopic
presentation and dichotic listening in which stimulus processing is
initially restricted to the left or right hemisphere, investigations

6
of emotional processing in normal, neurologically intact subjects have
also examined the right hemisphere's role in the processing of
emotional prosody. Using a dichotic listening (DL) procedure, Haggard
and Parkinson (1971) paired speech babble with short sentences spoken
in one of four emotional tones. They found that accuracy in
identifying the emotional tone was significantly better on left ear
trials, suggesting a right hemisphere superiority.
Similarly, Safer and Leventhal (1977) used monoaural presentation
of sentences with positive, negative, and neutral content spoken in a
positive, negative, or neutral tone. Results demonstrated that
subjects who listened to sentences in their left ear tended to use the
intonation in making their judgements whereas subjects who listened to
sentences in their right ear tended to use the content of the sentence
in making judgement. Interpretation of these findings, however, has
been questioned because of the use of a between groups comparison for
each ear presentation. The findings are suggestive, nevertheless.
Ley and Bryden (1982), using a DL procedure, had subjects report
on the emotional tone and content of a sentence arriving at either the
left or right ear (specified on each trial). Subjects were more
accurate in judging the emotional tone of the sentence when monitoring
the left ear and more accurate in judging the content of the sentences
when monitoring the right ear.
In a subsequent experiment, Bryden, Ley, and Sugarman (1982)
investigated hemispheric differences in ability to judge the emotional
tone of musical stimuli They did this by taking advantage of the
fact that in Western culture music written in a major key is generally

7
described as happy while music written in a minor is more often
described as sad (Davies, 1978). In a DL procedure, subjects were
required to identify the emotional tone of a short seven-note passage
while monitoring either the left or right ear. Subjects were more
accurate when identifying the emotional tone of passages presented to
the left ear relative to those presented to the right ear, again
supporting the notion of the right hemisphere dominance in processing
of emotional stimuli.
Dichotic listening procedures in normal subjects have also
demonstrated left ear advantages in recognition of other nonverbal,
emotional aspects of human speech such as laughing and crying (Garmon
& Nachson, 1973; King & Kimura, 1972). In a study which used a
variant of the monoaural paradigm in which subjects heard spoken
captions and laughter in either the left or right ear, cartoons were
judged as funnier when the laughter was heard by the left ear relative
to the right ear (DeWitt, 1978).
Recently, Mahoney and Sainsbury (1987) investigated hemispheric
asymmetries in perception of human, nonspeech emotional sounds.
During conditions of divided attention, a left ear advantage emerged
during the second block of trials. Under conditions of selective
attention, however, this left ear advantage was seen on the first
block of trials. In addition to providing support for a right
hemisphere superiority in processing of emotional nonspeech sounds,
these findings also suggest that effects of attention influenced the
rate and development of observed laterality effects but not the
direction of these effects.

8
In summary, investigations in both normal and brain impaired
subjects have provided evidence that the right hemisphere is
preferentially involved in the comprehension and expression of
emotional prosodic elements of speech and other nonverbal
vocalizations. Similarly, a large body of literature has also
investigated the role of the right hemisphere in the processing of
affective faces.
Affective Faces Studies
In 1980, DeKosky, Heilman, Bowers, and Valenstein reported a
study which investigated RHD and LHD patients' ability to make neutral
facial discriminations as well as affective facial discriminations.
They found that RHD patients performed more poorly than LHD patients
on both facial affect judgements as well as neutral facial
discrimination. In fact, when the two groups were statistically
equated for performance on the neutral discrimination task,
differences between RHD and LHD patients on the affective facial
discrimination task disappeared. These findings suggested that RHD
patients' poor performance on facial affect judgements can be solely
accounted for by their poor performance in facial discrimination
ability. This has led to the question of whether processing the
emotionality of a face, in fact, involves a "stimulus-content
dimension" in its own right or whether such processing merely involves
an increase in the configurational complexity of the stimulus and
consequently increases the demand on right-hemisphere mediated
visuospatial skills.

9
More recent investigations have challenged this latter
interpretation. Support for such a dissociation has recently been
provided in a study by Kolb and Taylor (1981). Their findings
revealed that patients with parietal excision were impaired on both
matching of facial affect and matching of facial identity. However,
these authors also report that patients with damage restricted to
right temporal and right frontal regions were more impaired on
processing of facial affect relative to processing of facial
identity. Similarly, a study by Freid, Mateu, Ojemann, Wohns, and
Fedio (1982) reported that neutral facial matching was disrupted by
stimulation of right parietal-occipital regions, whereas naming the
affect depicted on pictures of faces was disrupted by stimulation of
right posterior middle temporal gyrus.
In contrast, Cicone, Wapner, and Gardner (1980) found no
relationship between RHD patients' performance on an emotional
perception task and a facial identity task, suggesting that deficits
associated with RHD cannot be accounted for by facial recognition
deficits alone.
More recently, Bowers and Heilman (1984) reported a case which
demonstrated a dissociation between processing of affective and
nonaffective faces. Although this RHD patient was able to perform
well on neutral facial tasks and on same-different emotional faces
tasks, he was impairea on naming and comprehension of verbal labels
for facial expressions.
In a subsequent investigation, Bowers, Bauer, Cosiett, and
Heilman (1985) again addressed the question of whether defects shown

10
by RHD patients on facial affect tasks are dissociable from defects in
visuoperceptual processing. They cite two criticisms of the DeKosky
study which directly address this issue. First, they noted that a
small subset of RHD patients in the DeKosky study performed normally
on the neutral visuoperceptual task, yet, were impaired on the facial
affect tasks. This suggested that visuoperceptual deficits alone
cannot account for impaired processing of affective faces in all RHD
patients.
Secondly, they suggested that the use of same actors on affective
faces trials may have allowed subjects to rely on a pure template
matching strategy in which judgements about emotionality could have
been made on the basis of whether the two faces had the same
physiognomic configuration. A defect in this type of perceptual
process could then potentially affect performance on both facial
identity as well as affective facial tasks. As an alternative, Bowers
et al. required subjects to make affective facial judgements across
different actors so that such judgements would take place in an
"associative" context with less reliance on potential defective
perceptual matching.
Results of this study revealed that when patient groups were
statistically equated on visuoperceptual ability (facial identity
task), RHD patients still performed worse than LHD patients and
control subjects on (a) emotional discrimination of different actors,
(b) naming the emotion of a single face, and (c) picking the named
emotion from four pictures of the same actor. These findings provide
strong evidence that differences in LHD and RHD patients' abilities to

make affective judgements cannot be accounted for solely by
differences in the visuoperceptual processes underlying facial
identity discrimination.
Recent support for the relative dissociation of facial identity
judgements from facial affect judgements has also been provided by
Tranel, Damasio, and Damasio (1988). Tranel et al. described four
patients with bilateral lesions of occipitotemporal or temporal
regions whose performance on facial affect tasks were significantly
better than their performance on facial identity tasks.
Research in normal subjects has also investigated the role of the
right hemisphere in the processing of affective faces. Ley and Bryden
(1979) tachistoscopically presented faces to the left visual field
(LVF) and right visual field (RVF), and subjects made either facial
identity judgements or facial affective judgements. They found a LVF
superiority for both tasks. However, when performance on the facial
affect task was reanalyzed using performance on the facial identity
task as a covariate, the LVF superiority for making affective facial
judgements remained. These findings, which are similar to those of
Bowers et al. (1985), suggest that the right hemisphere superiority
for processing facial affect exists above and beyond the superiority
for processing facial identity.
in 1977, Suoeri and McKeever reported a reaction time (RT) task
in which subjects responded to previously memorized emotional or
nonemotional faces presented in LVF or RVF. Subjects who memorized
emotional faces showed significantly faster reaction times to LVF
targets than subjects who memorized nonemotional faces. Based on

12
these findings, Suberi and McKeever argued that the LVF effect for
processing nonemotional faces is significantly enhanced by
presentation of emotional faces.
McKeever and Dixon (1981) used emotional imagery and neutral
faces to investigate right hemisphere effects in processing of
affective material. They instructed subjects to imagine that
something very sad happened to a number of predetermined target
faces. In a subsequent target/nontarget discrimination task with
lateralized presentations, they report that the use of emotional
imagery significantly enhanced LVF (right hemisphere) performance.
This effect, however, was demonstrated in female subjects only.
Safer (1981) reported a study in which subjects memorized faces
by either empathizing with their emotional expressions or by labeling
the emotional expressions. Results demonstrated that subjects who
used empathy recognized more faces presented to the LVF than RVF. No
laterality effect was demonstrated for those who labeled faces. This
laterality effect for the empathy condition, however, was found for
male subjects only. Similarly, Buchtel, Campari, DeRisio, and Rota
(1978) reported faster responding to both positive and negative
stimuli presented in LVF relative to neutral targets. Hansch and
Pirozzolo (1980) and Strauss and Moskovitch (1981) also reported a LVF
effect for neutral and emotional faces.
In summary, investigations in both normal and brain impaired
subjects have supported the notion that the right hemisphere is
preferentially involved in the processing of affective faces. In
addition, several studies have also suggested that right hemisphere

13
advantage for processing of facial affect exists above and beyond the
right hemisphere's advantage for processing of facial identity.
Evidence about possible differences due to sex of subjects, however,
remains equivocal.
Related Research
Several studies have implicated the right hemisphere in memory
for emotionally charged materials. Weschler (1973) reported one of
the few studies of emotional memory in brain impaired subjects. Right
hemisphere and LHD patients were presented with two types of stories--
one emotional and the other nonemotional. When asked for subsequent
recall, RHD subjects made significantly more errors in recalling
emotional stories relative to LHD patients.
Cimino, Verfaellie, Bowers, and Heilman (1988) investigated
whether RHD patients have difficulty remembering past affective
episodes by asking them to recall prior emotional and neutral
experiences. Findings revealed that RHD patients produced
significantly less emotional reports than control subjects as judged
by independent raters. However, their own emotionality ratings were
no different from those of control subjects suggesting some
discordance between their actual production of emotional memories
versus their own perceived emotionality of such memories.
Unfortunately, most patients with LHD are aphasic and could not be
used in this study. Therefore, this report cannot conclude that this
defect is specific to RHD.
An investigation in normal subjects (Gage & Safer, 1979) looked
at hemispheric differences in mood-state dependent effects for

recognition of emotional faces. The mood-state dependent effect
states that stimuli will be better recalled when subject's mood at
encoding and retrieval are congruent than when they are disparate
(Bower, 1981). Using a mood induction procedure, Gage and Safer
demonstrated that recognition of faces initially presented in a
discrepant mood was significantly worse when presented to the LVF
(right hemisphere) than when presented to the RVF (left hemisphere).
Based on these observations, the authors suggest that the right
hemisphere stores the subject's mood as an integral part of the memory
representation to a greater extent than the left hemisphere.
Although it is well known that the left hemisphere is specialized
for mediating speech and linguistic stimuli in the vast majority of
right handed adults, the right hemisphere also appears to have some
reduced language capacity (Geschwind, 1969; Papanicolaou, Moore,
Deutsch, Levin, & Eisenberg, 1988). Several recent studies have
suggested that the right hemisphere may better process emotional
versus nonemotional verbal stimuli, at least at the single word
level. For example, Graves and coworkers (Graves, Landis, 4
Goodglass, 1980) found that aphasic patients with alexia due to left
hemisphere lesions could read emotional words better than nonemotional
words. In a subsequent study, these same investigators (Graves,
Landis, & Goodglass, 1981) found that neurologically intact males
subjects better recognized emotional words than nonemotional words
when these stimuli were presented to the LVF (right hemisphere).
Other researchers (Brody, Goodman, Holm, Krinzman, & Sebrechts,
1987) have looked at the effects of lateralized affective priming

15
stimuli (faces/words) on subsequent judgements of the affective value
of laterally presented emotional and nonemotional verbal target
stimuli. They found that affective primes presented to the RVF (left
hemisphere) resulted in decreased accuracy judgements of the target
stimuli that were also presented to that hemisphere. In contrast,
affective primes presented to the LVF (right hemisphere) resulted in
increased accuracy judgements regarding the affective value of verbal
stimuli presented to the right hemisphere.
Taken together, findings with both aphasic and neurologically
intact individuals suggest a right hemisphere advantage in processing
emotional verbal stimuLi. However, this view is not entirely clearcut
in that other studies have failed to replicate the "right hemisphere"
laterality effect for identifying emotional versus nonemotional verbal
stimuli (Strauss, 1983).
Another area of investigation concerning the role of the right
hemisphere in processing of emotional stimuLi is that of humor
appreciation. Brownell et al. (1984) have recently reported that RHD
patients have significant difficulty in understanding narrative humor
as portrayed in short story jokes. Similarly, Bihrle et al. (1986)
have reported that RHD patients performed significantly worse than LHD
patients on a nonverbal cartoon completion task.
In summary, a large body of literature suggests that the right
hemisphere plays a greater role in the comprehension and expression of
emotional information. These findings have been consistently
demonstrated in both neurologic as well as normal populations across a
wide range of tasks including affective facial judgements and

16
production, emotional prosody judgements and production, humor
appreciation, and emotional memory.
Right Hemisphere Dominance in Regulation of Mood and Affect
In addition to studies which provide evidence that the right
hemisphere is superior for recognizing emotional aspects of
information, recent investigations have also suggested that the right
hemisphere is dominant in regulating mood and affect. These studies
fail into two major categories. The first category includes those
studies which suggest that the right hemisphere is preferentially
activated during episodes of felt emotion, primarily in normal
subjects. The second category includes those studies which correlate
psychiatric disorders of mood and affect with decrement in right
hemisphere functions.
Smotional Activation Studies
Investigations which have looked at the right hemisphere's
activation during period of felt emotion have used several different
indices of cerebral activation. These have included such measures as
electrocortical activity (usually in terms of decreased alpha power),
measures of lateral eye movements, and asymmetries of facial
expression.
Davidson and Schwartz (1976) reported that subjects showed
greater right than left hemisphere EEG activity when recalling past
events associated with anger or relaxation and during self-reported
emotional reactions to visual material (Davidson, Schwartz, Saron,

17
Bennett, & Golemena, 1979). Right hemisphere activation has also been
reported during hypnotically induced depression (Tucker, Stenslie,
Roth, & Shearer, 1981), during generation of emotional imagery and
during painful stimulation (Karlin, Weinapple, Rochford, 4 Goldstein,
1979).
In addition to comparisons of left versus right sided activation,
several authors also emphasize the importance of relative differences
in level of activation in anterior versus posterior regions. Tucker
(1981) reported frontal activation but not posterior activation in
depressed mood. Likewise, Davidson et al. (1979) reported that mood
valence varied with right versus left activation in frontal regions,
but that posterior regions showed right hemisphere activation
irrespective of valence.
Lateral eye movements (LEM) as indices of hemispheric activation
have also been used to assess the role of the right hemisphere in
regulation of mood and affect. Prior investigations have revealed a
tendency toward right LEM (left hemisphere activation) with verbal
processing and left LEM (right hemisphere activation) with
visuospatial processing (Kinsbourne, 1972). Schwartz, Davidson, and
Maier (1975) have reported a greater frequency of left LEM in subjects
performing emotional versus neutral mental tasks.
Similarly, Borod, Vingiano, and Cytryn (1988b) measured LEM while
subjects were asked to generate emotional images of positive and
negative valence in auditory, visual, and tactile modalities.
Overall, subjects looked significantly more to the left than to the

18
right in response to emotional instructions. These findings suggest a
greater role of the right hemisphere in generating emotional imagery.
In addition, Tucker, Roth, Arneson, and Buckingham (1977a) have
reported more left LEM in anxious than nonanxious subjects. Woods
(1977) has also suggested that habitual left eye movers are higher in
intensity and frequency of emotional reactions than right eye movers.
While findings from LEM studies may appear to be conceptually
apparent, interpretations of findings from such investigations must be
cautioned in terms of the questionable reliability and validity of LEM
as indicators of hemispheric activation. Berg and Harris (1980) were
unable to replicate previous findings in LEM studies and concluded
that the validity of the LEM procedure as a measure of hemispheric
activation has yet to be established. Ehrlichman and Weinberger
(1978), in a detailed review of the LEM literature, similarly
concluded that the use of LEM in investigations of hemispheric
functioning was questionable at best.
Recently, lacrimal flow has also been utilized as an index of
hemispheric involvement in production of mood states. Delp and
Sackeim (1987) looked at lacrimal flow following sadness and happiness
mood manipulation in male and female subjects. For female subjects,
the sadness manipulation resulted in greater relative left eye
lacrimal flow, whereas the happiness manipulation resulted in a shift
toward greater relative reduction in left eye flow. Although these
findings may be interpreted as support for greater right hemisphere
involvement in lacrimal flow, this interpretation must be observed
with some caution as the assumed lateralization of specific

19
neuroanatomical pathways regulating lacrimal flow have not been
clearly established.
Measures of facial asymmetries observed during emotional
expression have also provided support for the hypothesis of right
hemisphere dominance in regulating affect and mood states.
Musculature of the lower part of the face is contralaterally
innervated and asymmetries observed with respect to facial expression
may be used to infer relative hemispheric involvement in production of
emotional expression.
Studies with normal subjects have revealed that the left hemiface
moves more extensively during posed facial expression (Borod & Caron,
1980; Borod, Caron, i Koff, 1981; Borod, Kent, Koff, Martin, & Alpert,
1988a; Borod, Koff, & White, 1983; Moskovitcn & Olds, 1982). One
criticism of such studies, however, is that they have used posed
facial expressions which may not actually reflect the underlying
affect or mood of the subject. In response to this criticism, several
authors have regarded spontaneous facial expression as a more valid
index of the subject's affective state. Ekman, Hager, and Friesen
(1981) failed to find such asymmetries of emotional expression during
spontaneous facial expressions. In contrast, other investigators
(Borod et al., 1983; Moskovitch & Olds, 1982) have observed greater
left sided (right hemisphere) involvement for both spontaneous and
posed facial expressions.
Several studies have also used composite photographs in which the
mirror image of the left or right half of the face is combined with
the original ipsilateral image. This process results in a complete

20
facial photo composed entirely of the left or right half of the
face. These studies report that left sided composites of posed facial
expressions are judged as more intense than right sided composites
(Heller & Levy, 1981; Rubin & Rubin, 1980; Sackeim, Gur, & Saucy,
1978). This effect has also been reported for spontaneous expressions
as well (Dopson, Beckwith, Tucker, & Bullard-Bates, 1984).
Clinical reports have indicated that spontaneous emotional facial
expressions are less likely to occur in RHD patients compared to LHD
patients (Borod, Koff, Perliman, Lorch, & Nicholas, 1986; Buck &
Duffy, 1980; Ross & Mesulam, 1979). Kolb and Milner (1981), however,
were unable to find differences between RHD and LHD patients in
spontaneous facial expressions and movements (only some of which were
emotional). They did find, however, differences with respect to the
anterior versus posterior distribution of the lesion, with anterior
lesions resulting in less spontaneous movements than more posterior
lesions. Borod et al. (1986) recently reported the only systematic
investigation, to date, of posed and spontaneous facial expressions in
brain impaired patients. They found that RHD patients produced fewer
posed as well as spontaneous emotional facial expressions than did LHD
patients.
Mood and Affect Studies
A second group of studies has also examined disorders of mood and
affect associated with right hemisphere function. These
investigations have studied mood and affective changes in patients
with known hemispheric pathology. Patient groups have included

21
epileptic patients with unilateral foci, head injury patients, and
patients with unilateral subcortical infarction or surgical removal.
In a factor analytic study which assessed interictal personality
changes associated with right temporal lobe epilepsy, Bear and Fedio
(1977) reported that right foci were more associated with affective
changes while left foci were more associated with cognitive changes.
Similarly, Taylor (1972) described a predominance of right sided foci
in a sample of epileptics with associated symptoms of depression,
anxiety, and phooias. In 1969, Flor-Henry reported that of patients
with unilateral epileptic foci, manic-depressive psychosis was found
twice as many times in individuals with right sided foci than in those
with left sided foci.
Lishman (1968), in his review of 144 cases of head injury, found
that affective disturbances were more common following right
hemisphere injury while cognitive/intellectual changes were more often
seen following left sided injury. These findings parallel those of
Bear and Fedio (1977) which looked at changes associated with left and
right epileptic foci. Recently, mania has also been reported
following right thalamectomy (Whitlock, 1982) and right thalamic
infarct (Cummings & Mendez, 1984).
Investigators have also looked at signs of hemispheric
dysfunction in patients with primary mood disturbance as a means of
investigating lateralized hemispheric functioning in relation to
disorders of mood and affect. Several investigators, utilizing EEC as
a measure of hemispheric activity, have reported low left versus right
hemisphere activity in depressed patients with differences occurring

22
primarily in frontal and central regions (d'Elia & Perris, 1973; 1974;
Perris, 1975). These studies have also found that increased right
sided activity correlated significantly with the severity of
depression as well as performance on a verbal learning task. These
findings have been replicated by other investigators as well
(Rockford, Swartzburg, Chaudberg, 4 Goldstein, 1976). Perris (1974)
has also reported lower left to right amplitudes of visual evoked
responses in depressed patients relative to schizophrenics and normal
controls. Taken together, these findings suggest greater right
hemisphere relative to left hemisphere activity in depressed mood.
Two interpretations of these findings have been suggested. One
maintains that such hemispheric differences represent a relative left
hemisphere underresponsivenes3 (Rockford et al., 1976). A second
possibility is that such differences represent a right hemisphere
overresponsiveness (d'Elia & Perris, 1974).
Recently, subtle left sided neurological signs have been reported
in depressed patients, suggesting right hemisphere involvement. The
first report was presented by Brumback and Staton (1981). They
described two depressed children who presented with pronator drift of
the left arm, hyperreactive left deep tendon reflexes, and left
extensor plantar responses; these symptoms resolved following
treatment with tricyclic antidepressants. Similarly, Freeman,
Galaburda, Cabal, and Geschwind (1985) reported a case of a 62-year-
old depressed female who had left sided facial weakness, a gaze
preference to the right, and limited use of her left arm. Again,
these symptoms resolved following treatment with ECT.

23
Findings from investigations of galvanic skin responses (GSR)
have provided only partial support for the presence of lateralized
dysfunction in depressed patients. Schneider (1983) reported lower
right handed GSRs in a depressed sample. In addition, Myslobodsky and
Horesch (1978) have noted higher left handed GSRs in depressed
subjects. Toone, Cooke, and Lader (1981), however, were unable to
replicate these findings. Such discrepancies may be accounted for by
conflicting reports which suggest that GSR responses may be controlled
ipsilaterally, contralaterally, or bilaterally (Holloway & Parsons,
1969; Lacroix & Comper, 1979; Myslobodsky & Rattok, 1977).
The performance of depressed patients on tests likely to require
greater right hemisphere processing has also been investigated.
Several studies sugggest that depressed subjects perform more poorly
on vi3uospatial tasks than verbal tasks (Flor-Henry, 1976; 1983;
Goldstein, Filskov, Weaver, & Ives, 1977; Kronfol, Hamsher, Digre, &
Wazir, 1978). Silberman, Weingartner, and Post (1983a) have also
suggested that the pattern of errors in depressed subjects closely
resembles that of right temporal lobectomized patients and that the
degree of impairment is correlated with the overall severity of
depression. However, as cautioned by Weingartner and Silberman
(1982), the impaired verbal learning and memory performance that is
frequently observed in depressed patients also points to left
hemisphere dysfunction. Because of decrements observed on left as
well as right hemisphere tasks, Weingartner and Siloerman (1982)
suggested that it is best to describe the deficits observed as a

24
relative impairment on right hemisphere tasks as compared to left
hemisphere tasks.
Several studies have also reported the occurrence of "reversed
lateralization" or "functional delateralization" in depressed patients
on tasks of verbal and nonverbal processing. Bruder (1983) published
a review of dichotic listening studies and concluded that depressed
patients demonstrated decreased lateralization on both verbal as well
as nonverbal dichotic listening tasks. Several authors, however, have
suggested that decreased lateralization is present primarily on
nonverbal tasks (Coulbourn & Lishman, 1974; Johnson & Crockett,
1982). Evidence of reversed lateralization has also been reported by
some authors. Silberman, Weingartner, Stillman, Chen, & Post (1983b)
have reported a left visual field superiority on a verbal task in a
sample of depressed females. Similarly, research in depressed
patients has also found (Flor-Henry, 1979; Flor-Henry & Koles, 1980)
increased parietal activity during rest, with left temporal activation
during spatial task3 and right parietal activation during verbal
tasks. These findings are at variance with predicted asymmetries in
normal populations. Hommes and Panhuysen (1971), using a small sample
of depressed patients, have reported that right sided sodium amytal
injections resulted in a degree of aphasia for all subjects.
Furthermore, this finding was significantly correlated with the
severity of depressive symptomatology.
In summary, a large body of the neuropsychology literature on
emotional processing has suggested a right hemisphere dominance in the
comprehension and expression of emotional information. These findings

25
have been consistently demonstrated in neurologic and psychiatric, as
well as normal, populations using a range of tasks including affective
facial judgements and production, emotional prosody judgements and
production, emotional memory, emotional verbal judgements, production
of affect, and humor appreciation.
Left Hemisphere Superiority for Positive Affect;
Right Hemisphere Superiority for Negative Affect
Early clinical reports of emotional/mood changes following brain
injury have also been interpreted as supporting a distinction between
the superiority of the left hemisphere in the processing of positive
affect and the superiority of the right hemisphere in the processing
of negative affect. Interpretation of these studies is based on the
notion of reciprocal inhibition which states that each hemisphere
exerts some degree of inhibition on the contralateral hemisphere
(Kinsbourne, 1973). In this view, damage to the left hemisphere would
result in disinhibition of the right hemisphere's negative affective
bias. Right hemisphere lesions would result in disinhibition of the
left hemisphere's positive affective bias.
Early observations suggested that RHD patients often appeared
indifferent or euphoric (Babinski, 1914; Denny-Brown et al., 1952).
In contrast, several authors reported that left hemisphere damage was
more associated with a "catastrophic" emotional response (Goldstein,
1952; Hecaen, 1962). These observations were later corroborated by
Gainotti (1972) in a large scale study of 160 patients. Based on
systematic investigation of patients' symptomatology, Gainotti

26
reported a consistent relationship between (a) depressive-catastrophic
reaction and left hemisphere damage and (b) indifference/minimization
of deficits and right hemisphere damage. Recently, Heilman and
colleagues (Heilman et al., 1 975; Heilman, Schwartz, 4 Watson, 1978;
Heilman, Watson, & Bowers, 1983) have noted indifference reactions
occur with striking frequency in RHD patients with the neglect
syndrome suggesting that the two syndromes may be associated in some
manner. This also suggests that right hemisphere changes associated
with inappropriate euphoria and indifference may represent either (a)
two points on a continuum or (b) two distinct emotional reactions
following brain injury, one of them sharing a common mechanism with
the unilateral neglect syndrome.
Using the Depression scale of the Minnesota Multiphasic
Personality Inventory as an index of depressive symptomatology,
Gasparrini, Satz, Heilman, and Coolidge (1978) reported significantly
elevated scores for LHD but not for RHD patients. More recently,
Robinson and colleagues (Robinson, 1983; Robinson 4 Price, 1982) have
also found that LHD patients were more likely to become clinically
depressed and that RHD patients were more likely to be inappropriately
euphoric. In addition, these results were not correlated with overall
cognitive impairment, suggesting that the patient's emotional
reactions to their deficits cannot solely account for these
findings. Location of damage within the hemisphere has been found to
be important in that these emotional reactions are more frequently
associated with damage to anterior, frontal regions (Kolb 4 Milner,
1981; Robinson 4 Benson, 1981).

27
Findings from studies of unilateral carotid injection of sodium
amytal (WADA procedure) also support reported differences in emotional
changes following LHD and RHD. Terzian (1964) and Rossi and Rosadini
(1967) reported depressive-catastrophic reactions following left sided
injections and inappropriate euphoria following right sided
injections. These findings are also supported by other reports
(Alema, Rosadini, & Rossi, 1961; Perria, Rosadini, & Rossi, 1961).
However, Milner (cited in Rossi & Rosadini, 1967) failed to replicate
these findings. In her investigation, only 5% of patients displayed
depressive type responses, while the majority displayed euphoric
reactions. This discrepancy between Milner's study and those of other
investigators may be related to the significantly higher doses of
sodium amytal used in the Milner study (Silberman & Weingartner,
1986).
Investigation of normal, neurologically intact subjects has also
provided some support for the relative superiority of the left
hemisphere in the processing of positive affect and the relative
superiority of the right hemisphere in the processing of negative
affect. Davidson and colleagues (1979) recorded EEG responses while
subjects viewed television programs of varying emotional content and
subsequently indicated their emotional responses. Greater left
hemispheric activity was found in response to positive emotional
content, and greater right hemispheric activation was found in
response to negative emotional content. Interestingly, this
difference was only apparent on more anterior, frontal recordings
while posterior, parietal activity suggested relative right hemisphere

28
activation during all periods of felt emotion. These results again
suggest the significance of the anterior-posterior dimension in
processing of emotional information.
Asymmetries of emotional facial expressions have also tended to
support the relative superiority of the left hemisphere in processing
of positive affect and the relative superiority of the right
hemisphere in processing of negative affect. Sackeim et al. (1978)
have reported a tendency for facial expressions to be greater on the
left side of the face. Furthermore, these authors suggest that this
asymmetry was more pronounced for negative than positive facial
expressions. Similarly, Schwartz, Ahern, and Brown (1979) have
investigated facial expressions during spontaneous mood
fluctuations. Right sided contractions were stronger during periods
of nappiness or excitement, while left sided contractions were
stronger during facial expressions of sadness and fear.
Ahern and Schwartz (1979) have reported more right LEM (left
hemisphere activation) when subjects respond to questions that evoked
happiness or excitement. In contrast, more left LEM (right hemisphere
activation) occurred when subjects responded to questions that evoked
sad or fearful affects. As previously discussed, the questionable
validity of LEM as an index of cerebral activation, however, must be
considered in any interpretation of this study.
Studies have also investigated left and right visual field
differences for positive and negative emotions. Using a contact lens
system that restricts visual input to the RVF (left hemisphere) or LVF
(right hemisphere), Dimond, Farrington, and Johnson (1976) reported

29
that unpleasant films were rated as more unpleasant when presented to
the LVF than when presented to the RVF. More recently, Reuter-Lorenz,
Givis, and Moskovitch (1983) have also reported shorter reaction times
to RVF presentations of happy faces and LVF presentations of sad
faces. These findings are congruent with proposed left hemisphere
positive affect and right hemisphere-negative affect distinctions.
Although numerous studies are consistent with the hypothesis that
the right hemisphere preferentially mediates negative emotions and the
left hemisphere mediates positive emotions, an equal number of studies
find no support for this hemispheric valence hypothesis. Rather both
positive and negative stimuli seem to be preferentially mediated by
the right hemisphere. To account for these discrepant views on the
hemispheric processing of positive versus negative stimuli, Bryden and
Ley (1983) have argued that methodological differences across studies
might contribute to the discrepant findings. For example, Reuter-
Lorenz and Davidson (1981) report faster reaction times for LVF
presentations of sad faces and RVF presentations of happy faces. In
this study, subjects were required to identify which of two laterally
presented faces (one neutral and one emotional) showed an affective
expression. In investigations which show a significant overall right
hemisphere effect, the task is quite different. In these studies, the
task is usually to determine whether a laterally presented face is the
same or different than a centrally presented face. It is possible
that these differences in task requirements may, in some way, account
for the findings.

30
A second possibility is that negative emotional faces may be more
configurationally complex (requiring greater right hemispheric
processing), a feature which may be of even greater significance in
the task requirements of the Reuter-Lorenz task. This possibility
does not, however, account for the large number of studies which have
actually analyzed for type of emotion and still failed to find any
laterality effect due to emotional valence (Bowers et al., 1985;
Bryden et al., 1982; Buchtel et al., 1978; Heilman et al., 1984; Ley &
Bryden, 1979).
A third possibility accounting for these discrepant findings is
that studies which do find emotion specific hemispheric effects (i.e.,
left hemisphere-positive and right hemisphere-negative) are those
which deal primarily with mood and/or experiential phenomena. In
contrast, studies which do not find emotion specific hemispheric
effects but instead do find right hemisphere superiority are those
which involve cognitive encoding of emotional stimuli (i.e., "cold,"
cognitive tasks).
In their chapter, Bryden and Ley (1983) conclude that less
evidence exists to strongly support the notion that the left
hemisphere is more involved in the processing of positive affect and
the right hemisphere is more involved in the processing of negative
affect. The available evidence, however, provides strong support for
the role of the right hemisphere in processing both positive and
negative affective material.

31
Mechanisms of Emotional Processing: The Role of Arousal
Schacter and colleagues (Schacter, 1964; Schacter & Singer, 1962)
proposed a theory of emotions termed the Cognitive-Arousal model.
According to this model, an emotional state is the product of an
interaction between arousal and cognition. An important assumption is
that both arousal and cognition are necessary components of emotion.
In this view, arousal is viewed as important in determining the felt
intensity of the emotion while the cognitive element is important in
determining the specific quality of the emotion.
Early support for the joint roles of arousal and cognition were
provided by Maranon (1924, cited in Fehr and Stern, 1970) who
artificially aroused subjects with administration of drugs that
stimulated the sympathetic nervous system. Maranon's subjects did not
report feeling emotions although some did report feeling "as if"
emotions. In contrast, if subjects were given a congitive set
(induction of an affective memory) they did report emotional reactions
when artifically induced arousal was present. Schacter (1970) has
also provided support for this notion in a study which looked at the
specific effects of pharmacologically induced arousal in neutral and
stressful situation. Schacter demonstrated that physiological arousal
alone (neutral condition) was not sufficient to evoke emotional
responses from subjects but that the combination of arousal and
availability of a cognitive label (stressful situation) was
necessary. Although Schacter's research has been heavily criticized,
especially on methodological grounds, this does not refute the

32
assumption that arousal and cognition may play an important role in
emotion.
Early investigations in neurophysiology laid much of the
groundwork for our current knowledge of physiological arousal. In
1933 Berger reported that the electroencephalographic (EEG) pattern
during behavioral arousal displayed decreases in amplitude and
increases in frequency. This "electroencephalographic
desynchronization" observed during periods of behavioral arousal was
also later reported to occur during emotional states (Lindsley, 1970).
Studies have also identified critical neuroanatomic structures
involved in the elicitation of arousal responses. Stimulation in
nonspecific thalamic nuclei or the mesencephalic reticular formation
(MRF) result in behavioral manifestations of arousal as well as EEG
desynchronization (Moruzzi & Magoun, 19^9). Similarly, stimulation of
frontal or temporoparietal cortex activates the MRF (French,
Hernandez-Peon, & Livingston, 1955) and elicits an arousal response
(Sequndo, Nasuet, & Buser, 1955). Another pathway by which cortical
stimulation can produce arousal is via limbic system projections to
cortex and MRF (Heilman & Valenstein, 1972; Watson, Heilman, Cauthen,
& King, 1973).
This conceptualization of reciprocal connections between MRF
system and cortical regions is central to a model of arousal proposed
by Sokolov (1963). Sokolov described a specific pattern of
physiological changes which occurred in response to novel or
"significant" stimuli. This specific pattern of physiological changes
was termed the orientating response (OR). At the behavioral level,

33
the organism may realign the head and/or body toward the source of
stimulation. At the neurophysiological level, several changes occur
which include a transient increase in skin conductance, pupil
dilation, heart rate deceleration, pauses of respiration, and EEG
desynchronization. The presumed functional value of these collective
components of the OR is to make the organism more receptive to
incoming stimuli as well as to prepare the organism for action.
A second component of Sokolov's model is that of the defensive
response (DR), which is likely to be of equal, if not greater,
potential significance in the processing of emotional stimuli. In
Sokolov's view, when high intensity or aversive stimuli are presented,
the orienting response is soon replaced by the defensive response.
This response is characterized by greater increases in sympathetic
activity across response systems including heart rate acceleration and
cephalic vasoconstriction. The functional value of this response at
the behavioral level is avoidance of the stimulus.
It is interesting to note that both orienting and defensive
responses are conceptualized as arousal responses, yet each results in
characteristically distinct patterns of responding. This occurrence
presents difficulty for the view of arousal as a unidimensional
phenomenon. Subsequent investigators have looked at these seemingly
paradoxical heart rate responses and attempted to correlate them with
psychological processes.
Lacey (1967) reconceptualized this "directional fractionation" of
cardiac activity in terms of the conditions under which stimulation
occurred and their effects on the organism's processing of stimuli.

34
Lacey proposed that cardiac deceleration is associated with
environmental intake while heart rate acceleration is associated with
environmental rejection (the intake-rejection hypothesis). Lacey
suggested that cardiac deceleration served to facilitate sensory
processing while cardiac acceleration served to inhibit sensory
processing. In this view, stimuli which elicit attention and interest
are associated with cardiac deceleration and environmental intake. In
contrast, stimuli which are painful or aversive or which require a
significant amount of mental activity such as problem solving or
arithmetic are associated with cardiac acceleration and environmental
rejection. Lacey (1967; 1972) also proposed a neurophysiological
mechanism for such cardiac changes whereby cardiac responses altered
cortical activity indirectly by means of a visceral afferent feedback
loop mediated by the baroreceptors.
An alternative explanation for heart rate changes has been
offered by Obrist and colleagues (1974) who have emphasized the
relationship between motor requirements and cardiac activity.
According to Obrist et al., there is a positive correlation between
changes in cardiac activity and changes in levei of somatic activity
and both are controlled by integrative mechanisms in the central
nervous system (cardiac-somatic coupling). Obrist et al. (1974) also
noted that instances occur in which the cardiac-somatic coupling is
dissociated, whereby increases in heart rate (termed cardiac
preparatory responses) are observed without related overt changes in
somatic activity. Interestingly, this occurs specifically in
situations related to active avoidance of aversive stimuli.

35
Similarly, Freyschuss (1970) has observed heart rate acceleration when
subjects are instructed either to tense or move an arm even though
such movement is impossible because of experimentally induced
paralysis. These observations suggest that cardiac activity is not
solely coupled with overt somatic activity per se, but that cardiac
activity is coupled with real as well as intended somatic activity.
While orienting and defensive responses result in
characteristically distinct patterns of autonomic responding, they
have both been conceptualized as arousal responses. However, this
view presents difficulty for the view of arousal as a unidimensional
phenomenon. It also provides some support for the notion that
autonomic reactivity is not as uniform as once suggested (Cannon,
1927; Schacter & Singer, 1962).
Ax (1953) provided some of the first evidence to suggest that
various affective states may be associated with distinct autonomic
patterning. Ax reported that diastolic blood pressure increased more
during anger than fear imagery, while heart rate and systolic blood
pressure increased with equal magnitude. More recent studies have
replicated these findings (Schacter, 1957; Weerts & Roberts, 1976).
Schwartz, Weinberger, and Singer (1981) have recently reported
cardiovascular differentiation between imagery induced happiness,
sadness, fear, and anger. Several investigators have also found
significantly greater heart rate accelerations in response to fearful
stimuli such as mutiliation slides, spiders, and fearful imagery (Hare
& 31evings, 1975; Klorman & Ryan, 1980; Klorman, Weissberg, &
Weisenfeld, 1977; Vrana, Cuthbert, & Lang, 1986). Recently, Ekman,

36
Levenson, and Friesen (1983) have also reported heart rate increases
in response to production of emotional facial expressions of anger,
fear, and sadness and heart rate decreases in response to disgust,
surprise, and happiness.
In summary, there is some evidence to suggest that autonomic
arousal is not as uniform as once suggested. In fact, several studies
support the notion that different patterns of autonomic arousal may be
associated with different types of emotional states. In addition,
recent conceptualizations of heart rate arousal responses have
suggested that such changes may be linked to overt and/or covert
motoric responses. This view is also consistent with the bio-
informational theory proposed by Lang (1979). Lang proposes that
emotional imagery results in patterns of autonomic activity very
similar to those found in the actual emotional situation. This raises
the possibility that certain emotions, by virtue of their strong motor
components, may be more associated with heart rate acceleration while
others with less motor demands may be more associated with heart rate
deceleration.
Neuropsychological Models of Emotional Processing
Several different neuropsychological models of emotional
processing have been proposed to account for findings of
investigations in brain impaired and neurologically intact subjects.
The models of Fox and Davidson (1984), Kinsbourne (Kinsbourne &
Bemporad, 1984), Tucker (1981), and Heilman (Heilman et al. 1983;
Heilman, 1988, personal communication) will be briefly reviewed.

37
The Model of Fox and Davidson
Fox and Davidson (1984) have proposed a developmental model of
neural substrates underlying affective response systems. Their model
is based on evolutionary considerations of the critical role of basic
approach and avoidance systems. These authors propose that approach
and avoidance comprise the two underlying behavioral dimensions upon
which all subsequent affective subsystems and responses have
evolved. In addition, they suggest that hemispheric specialization
constitutes the critical, neuroanatomical substrate of approach-
avoidance behavior. More specifically, these authors propose that the
left hemisphere is specialized for approach behaviors or positive
affects while the right hemisphere is more specialized for avoidance
behaviors or negative affects.
In support of tnis model, these authors relate the development of
hemispheric specialization and interhemispheric transfer to the
development of affective response systems. These authors argue that
all of the primary emotions emerge over the first year of life
(Bowlby, 1972; Charlesworth, 1 964; Izard, Hubner, Risser, McGuiness, &
Dougherty, 1980; Sroufe & Wunsch, 1972; Steinberg & Campos, 1983).
Subsequent to this, they propose that primary emotions are modified by
three processes: (a) the addition of new behaviors to the response
repertoire of the "affect program," (b) regulation in the form of
inhibition and appraisal, and (c) blending of primary emotions.
Interest and disgust are reliably elicited in the neonate (Izard,
1977), and Fox and Davidson (1984) cite evidence from EEG findings in
infants to suggest that these emotions are lateralized. They found

38
more interest expressions and greater relative left-sided EEG activity
following sucrose administrations compared with citric acid
administrations in infants and argued that these findings provide
support for left hemisphere superiority in processing of positive
affect. They further proposed that these emotions are under
unilateral hemispheric control since little functional interconnection
between the hemispheres exists at birth.
Through the course of development, changes in interhemispheric
communication are proposed as the necessary substrate for emergence of
fear and sadness in the emotional repertoire. In support of this, the
authors cite evidence that the onset of locomotion, a behavior
associated with commissural transfer, is tightly coupled to the
emergence of fear (Bayley, 1963; 1969; Rader, Bausano, & Richards,
1980). In addition, the authors argue that the expression of sadness
is often associated with alternating sequences of approach and
avoidance, again implicating a critical role for interhemispheric
communication (Ainsworth, Blehar, Waters, & Wall, 1978; Izard,
1977). The capacity to inhibit negative affective responses, which
emerge during the second year, are also presumed to be linked to the
functional integrity of the commissural system. In addition, these
authors propose that the left hemisphere normally exerts an inhibitory
influence on the right hemisphere through transcallosal connections,
resulting in attenuation of negative affect in the normal state.
Kinsbournes Model
Kinsbourne and Bemporad (1984) proposed a separate model also
based on the behavioral dichotomy that he terms action-approach and

39
inaction-avoidance. This model, however, invokes both anterior-
posterior and lateral specialization of cerebral processing.
Posterior regions are proposed to provide data necessary to maintain
homeostasis and anterior regions are proposed to exert control to
stabilize as needed. In addition, anterior systems are viewed as
exerting inhibitory control over its posterior source of
information. As in the conceptualization of Fox and Davidson, this
model also proposes a left hemispheric approach and right hemispheric
avoidance dichotomy. However, in Kinsbourne's model the left
hemisphere is specialized for processing of external change and
ongoing action (approach) and the right hemisphere is specialized for
processing of internal changes, interruption of action (avoidance) as
well as control of emotional arousal. This conceptualization
establishes, in effect, four quadrants providing different mechanisms
of processing:
Left-Anterior
Left-Posterior
Right-Anterior
Right-Posterior
Action control over external change
Enables left-anterior action control to make
contact with necessary exteroceptive information
Emotional control over internal arousal
Enables right-anterior emotional control to make
contact with interoceptive information
With regard to the two control systems (action control and emotional
control), Kinsbourne argues that the two hemispheres are not in
inhibitory, but rather in compensatory interaction. Furthermore,
whether the control is under left hemisphere-approach or right
hemisphere-avoidance processing depends on the stimulus circumstances

40
and on the status of the organism's attempt to exert control over its
environment or itself.
Kinsbourne provides a very interesting model of emotional
processing which is based, to some extent, on the approach-avoidance
model of Fox and Davidson and further elaborated to include an
anterior-posterior dimension. However, less evidence exists to
support or refute his claims.
Tuckers Model
Tucker (1981) has also suggested a neuropsychological model of
emotional processing based on lateralized neuroanatomical systems.
These systems control (a) tonic activation and motor readiness and (b)
phasic arousal responses to perceptual input. In contrast to notions
of reciprocal inhibition of the hemispheres, Tucker invokes mechanisms
of subcortical release of lateralized arousal systems to account for
left-negative versus right-positive valence findings.
According to Tucker, the left hemisphere is specialized for
activation and complex motor operations. Support for this is provided
by the preponderance of right hand motor dominance in the general
population as well as observed deficits in both right and left hand
production of learned, skilled motor movements (apraxias) following
left hemisphere lesions (Geschwind, 1975). The presumed neurochemical
substrate for this specialization is the dopamine system which several
investigations suggest is predominantly a left lateralized system
(Click, Meibnach, Cox, & Maayani, 1979; Wagner et al., 1983). Tucker
argues that activation operates in a tonic fashion to increase
informational redundancy. This view is supported by the observation

41
that increased dopaminergic activity is associated with restriction of
behavioral output by the production of motor stereotypies in humans
and other animals (Ellinwood, 1967; Iversen, 1977). Tucker also cites
evidence from psychiatric literature to suggest that negative emotions
of anxiety as well as ritualized, stereotyped behaviors associated
with obsessive-compulsive disorder and, to some extent, left partial
complex seizure disorder represent subcortical release and subsequent
overactivity of this left hemisphere activation system.
In contrast, Tucker proposes a right hemispheric specialization
for phasic arousal responses to perceptual input which exerts its
control through habituation. The presumed neurochemical substrate for
this specialization is the norephinephrine system which some
investigations have suggested is represented to a greater extent in
the right hemisphere (Oke, Keller, Mefford, 4 Adams, 1978; Oke, Lewis,
& Adams, 1980). Lateralized norephinephrine pathways are known to
show a pattern of widespread distribution throughout the brain
providing the necessary substrate for arousal responses and
facilitation of orienting to novelty. In support of this, Tucker
notes greater right hemisphere ability in tasks of "global" versus
"local" processing, requiring integration of perceptual input (Levy,
1969; Nebes, 1974). In addition, he cites evidence from the
psychiatric literature suggesting greater right hemisphere involvement
in hysteric personalities euphoric and often indifferent emotional
responses which appear analogous to the responses of right hemisphere
damaged patients (Galin, Diamond, & Braff, 1977; Gur & Gur, 1975;
Smokier & Shevrin, 1979).

42
Tucker suggests that the two hemisphere's differing modes of
processing may be the primary factor in lateralized valence effects
reported in the literature. He hypothesizes that what we experience
as emotions arise from operation of these arousal and attentional
modulatory processes.
Heilman's Model
From investigations of indifference reaction associated with
right hemisphere damage and unilateral neglect syndrome, Heilman and
colleagues (1983) have suggested a model of emotional processing based
on hemispheric differences in arousal-activation responses. Heilman
has suggested that right hemisphere damaged patient's difficulties in
emotional expression may be a result of (a) deficits in arousal-
activation and (b) an ability to develop an appropriate cognitive
state due to basic deficits in comprehension of prosodic elements of
speech and affective facial expressions. Patients with indifference
reaction often have the unilateral neglect syndrome in which they may
fail to orient, report, or respond to stimuli in the contralateral
side of space (Denny-Brown et al 1952; Gainotti, 1972; Heilman &
Valenstein, 1972). Heilman et al. have suggested that unilateral
neglect is a defect in attenuation-arousal-activation due to
disruption of a corticolimbic-reticular loop (Heilman & Van Den Abel,
1979). Based on the fact that neglect occurs most often following
right hemisphere damage, he has proposed that the right hemisphere may
be dominant for mediating attention-arousal-activation responses.
To investigate arousal responses in brain impaired patients,
Heilman et al. (1978) stimulated the forearm ipsilateral to the side

43
of lesion in RHD and LHD patients while recording galvanic skin
response from the same side. The authors noted that RHD patients had
significantly smaller GSR arousal responses than LHD patients or
nonbrain damaged control subjects. Similarly, Morrow, Urtunski, Kim,
and Boiler (1981) presented LHD and RHD patients with neutral and
emotional stimuli. Right hemisphere patients showed decreased
galvanic skin responses to both neutral as well as emotional stimuli
relative to LHD patients.
More recently, Yokoyama, Jennings, Ackles, Hood, and Boiler
(1987) have looked at heart rate and reaction time responses in
patients with right unilateral hemispheric lesions. These authors
found that RHD patients had significantly slower reaction times and
decreased heart rate responses (both deceleratory as well as
acceleratory) relative to LHD patients. These findings indicate that
the greater role of the right hemisphere in attention may be reflected
in both reaction time as well as anticipatory heart rate changes.
Investigations in neuroiogically intact subjects corroborate
these findings. Hugdahl, Franzon, Anderson, and Walldebo (1983)
report greater anticipatory heart rate accelerations for emotional
stimuli presented to the LVF (right hemisphere) compared with RVF
(left hemisphere) trials. Similarly Walker and Sandman (1982) also
report greater right hemisphere activity (as measured by the P100
component of the average evoked potential) when the heart was
spontaneously accelerated. In addition, Hugdahl, Wahlgren, and Wass
(1982) found delayed habituation of the electrodermal orienting

44
response to visual stimuli initially projected to the right
hemisphere.
Heilman and Van Den Abel (1979) have also suggested a right
hemisphere superiority for activation. Using a neutral warning
stimulus paradigm, these authors reported that warning stimuli
projected to the right hemisphere reduced reaction times of the right
hand more than warning stimuli projected to the left hemisphere. In
addition, warning stimuli projected to the right hemisphere reduced
reaction times of the right hand more than warning stimuli projected
to the left hemisphere reduced reaction times of the left hand. Based
on these findings, it can be seen that warning stimuli projected to
the right hemisphere reduced reaction times of both hands to a greater
extent than left hemisphere presentations, suggesting that the right
hemisphere was better able to activate responses in both hands
relative to the left hemisphere.
In addition, Verfaellie, Bowers, and Heilman (1987) reported a
study of neurologically intact subjects which provides some support
for the right hemisphere dominance of activation. By manipulating
preliminary intentional warning cues (which hand to use in
responding), they found faster left hand versus right hand responses
suggesting that the left hand (right hemisphere) was better able to
benefit from this preparatory information than the right hand (left
hemisphere). In support of this, Verfaellie and Heilman (1987) also
report the performance of two patients with right and left
supplementary motor area (SMA) damage on this paradigm. They report
that the patient with left SMA damage (intact right hemisphere) was

45
able to benefit from preparatory information wnile the patient with
right SMA damage was unable to benefit from preparatory information,
again suggesting a greater role of the right hemisphere in activation
of response.
More recently, Heilman (1988, personal communication) has
suggested that emotion specific hemispheric effects (i.e., left
hemisphere-positive, right hemisphere-negative) reported in the
literature may be artifactual and actually represent hemispheric
differences in arousal and preparation for action. Because the right
hemisphere is dominant for mediating arousal/activation, it may
therefore be more intrinsically involved in processing emotional
stimuli that have greater "preparatory" significance for survival
(i.e., fight-flight emotions such as anger and fever). In contrast,
the left hemisphere may be more involved in mediating nonpreparatory
emotions (i.e., happiness, sadness, disgust) that place less immediate
or "phasic" attentional demands on the organism for survival.
In summary, recent investigations in brain impaired and
neurologically intact subjects suggests a greater role of the right
hemisphere in arousal-activation responses. Furthermore, this finding
is observed across several indices of arousal-activation including
heart rate, skin conductance, and reaction time measures.
Critical Issues
As reviewed in the introduction, there appears to be general
consensus that the two hemispheres in man differ in terms of their
contribution to emotional processing. However, the precise role

46
played by each remains unclear. Some investigators have argued that
the right hemisphere is globally involved in ail aspects of emotional
processing including the cognitive encoding/decoding of emotional
stimuli, arousal-activation responses to emotional stimuli and
behavioral responses to these stimuli (Heilman et al., 1983; bey &
Bryden, 1979). Other investigators have argued that the two
hemispheres differ in terms of the type of emotions that are
preferentially mediated by each (Fox St Davidson, 1984; Kinsbourne Sc
Bemporad, 1984; Tucker, 1981). The most popular version of this view
i3 that the left hemisphere is dominant for positive (approach)
emotions, whereas the right hemisphere is dominant for negative
(avoidance) emotions.
Still others have argued that this positive-negative dichotomy in
hemispheric processing of emotions is artifactual and actually relates
to hemispheric differences in arousal and preparation for action
(e.g., activation) (Heilman, 1988, personal communication). In this
view, the right hemisphere is dominant for mediating
arousal/activation and as such, is more intrinsically involved in
processing emotional stimuli that have greater "preparatory"
significance for survival (i.e., fight-flight emotions such as anger
and fear). In contrast, the left hemisphere is more involved in
mediating nonpreparatory emotions (i.e., happiness, sadness, disgust)
that place less immediate or "phasic" attentional demands on the
organism for survival.
In order to distinguish among these models, it would be necessary
to determine whether different categories of emotional stimuli result

47
in differential patterns of arousal/activation, depending on the
hemisphere to which they were presented. According to the global
right hemisphere emotion model, emotional stimuli of any valence
directed to the right hemisphere should result in greater
arousal/activation responses than those directed to the left
hemisphere. According to valence models, negative emotional stimuli
(anger, fear, disgust) would induce greater arousal/activation
responses when directed to the right versus left hemisphere, whereas
the opposite should occur when positive stimuli (happiness) are
used. Finally, according to the preparatory/nonpreparatory model,
emotional stimuli having preparatory significance (anger, fear) should
result in greater arousal/activation responses when directed to the
right versus left hemisphere. The opposite should occur with
nonpreparatory emotional stimuli (happiness, disgust, neutral).
The focus of the present study was to further examine these
divergent views regarding the hemispheric processing of emotional
stimuli. The basic paradigm was one in which neutral and emotional
stimuli of different valence were laterally presented to either the
left or right hemisphere (using a tachistoscopic procedure). The
purposes of this study were to determine (a) the extent to which
laterally presented emotional/nonemotional stimuli might result in
differential patterns of behavioral activation (as assessed by
reaction time responses) as well as differential patterns of autonomic
arousal (as assessed by HR and SCR responses); (b) whether there were
hemispheric asymmetries in mediating arousal and/or activation in
response to these emotional/nonemotional stimuli; and (c) whether

48
certain categories of emotion (positive/negative; preparatory/
nonpreparatory) induce asymmetric arousai/activation, depending on the
hemisphere to which they are initially presented.
In order to address these issues, two experiments were
completed. In the first study, laterally presented emotional stimuli
of different valences served as warning stimuli to the subjects who
then made manual RT responses to a neutral midline stimulus. This
warning stimulus paradigm was chosen because it enables one to
determine the extent to which lateralized emotional warning stimuli
serve to behaviorally activate and prepare the individual to respond
to a subsequent stimulus (Lansing, Schwartz, & Lindsey, 1959). In the
second study, laterally presented emotional stimuli were also shown to
subjects, and autonomic indices of arousal (HR, SCR) were measured.
Although it would have been more "ideal" to obtain both autonomic and
RT measures to the lateralized emotional stimuli in the same study,
this wa3 not realistically feasible. The "3low" rise time of the SCR
(2-4 seconds) in conjunction with the relatively short lived
activating effects of warning stimuli (500-2000 msec) precluded such a
direct manipulation. Thus, two separate experiments were conducted.
Four different emotional categories were chosen for the present
investigation. Two categories which have previously been shown to
result predominantly in cardiac deceleratory responses (happy,
disgust) and two categories which have previously been shown to result
predominantly in cardiac acceleratory responses (fear, anger) (Ekman
et al., 1983). Due to the relative paucity of discernable positive
emotions among the wide range of emotional categories (Ekman, 1972),

49
it was not possible to equate occurrence of positive versus negative
emotions.
Hypotheses and Predictions
According to the right hemisphere emotion model, the right
hemisphere plays a greater role than the left hemisphere in mediating
arousal/activation responses to emotional materials. If this model is
correct, then one would predict faster RT responses to a midline
neutral stimulus when it is preceded by an emotional warning stimulus
(WS) directed to the right hemisphere (LVF) than by an emotional WS
directed to the left hemisphere (RVF). Additionally, RTs should also
be faster when emotional WS versus nonemotional WS are directed to the
right hemisphere (LVF). These predictions were examined in Experiment
I. Similarly, one would also predict that autonomic responsivity to
emotional versus nonemotional stimuli should be greater when they are
directed to the right hemisphere (LVF) versus the left hemisphere
(RVF). These predictions were examined in Experiment II.
According to hemispheric valence models of emotional processing,
negative emotional stimuli are preferentially mediated by the right
hemisphere, and positive emotional stimuli are mediated by the left
hemisphere. If this hypothesis is correct, then one would predict
that negative emotional WS directed to the right hemisphere (LVF)
should result in faster RTS to a neutral midline stimulus than when
the negative WS is directed to the left hemisphere (RVF). Conversely,
positive emotional WS directed to the left hemisphere (RVF) snould
result in faster RTs than positive WS directed to the right hemisphere

50
(LVF). Similarly, in Experiment II, one would predict greater
autonomic responsivity (HR acceleratory and deceleratory responses,
SCR) to negative stimuli that are directed to the right hemisphere
(LVF) versus stimuli that are directed to the right hemisphere (LVF)
versus those directed to the left hemisphere (RVF). The opposite
pattern of autonomic arousal should occur for positive emotional
stimuli.
According to the hemispheric preparatory model of emotion, the
right hemisphere is dominant for mediating emotional stimuli that have
a greater preparatory significance for survival (i.e., fight-flight
emotions such as anger and fear). In contrast, the left hemisphere is
dominant for mediating nonpreparatory stimuli that* place less "phasic"
demands on the individual for immediate survival (i.e., happiness,
disgust, neutrality). If this model is correct, then in the first
experiment one would predict that anger and fear WS directed to the
right hemisphere (LVF) should result in faster RTs than when they are
directed to the left hemisphere (RVF). Conversely, happy, disgust,
and neutral WS should result in faster RTs when they are directed to
the left versus right hemisphere. Likewise, if one assumes that
behavioral activation and autonomic responsivity are strongly coupled,
then similar predictions would be made for Experiment II. That is,
one would predict greater autonomic responsivity (HR acceleratory and
deceleratory responses) to anger and fear stimuli (assumed to be
preparatory) when they are directed to the right versus left
hemisphere. The opposite hemispheric pattern of autonomic arousal

51
should occur for neutral, disgust, and happy stimuli (assumed to be
nonpreparatory).
Alternatively, it is possible that the right hemisphere may be
dominant for mediating arousal/activation responses to stimuli,
regardless of their emotional-nonemotional content. In this view,
stimuli directed to the right hemisphere should result in greater
arousal/activation responses than stimuli directed to the left
hemisphere. However, any differences in arousal/activation responses
to emotional versus neutral stimuli should be comparable across the
left and right hemispheres. Thus, in Experiment I, WS directed to the
right hemisphere (LVF) should result in faster RTs than WS directed to
the left hemisphere (RVF). Any RT differences between emotional
versus nonemotional WS should be comparable for LVF and RVF
presentations. Likewise, in Experiment II, one would predict greater
autonomic responsivity (HR, SCR) to stimuli directed to the right
versus left hemisphere. Again, however, any differences in arousal
responses to emotional versus neutral stimuli should be comparable
across LVF and RVF presentations of the stimuli.

METHOD
Subj ects
A total of 60 (30 male, 30 female) students at the University of
Florida served as subjects (Ss) in the present investigation.
Subjects were given either course credit or paid for their
participation. All Ss were right handed according to self-report and
their performance on the Briggs and Nebes (1975) Handedness
Questionnaire. An overall score of +9 or above (right hand
preference) was used as the criterion for participation in this study.
Different groups of 30 Ss each (15 male, 15 female) participated
in Experiment I and II. Subjects were randomly assigned to Experiment
I or Experiment II. Due to the potential confounding effects of
prior exposure to the emotional stimuli on subsequent RT and
psychophysiological responses, a between groups comparison was felt to
be advantageous. The mean age of Ss in Experiment I was 20.7 with a
range of 18-26 years. The mean age of Ss in Experiment II was 21.0
with a range of 19-27 years.
Experiment I: Reaction Time Task
Stimuli
Stimuli consisted of 192 black and white slides. These slides
depicted 96 neutral and 96 emotional scenes which included 24 slides
52

53
of each of four different valences (happy, angry, fearful, and
disgusting). The emotional scenes were selected from a variety of
materials including magazines and photography books. The scenes used
in this experiment did not include familiar landmarks or personalities
in an effort to avoid possible confounding effects of familiarity on
RT and HR responses to these stimuli. The stimuli were rated for type
and intensity of affect by 20 (10 male, 10 female) University of
Florida students who did not participate in the present experiments.
All stimuli averaged at least 91% agreement. Mean intensity ratings
(on a scale of 1 to 5) for the five categories were happy, 3.4; angry,
3.6; fearful, 3.9; disgusting, 3.8; and neutral, .3.
Apparatus
Slides were projected onto a 40x35-cm Kodak milk-glass, rear view
projection screen. A 5-mm diameter red light-emitting diode (LED) was
placed at the center of the screen to serve as a central fixation
point and imperative stimulus in the RT task. Two spring loaded keys
were placed 30 cm to the left and right of body midline. The timing,
presentation of stimuli, and recording of Ss' responses from release
of spring loaded keys were accomplished by an I3M-PC microcomputer
interfaced with BRS logic. Slides were projected at 5 degrees of
visual angle lateral to the central fixation LED. Attached to the
projector lens were Uniblitz 325B high speed shutters which allowed
maximum rise and fall time of slide presentation. The room was dimly
lit to avoid the effects of visual startle during stimulus
presentation.

54
Eye movements were continuously monitored by electro-oculography
(EOG) to ensure maintained fixation and lateralized presentation of
slides. The EOG signal was detected by Beckman Ag/AgCl miniature
electrodes attached at the temporal canthus of the left and right
eye. The EOG signal was filtered and fed into a DC amplifier and
recorded on a Grass Model 78B polygraph. Event marker input from
IBM-PC microcomputer was also fed into the polygraph recording to
identify occurrence of slide presentation and to facilitate subsequent
identification of eye movements during this interval.
Procedure
Subjects were seated 91 cm from the projection screen with left
and right hands placed on left and right sided keys, respectively.
The task was a choice reaction time task. Subjects viewed a laterally
presented warning stimulus (neutral or emotional scene) of 500 msec.
This was followed 500, 1000, or 1500 msec later by a centrally
presented neutral, imperative stimulus (red LED) with interstimulus
interval randomly varied across trials. Half of the Ss were
instructed to release the left key following onset of the LED if the
preceding warning stimulus was a neutral scene or the right key if the
warning stimulus was an emotional scene. The remaining Ss were
instructed to release the left key following onset of the LED if the
preceding warning stimulus was an emotional scene and the right key if
the preceding warning stimulus was a neutral scene. Half-way through
the experiment, hand order was reversed for all subjects. Subjects
received a total of 192 trials. Response hand and visual field of

55
presentation were randomized and counterbalanced across stimulus
type. Subjects were given 10 practice trials prior to the experiment.
Experiment II: Psychophysiological Responses to
Laterally Presented Emotional Material
Stimuli
The stimuli were identical to those used in Experiment I. They
included 24 neutral and 96 emotional scenes [24 of each of four
differing valences (happy, angry, fearful, disgusting)].
Apparatus
Slides were projected onto a 40x35-cm Kodak milk-glass, rear view
projection screen. A 5-mm diameter adhesive circle was placed at the
center of the screen to serve as a central fixation point. Slides
were projected at 5 degrees of visual angle lateral to the central
fixation point. Lafayette Model #43016 shutters were attached to the
projector lens to maximum the rise and fall time of slide
presentation. The timing and presentation of stimuli were
accomplished by IBM-PC microcomputer interfaced with BRS logic. Eye
movements were continuously monitored in the same fashion as
Experiment I. The room was again dimly lit to avoid the effects of
visual startle during stimulus presentation.
Psychophysiological measures (HR, SCR, and respiration depth)
were recorded for the 3 seconds prior to stimulus onset and 8 seconds
of stimulus presentation. Heart rate responses were recorded by two
Beckman Ag/AgCl electrodes attached to the right and left lateral
margins of the chest. A third electrode was attached to the

56
sternum. Electrode sites were prepared by mild abrasion of the skin
with Hewlett Packard Redux paste. Electrodes were fastened by the use
of adhesive collars and were filled with Hewlett Packard Jel Redux
cream as the electrolyte. The electrocardiogram (ECG) was amplified
by a Colbourn S75-03 high gain bioamplifier. This signal was input to
a Coibourn S75-38 Ban-Pass Biofilter with subsequent detection of the
R-wave component which interrupted the computer to provide inteibeat
intervals. A-D conversion was accomplished by Colbourn R65-17 Data
Translation Board and signal was downloaded to an IBM-PC interfaced
with data acquisition modules.
Skin conductance was recorded with Met-Associates electrodes
placed on the thenar and hypothenar eminences of the left and right
hands. Electrode sites were wiped clean with distilled water.
Electrodes were attached by the use of adhesive collars filled with KY
jelly. The analog SC signal was fed into a Colbourn Model S71-22 Skin
Conductance Module. This signal was digitized by a Colbourn R65-17
Data Translation Board and the signal was downloaded to an IBM-PC
microcomputer interfaced with the data acquisition modules.
Respiration depth was also recorded to detect and subsequently
exclude those trials in which unusually large inhalations or
exhalations may have confounded HR or SCR. Respiration depth was
measured by means of a Colbourn Model T41 -91 Aneroid Chest Bellows,
and processed by a Colbourn S72-25 Module. A-D conversion and
computer interface utilized the same equipment as HR and SCR measures.

57
Procedure
Prior to the experiment, Ss were requested to list specific
episodes from their own lives which they considered happy, angry,
fearful, disgusting, or neutral. Subjects were requested to list five
episodes for each category for a total of 25.
The experiment took place in a quiet, dimly lit room. Subjects
were seated in a comfortable, recliner chair positioned 152 cm from
the projection screen. After electrode placement and a 20-minute
adaptation period, Ss were instructed to refrain from any unnecessary
movement during the experiment.
Subjects were presented with a neutral or emotional slide in left
or right visual field for 8 seconds. Subjects were instructed to
maintain fixation on the centrally positioned circle throughout slide
presentation. To decrease Ss' habituation and encourage continued
processing of the stimuli during this time, Ss were also instructed to
recall one of the five episodes of the same valence as the slide
presented. Approximately 10 seconds after slide offset, Ss were asked
to indicate the valence of the slide and responses were recorded
manually on a separate sheet. Inter-trial interval varied randomly
from 23 to 45 seconds to minimize occurrence of anticipatory HR and
SCR. Subjects received a total of 120 trials with visual field of
presentation randomized and counterbalanced across stimulus type.
Subjects also received 10 practice trials for maintained fixation
prior to the experiment utilizing neutral stimuli only.

RESULTS
Experiment I: Reaction Time Task
Reaction Time Responses
Subjects' reaction time responses served as the data for analysis
in a repeated measures analysis of variance (ANOVA). The between
subjects factor was Sex (male, female). The within subject factors
were Visual Field (left, right), Hand (left, right), and Stimulus Type
(happy, angry, fearful, disgusting, neutral) as within subjects
factors. Incorrect trials and trials in which eye movements occurred
were not included in the analysis. The remaining 81 of trials served
as the data for analyses. This included 83* of happy trials, 82% of
angry trials, 82% of fearful trials, 74 of disgusting trials, and 85%
of neutral trials. A log transformation was used to correct for
skewness in data distribution and to adequately meet homogeneity of
variance assumptions for the ANOVA (Kirk, 1968; Winer, 1971).
A summary of the results of this ANOVA are depicted in Table 1 .
Findings revealed a main effect of Sex [_F (1, 28) = 4.44, £ = .0442)]
with mean RT responses of male (M = 481.20) Ss significantly faster
than RT responses of female (14 = 603.48) Ss, depicted in Figure 1. A
significant Sex x Visual Field effect [F_ (1, 28) = 4.06, £ = .0535)]
was also obtained and is depicted in Figure 2. Post-hoc simple
effects testing was performed to clarify the nature of this
interaction (Kirk, 1978; Winer, 1971). Findings revealed no
58

59
Table 1
Summary of Analysis of Variance: Experiment I, Reaction Time
Source
df
SS
F
P
Sex
1 ,
28
6.0277
4.44
#
Visual Field
1,
28
.0016
.25
-
Visual Field x Sex
1,
28
.0267
4.06
*
Hand
1 ,
28
.0375
OO
-
Hand x Sex
1 ,
28
.0032
.07
-
Stimulus Type
4,
112
.0874
1.38
-
Stimulus Type x Sex
4,
112
.1674
2.64
#
Visual Field x Hand
1.
28
.0000
.01
-
Visual Field x Hand x Sex
1 ,
28
.0008
.05
-
Visual Field x Stimulus Type
4,
112
.0334
35
-
Visual Field x Stimulus Type x Sex
4,
11 2
.0966
1 .02
-
Hand x Stimulus Type
4,
1 12
.1497
L*J
CD
-
Hand x Stimulus Type x Sex
4,
112
.0386
36
-
Visual Field x Hand x Stimulus Type
4,
1 12
.1995
2.49
#
Visual Field x Hand x Stimulus Type
x Sex
4,
112
.0357
.45
"
*p < .05.

60
£
c
o
o
03
cc
Happy Angry Fearful Disgusting Neutral
Emotion
Figure 1. Experiment I--Reaction Time Analysis, Sex Main Effect,
Figure 2. Experiment I--Reaction Time Analysis, Sex x Visual Field
Interaction.

61
significant differences across LVF and RVF trials for female Ss [F_
(1, 14) = 3.71, £ = .0644)]. However, for male Ss, RVF (fi = 477.10)
were significantly faster than LVF (M = 485.49) trials [ F_ (1 14) =
5.23, £ = .0299)].
Findings also revealed a significant interaction of Sex x
Stimulus Type [F (4, 112) = 2.64, £ = .0376)], depicted in Figure 3.
Post-hoc simple effects testing revealed no significant differences
across Stimulus Type for female Ss [F (4, 56) = 1.26, £ = .294)].
However, male Ss showed a significant effect for Stimulus Type [F_
(4, 56) = 2.60, p = .0454)]. Duncan's post-hoc comparisons revealed
that for male Ss, RT responses to disgusting slides (_M = 506.28) were
significantly slower than happy (M = 468.29) or angry = 483.^7)
sLides at £ < .05.
A significant Hand x Visual Fieid x Stimulus Type interaction [F_
(4, 112) = 2.49, £ = .0473)] was also obtained and is depicted in
Figure 4. Post-hoc simple effects testing of RVF trials only revealed
no statistically significant differences across Hand and Stimulus Type
conditions [F_ (4, 116) = .97, £ = .4249)]. In LVF, however, post-hoc
simple effects tests revealed a significant Hand x Stimulus Type
interaction [F (4, 116) = 2.61, £ = .0390)]. Duncan's post-hoc
comparisons revealed that happy trials in LVF were significantly
faster when using the right (_M = 507.45) versus left (fl = 562.94) hand
at £ < .05.
Further inspection of the data revealed that the significant Sex
effects observed in the preceding analysis may have been influenced by
the markedly slowed performance of two female Ss. When compared to

Reaction Time
62
(a)
(b)
Figure 3- Experiment I--Reaction Time Analysis, Sex x Stimulus Type
Interaction: (a) males only and (b) males and females.

ReactionTime
63
580 -
560 *
540 -
520 -
500 -
560
550
540
530
520
Figure 4.
Left Visual Field Trials
557.91
Happy Angry Fearful Disgusting Neutral
Emotion
(a)
Right Hand
Left Hand
Right Visual Field Trials
Experiment I--Reaction Time Analysis, Hand x Visual Field x
Stimulus Type Interaction: (a) left visual field trials
and (b) right visual field trials.

the overall mean RT of females for each of the 20 Visual Field x Hand
x Stimulus Type conditions, these two Ss possessed 8 (40$ of total)
and 10 (50$ of total) mean reaction times which fell two standard
deviations above the overall mean performance of female Ss. Of the
remaining 13 female Ss, no S possessed a single mean reaction time
greater than 2 standard deviations above the overall mean of female
Ss.
For this reason, a second analysis of the RT data was conducted
in which the data from the two female Ss noted above was excluded. A
summary of the results of this analysis are depicted in Table 2.
Findings revealed no significant main effects. The main effect of Sex
observed on the preceding analysis was no longer significant
suggesting that this effect may have been significantly influenced by
the markedly slowed performance of the two female Ss noted above.
Findings, however, did reveal a significant Visual Field x Sex
interaction [F (1, 26) = 5.20, £ < .0311)] as in the preceding
analysis, depicted in Figure 5. Simple effects testing revealed a
pattern of findings similar to those noted in the first analysis with
no significant difference between LVF and RVF performance for female
Ss. Males, however, performed significantly faster to RVF stimuli
relative to LVF stimuli. A trend for the Sex x Stimulus Type
interaction was also noted [F (4, 104) = .0722, £ = .0722)] which had
previously attained significance in the first analysis. This
interaction is depicted in Figure 6. The Hand x Visual Field x
Stimulus Type interaction significant, in the first analysis, failed
to reach significance in the present analysis.

65
Table 2
Summary of Analysis of Variance
i: Experiment
I,
Reaction
Time (minus
outliers)
Source
df
SS
F
£
Sex
1 ,
26
2.3201
2.17
-
Visual Field
1.
26
.0002
.03
-
Visual Field x Sex
1 .
26
.0332
5.20
#
Hand
1,
26
.0441
1.15
-
Hand x Sex
1 ,
26
.0060
.16
-
Stimulus Type
4,
104
.1194
1.96
-
Stimulus Type x Sex
4,
104
.1350
2.22
-
Visual Field x Hand
1.
26
.0019
.12
-
Visual Field x Hand x Sex
1 ,
26
.0000
.00
-
Visual Field x Stimulus Type
4,
1 04
.0137
.20
-
Visual Field x Stimulus Type x
Sex
4,
104
.0699
.76
-
Hand x Stimulus Type
4.
104
.1714
1.59
-
Hand x Stimulus Type x Sex
4,
104
.0357
33
-
Visual Field x Hand x Stimulus
Type
4,
104
.1514
1.86
-
Visual Field x Hand x Stimulus
Type
x Sex
4,
104
.0362
. 45
p < .05.

Reaction Time r Reaction Time
66
600 -
550 -
500 -
450 -
400 -
552.16

556.34
485.49
477.10

Left
Right
Visual Field
Females
Males
;ure 5.
Experiment I--Reaction Time Analysis (minus outliers),
Visual Field x Sex Interaction.
Happy Angry Fearful Disgusting Neutral
Emotion
Females
Males
Figure 6. Experiment IReaction Time Analysis (minus outliers), Sex
x Stimulus Type Interaction.

67
Percent Correct Responses
A separate ANOVA was conducted which used proportion of correct
responses for each S as the dependent variable. An arcsin square root
transformation was performed to correct for lack of a normal
distribution inherent in proportion data and to meet homogeneity of
variance assumptions for the ANOVA (Kirk, 1968; Winer, 1971).
Results of this analysis utilizing ail 30 Ss are depicted in
Table 3. Findings revealed only a significant main effect of Stimulus
Type [F (4, 112) = 7.60, £ = .0001)], depicted in Figure 7. Duncan's
post-hoc comparisons revealed that percent correct identification of
disgusting (M = .804) trials was significantly worse than all four
remaining categories (happy, = .907; angry, = .904; fearful, M =
.903; neutral, = .938) at £ < .05. In addition, a trend for
Stimulus Type x Visual Field [F (4, 112) = 2.29, £ = .064)] was also
revealed, depicted in Figure 8. This pattern of findings suggest that
happy trials were more accurate in the LVF while neutral trials were
more accurate in the RVF.
A second ANOVA was conducted for percent correct data which
excluded the two female Ss noted above. Results of this analysis are
depicted in Table 4. This analysis revealed only a significant main
effect of Stimulus Type [F (4, 104) = 6.13, £ = .0002)], depicted in
Figure 9. Duncan's post-hoc comparisons revealed that percent correct
identification of disgusting (_M = .810) trials was significantly worse
than all four remaining categories (happy, M = .911; angry, M = .904;
fearful, M = .899; neutral, M = .940) at £ < .05. No other effects
reached significance or trend status.

68
Table 3
Summary of Analysis of Variance: Experiment I, Percent Correct
Source
df
SS
F
P
Sex
1,
28
.8665
2.28
-
Visual Field
1,
28
.0181
.22
-
Visual Field x Sex
1,
28
.1683
2.01
-
Hand
1,
28
.0757
1.23
-
Hand x Sex
1,
28
.0203
33
-
Stimulus Type
4,
112
2.5844
7.60
#*
Stimulus Type x Sex
4,
112
.2799
.82
Visual Field x Hand
1,
28
.0009
.01
-
Visual Field x Hand x Sex
1 ,
28
.0004
.01
-
Visual Field x Stimulus Type
4,
112
.4369
2.29
-
Visual Field x Stimulus Type x Sex
4,
112
.1908
1 .00
-
Hand x Stimulus Type
4,
1 12
.4237
1.44
-
Hand x Stimulus Type x Sex
4,
112
.1018
.35
-
Visual Field x Hand x Stimulus Type
4,
1 12
.0793
.45
-
Visual Field x Hand x Stimulus Type
x Sex
4,
112
.1774
1.02
"
#*
p < .01.

Percent Correct ^ Percent Correct
69
0.95
0.85
0.75
Lgure 7.
100
90
80
70
Figure 8.
.938
1 1 1 1 1
Happy Angry Fearful Disgusting Neutral
Emotion
Experiment I--Percent Correct Analysis, Stimulus Type Main
Effect.
Trend
.957
i i i i i
Happy Angry Fearful Disgusting Neutral
Emotion
RVF
LVF
Experiment I--Percent Correct Analysis, Stimulus Type x
Visual Field Trend.

70
Table 4
Summary ofAnalysis of Variance: Experiment I, Percent Correct (minus
outliers)
Source
df
ss
F
P
Sex
1 ,
26
.9794
2.42
-
Visual Field
1,
26
.0009
.01
-
Visual Field x Sex
1 ,
26
.0552
.87
-
Hand
1,
26
.0498
.76
-
Hand x Sex
1 ,
26
.0091
.14
-
Stimulus Type
4,
104
2.0531
6.13
**
Stimulus Type x Sex
4,
104
.2958
.88
-
Visual Field x Hand
1 ,
26
.0205
.27
-
Visual Field x Hand x Sex
1 ,
26
.0180
.24
-
Visual Field x Stimulus Type
104
.3694
1.86
-
Visual Field x Stimulus Type x
Sex
4,
104
.2289
1 .15
-
Hand x Stimulus Type
4,
104
. 3254
1 .16
-
Hand x Stimulus Type x Sex
4,
104
.1308
.47
-
Visual Field x Hand x Stimulus
Type
4,
104
.0452
.28
-
Visual Field x Hand x Stimulus
x Sex
Type
4,
104
. 0866
.54
-
** p < .01 .

Percent Correct
71
Figure 9- Experiment I--Percent Correct Analysis, Stimulus Type Main
Effect.

72
Experiment II
Heart Rate Data Reduction
Heart rate responses were edited by a computer program which
converted all trials from interbeat interval data to beats per minute
format for each second of the sampling period. Each interbeat
interval was weighted proportionally to the fraction of the second it
occupied according to method recommended by Graham (1980). Baseline
for each trial was defined as the average HR for the 2 seconds
preceding stimulus onset. For each trial, baseline HR was then
subtracted from each of eight post-stimulus HR values, yielding second
by second post-stimulus HR changes from baseline. Trials in which eye
movements occurred were not included in the analysis. The remaining
77% of trials served as the data for analysis. This included 73? of
happy trials, 77? of angry trials, 73$ of fearful trials, 75? of
disgusting trials, and 78? of neutral trials.
From this, two separate data sets were generated. The first,
selected, for each trial, was the maximum deceleratory HR response
from the eight post-stimulus second by second HR changes from
baseline. This maximum deceleratory response then served as the
dependent variable in a repeated measures ANOVA. The second data set
selected, for each trial, was the maximum acceleratory HR response
from the eight post-stimulus second by second HR changes from
baseline. This maximum acceleratory response also served as the
dependent variable in a separate repeated measures ANOVA.

73
Maximum deceleratory responses
A repeated measures ANOVA was performed with Sex (male, female)
as a between subjects factor and Visual Field (left, right), and
Stimulus Type (happy, angry, fearful, disgusting, neutral) as the
witnin subjects factors. A summary of the results of this ANOVA are
depicted in Table 5. Results of this analysis revealed a main effect
of Sex [F (1, 28) = 5.16, p = .0310)]. Males (M = -9.153) showed
significantly greater maximum deceleratory HR responses than females
(M = -7.756). These findings are depicted in Figure 10. Results also
revealed a trend for a main effect of Visual Field [F_ ( 1, 28) = 3-33*
p = .0789)] with a pattern of greater deceleratory HR responses to LVF
trials (M. = -8.7OO) relative to RVF trials (_M = 8.209) (see Figure
11).
This analysis also yielded a main effect of Stimulus Type [F
(4, 112) = 4.96, p = .0010)]. Duncan's post-hoc comparisons revealed
happy slides (£ = -9.016) elicited significantly greater HR
decelerations than angry (N1 = -7.969) or fearful (M = -7.663) slides
at £ < .05. Neutral slides (M. = -8.984) also elicited significantly
greater HR decelerations than angry or fearful slides at p < .05.
These relationships are depicted in Figure 12.
Findings also revealed a significant Visual Field x Sex
interaction [F_ ( 1, 28) = 7.86, £ = .0091 )]. Simple effects testing
revealed that males showed significantly greater deceleratory HR
changes in LVF (M = -9.776) relative to RVF (M = -8.530) trials,
depicted in Figure 13- Results also revealed a significant Sex x

74
Table 5
Summary of Analysis of Variance: Experiment II, Heart Rate
Deceleration
Source
df
SS
F
P
Sex
1 ,
28
146.4789
5.16
*
Visual Field
1.
28
18.0850
3-33
-
Visual Field x Sex
1 ,
28
42.7167
7.86
*#
Stimulus Type
4,
1 12
89.6031
4.96
#*
Stimulus Type x Sex
4,
112
60.4664
3-35
**
Visual Field x Stimulus Type
4,
1 12
7.7819
.47
-
Visual Field x Stimulus Type x Sex
4,
112
12.6660
.77
-
*p < .05. ** p < .01.

Maximum Heart Rate Deceleration
75
Figure 10. Experiment IIMaximum Deceleratory Heart Rate Analysis,
Sex Main Effect.
Figure 11. Experiment II--Maximum Deceleratory Heart Rate Analysis,
Visual Field Trend.

Maximum Heart Rale Decelerlalion '3. Maximum Heart Rate Deceleration
from Baseline c from Baseline
76
Emotion
12. Experiment II--Maximum Deceleratory Heart Rate Analysis,
Stimulus Type Main Effect.
Visual Field
Figure 13. Experiment IIMaximum Deceleratory Heart Rate Analysis,
Visual Field x Sex Interaction.

77
Stimulus Type interaction [F (4, 112) = 3-35, £ = .0125)]. Simple
effects testing revealed no significant effects across Stimulus Type
for males [£ (4, 56) = 1.86, £ = .130)]. However, for female subjects
this Stimulus Type effect was significant [F (4, 56) = 5-32, £ =
.001)]. Duncan's post-hoc comparisons revealed that deceleratory HR
responses to happy = -8.306), disgusting (>4 = -8.3^5) and neutral
(M = -8.638) were all significantly greater than deceleratory HR
responses to angry (_M = -6.120) slides at £ < .05. These
relationships are depicted in Figure 14.
Maximum acceleratory responses
A repeated measures ANOVA was performed with Sex (male, female)
as the between subjects factor and Visual Field (left, right) and
Stimulus Type (happy, angry, fearful, disgusting, neutral) as the
within subjects factors. A summary of the results of this analysis
are depicted in Table 6. Results revealed a significant interaction
of Stimulus Type x Sex [F (4, 112) = 3-48, £ = .0102)], depicted in
Figure 15. Simple effects testing revealed no significant effect of
Stimulus Type for males [F_ (4, 56) = 1.86, £ = .130)]. For female Ss,
however, this effect was significant [F (1, 14) = 5.32, £ = .001)].
Duncan's post-hoc comparisons revealed that angry (14 = 3-^73) and
fearful (_M = 3-888) slides elicited significantly greater HR
accelerations than neutral (_M = 2.158) slides at p < .05.
This analysis also revealed a trend for an interaction between
Visual Field x Sex [F (1, 28) = 3-36, £ = .0776)], with a pattern of
greater acceleratory HR responses for males in LVF (M = 2.531) trials

Maximum Heart Rate Deceleration
(rom Baseline
78
Emotion
Figure 14. Experiment II--Maximum Deceleratory Heart Rate Analysis,
Sex x Stimulus Type Interaction.

79
Table 6
Summary of Analysis of Variance: Experiment II, Heart Rate
Acceleration
Source
df
SS
F
P
Sex
1 ,
28
28.2236
.65
-
Visual Field
1.
28
.521 4
.13
-
Visual Field x Sex
1 ,
28
13-2770
3-36
-
Stimulus Type
1,
1 12
28.6964
1.74
-
Stimulus Type x Sex
112
57.5180
3.48
**
Visual Field x Stimulus Type
4,
1 12
13.3162
.73
-
Visual Field x Stimulus Type x Sex
4,
112
9.4844
.52
-
* *
p < .01.

Maximum Heart Rale Acceleration Maximum Heart Rale Acceleration
80
Happy Angry Fearful Disgusting Neutral
Emotion
Males
Females
Figure 15. Experiment II--Maximum Acceleratory Heart Rate Analysis,
Stimulus Type x Sex Interaction.
Figure 16. Experiment IIMaximum Acceleratory Heart Rate Analysis,
Visual Field x Sex Interaction.

81
relative to RVF (M = 2.027) trials. This pattern of findings is
depicted in Figure 16.
Second by second responses
Second by second post-stimulus HR changes for male and female Ss
are displayed for each Visual Field x Stimulus Type condition in
Figures 17 and 18, respectively. These values represent the average
change from baseline for each of the eight po3t-stimulus seconds.
Post-stimulus HR changes from baseline for both male and female Ss
suggest that, overall, Ss showed primarily HR deceleratory changes
throughout the presentation of the stimuli. This may be accounted
for, in part, by task factors which required perceptual intake
throughout the presentation of the stimulus. Perceptual intake has
been associated with HR deceleration as suggested by Lacey (1967).
It is also of note that post-stimulus HR changes from baseline do
not reflect occurrence of acceleratory HR responses (i .e., positive HR
change from baseline). It may be that the emotional stimuli
themselves were not inherently "strong enough to elicit acceleratory
HR responses. However, findings from the analysis of maximum
acceleratory HR responses do indicate that, at some point during the 8
post-stimulus seconds, Ss are experiencing HR acceleration. The most
plausible explanation for the lack of post-stimulus HR accelerations
in graphic representation of the 8 post-stimulus seconds is that these
acceleratory responses are occurring at different points in time
across different trials and possibly across different S3. That is,
accelerations may be "averaged out" by decelerations occurring at the
same time on separate but like trials (i.e., same VF x Stimulus Type

BPM Irom Baseline BPM from Baseline
82
Males
Post-Stimulus Seconds
Males
Post-Stimulus Seconds
Males
Post-Stimulus Seconds

c
Q
U1
01
CQ
E
o
2
CL
CD
Males
C
*3
a
CQ
e
o
2
Q-
CQ
Males
Post-Stimulus Seconds
Figure 17. Experiment II--Second x Second Post-Stimulus Heart Rate
Changes in Male Subjects for Each Visual Field x Stimulus
Type Condition.

BPM from Baseline BPM from Baseline
33
Females
Post-Stimulus Seconds
Females
Post-Stimulus Seconds
Females -o- LVF-Fearful
RVF-Fearful
Post-Stimulus Seconds

c


ra
CO
E
o
2
0.
CQ
Females
Post-Stimulus Seconds

C
UJ
ra
CO
E
o
i
0.
CO
o -
-1 -
-2-
-3-
-4 -
-5 -
Females
-Q-
LVF-Neutral
RVF-Neutral
TD (T
I i i 1 I 1 I I I I 1
D1 D2 D3 D4 D5 D6 D7 D8
Post-Stimulus Seconds
Figure 18. Experiment lI--Second x Second Post-Stimulus Heart Rate
Changes in Female Subjects for Each Visual Field x
Stimulus Type Condition.

84
condition). In the present experiment, Ss were required to also image
a personal episode of the same valence as the stimulus. The time
course of the Ss' imaging could not be controlled and it may be the
case that Ss differed in onset of their imaging and occurrence of
affective aspects of the image.
One observations which is of some interest is the fact that
female Ss do show some indications of "autonomic patterning" in their
HR changes across the 8 post-stimulus seconds. That is, females
appeared to show less deceleration for fearful trials relative to
other affective categories. Furthermore, this pattern appears to be
reflected to a greater extent for LVF (right hemisphere) than RVF
(left hemisphere) presentations.
Skin Conductance Data Reduction
For each trial, SCRs were depicted as the difference between the
average of the skin conductance level during the 2 seconds preceding
stimulus onset (tonic, baseline level) and the maximum skin
conductance level during the 8 post-stimulus seconds (phasic level).
This SCR value for each trial served as the dependent variable in a
repeated measures ANOVA. Sex was the between subjects factor and
Visual Field (left, right), Hand (left, right), and Stimulus Type
(happy, angry, fearful, disgusting, neutral) were the within subjects
factors.
Results of this ANOVA yielded no significant main effects or
interaction effects. Table 7 depicts a summary of this analysis. A
weak trend, however, was observed for a Sex x Visual Field x Stimulus

85
Table 7
Summary of Analysis of Variance: Experiment II, Skin Conductance
Responses
Source
df
SS
F
P
Sex
1,
28
.6410
2.60
-
Visual Field
1,
28
.001 1
.08
-
Visual Field x Sex
1,
28
.0004
.03
-
Hand
1,
28
.0056
.13
-
Hand x Sex
1 ,
28
.0399
.94
-
Stimulus Type
112
.0167
36
-
Stimulus Type x Sex
11 2
.0628
1 .35
-
Visual Field x Hand
1,
28
.0007
.19
-
Visual Field x Hand x Sex
1,
28
.0018
.43
-
Visual Field x Stimulus Type
4,
1 12
.0664
1.61
-
Visual Field x Stimulus Type x
Sex
4,
1 1 2
.0868
2.10
-
Hand x Stimulus Type
4,
1 12
.0186
1.31
-
Hand x Stimulus Type x Sex
4,
1 1 2
.0097
.69
-
Visual Field x Hand x Stimulus
Type
4,
112
.0249
1.59
-
Visual Field x Hand x Stimulus
x Sex
Type
4,
1 12
.01 12
.71
-

86
Type interaction [F (4, 112) = 2.10, £ = .0875)], depicted in Figure
19. This pattern of findings revealed for male Ss only a Visual Field
x Stimulus Type interaction which approached significance. This
pattern suggested greater SCR in LVF (M = .1302) relative to RVF (fi =
.0651) for angry slides. In addition, happy 3lides elicited greater
SCR in RVF (M = .1268) compared to LVF (M = .0736) trials.

Skin Conductance Response
87
0.14-
0.12-
0.10-
0.08 -
0.06-
Figure 19.
Trend
Emotion
Experiment II--Skin Conductance Response Analysis, Sex x
Visual Field x Stimulus Type Trend.

DISCUSSION
Several models have been proposed to account for lateral
symmetries observed on tasks of emotional processing. One model, the
right hemisphere model, suggests that the right hemisphere is globally
more involved in all aspects of emotional processing including the
cognitive encoding/decoding, arousal-activation, and behavioral
responses to emotional stimuli. In this view, emotional stimuli
presented initially to the right hemisphere via the left sensory
channel (left visual field, left ear) elicit significantly greater
arousal responses and result in significantly quicker/more accurate
detection than emotional stimuli presented initially to the left
hemisphere via the right sensory channel (right visual field, right
ear) .
A second model, the hemispheric valence model, proposes that the
left hemisphere is more adept at processing positive emotions and the
right hemisphere is more adept at processing negative emotions.
Within this framework, positive emotional stimuli initially presented
to the left hemisphere would elicit significantly greater arousal
responses and result in significantly quicker/more accurate detection
than positive emotional stimuli presented initially to the right
hemisphere. Likewise, negative emotional stimuli presented to the
right hemisphere would elicit significantly greater arousal responses
88

89
and result in significantly quicker/more accurate detection than
negative emotional stimuli presented initially to the left hemisphere.
A third model, the preparatory model, argues that this positive
negative dichotomy in hemispheric processing of emotions is
artifactual and actually relates to differences in arousal and
activation/preparation for action. According to this model, the right
hemisphere is dominant for mediating arousal/activation and as such,
is more intrinsically involved in processing emotional stimuli that
have greater "preparatory" significance for survival (i.e., fight-
flight emotions). In contrast, the left hemisphere is more involved
in mediating nonpreparatory emotions that place less immediate
attentional demands on the organism for survival.
Lastly, it is possible that the right hemisphere is dominant for
mediating arousal/activation responses to stimuli irrespective of
their emotional/nonemotional content. That is, both emotional and
nonemotional stimuli should result in significantly greater arousal/
activation responses when projected to the right hemisphere than the
left hemisphere.
The present study sought to further investigate these different
conceptualizations of the hemispheric processing of emotional
stimuli. The purposes of the study were to (a) determine the extent
to which laterally presented emotional/nonemotional stimuli might
result in differential patterns of behavioral activation (reaction
time responses) as well as differential patterns of autonomic arousal
(HR, SCR responses); (b) to determine whether there were hemispheric
asymmetries in mediating arousal and/or activation responses to these

90
emotionai/nonemotional stimuli; and (c) to determine whether certain
categories of emotion (positive/negative; preparatory/nonpreparatory)
induce asymmetric arousal/activation, depending on the hemisphere to
which they are initially presented.
Findings from Experiment I, in which RTs were made to midline
neutral stimuli that were preceded by lateralized stimuli of different
emotional valences, failed to support any of the laterality models of
emotion. For example, no overall superiority for lateralized
emotional warning stimuli presented to the right versus left
hemisphere was found. Likewise, no hemispheric specific emotional
valence effects were observed. Similarly, no evidence was present for
the view of hemispheric differences in "preparatory" versus
"nonpreparatory" emotions. What was found, however, was the
following: (a) females showed no laterality effects of any kind; and
(b) males, on the other hand, had overall faster RTs to neutral
stimuli that were preceded by emotional warning stimuli in the RVF
(left hemisphere) versus LVF (right hemisphere). This finding, which
suggests that emotional stimuli induce greater behavioral activation
when presented to the left hemisphere than to the right hemisphere, is
the opposite of that predicted by any model arguing for superiority of
the right hemisphere in mediating emotional responsivity.
There are several possibilities which might account for these
findings. In Experiment I, Ss were required to make a left-right
decision based on the emotionai/nonemotional nature of the warning
stimulus. That is, they had to respond with one hand to emotional
stimuli and with the other hand to neutral stimuli. In other words,

91
Ss had to make a left-right discrimination judgement during the
interstimulus interval between the WS and the imperative stimulus. It
is well known that left-right discrimination seems to fall within the
domain of left hemisphere functions (Benton, 1968; Gertsmann, 1940;
Saugeut, Benton, & Hecaen, 1971). Consequently, it is possible that
this left-right discrimination inherent in the task demands may be
related to the finding of faster RTs for male Ss when stimuli were
presented to the left hemisphere versus the right hemisphere. The use
of a go/no paradigm using a single hand for response would circumvent
this possible confounding factor.
Secondly, a variety of task strategies and stimulus factors can
significantly influence the magnitude or even the direction of
perceptual asymmetries (Moskovitch, 1986). Factors such as stimulus
duration, spatial frequency, stimulus clarity, and number and
configuration of stimulus features can significantly influence
observed lateral asymmetries (Bryden, 1978; Bryden & Allard, 1976;
Moskovitch, 1983; Sergent, 1983; Sergent & Bindra, 1981). For
example, Patterson and Bradshaw (1975) have reported that decreasing
the dimensions along which facial stimuli differed decreased and even
reversed the expected LVF superiority on a simultaneous face matching
task.
It may be the case that in the present investigation the stimuli
were sufficiently complex to warrant a change in processing strategy
which relied perhaps to a greater degree on left hemispheric analysis
of details and significant features in distinguishing among the
different categories of stimuli. If this is the case, then the

92
significant processing demands of the warning stimulus may have
mitigated any potential activation or preparatory effect that the
emotional stimuli may have had because processing of the warning
stimulu continued through the interstimulus interval. The use of less
complex warning stimuli such as emotional and neutral faces is one way
of testing this hypothesis.
In addition, it may al30 be of informational value to use such
warning stimuli in a simple reaction time task to look at the general
activation effect of emotional and neutral stimuli on simple reaction
time to a neutral imperative stimuli. This paradigm would serve to
eliminate the processing demands of the warning stimulus which may
have interfered with the potential activation effect of emotional
warning stimuli on right hemisphere processing.
Lastly, it may be the case that emotional warning stimuli used in
the present investigation were not arousing enough to provide adequate
activation effects in response to the imperative stimulus.
Findings the second experiment, in which measures of autonomic
arousal were obtained to laterally presented emotional/nonemotional
stimuli, were more in line with current views regarding hemispheric
differences in processing emotionaL stimuli. With SCRs, there were no
significant effects in terms of SCRs to lateralized emotional
stimuLi. However, a trend was observed (Sex x VF, p = .087), whereby
male Ss had greater SCRs to happy stimuli when they were presented to
tne RVF (left hemisphere) than to the LVF (right hemisphere); greater
SCRs occurred when angry stimuli were presented to LVF (right
hemisphere) than to the RVF (left hemisphere). Although this pattern

93
of findings is consistent with both the valence model and the
preparatory model of emotional processing, these findings must be
viewed with caution since significance was not obtained. The lack of
robust findings may be due, in part, to the fact that SCRs for Ss were
generally quite small, making detection of differences across visual
field, hand, and stimulu type conditions difficult.
With regard to HR responses that were also measured in Experiment
II, happy and neutral slides induced significantly greater HR
deceleratory responses than angry or fearful slides. This finding
occurred regardless of which hemisphere initially received the
stimulus. In other words, no emotional specific hemispheric arousal
effects were present for HR deceleratory responses. Additionally, a
trend was observed for greater HR deceleratory responses for stimuli
directed to the right hemisphere (LVF) versus those directed to the
left hemisphere (RVF). This finding was based primarily on the
responses of male Ss, as indicated by a significant Sex x VF
interaction. Specifically, males had significantly larger HR
deceleratory responses to LVF stimuli than to RVF stimuli, and this
effect was present regardless of the emotional/nonemotional content of
the stimuli. In contrast, females Ss displayed no laterality effect
of any kind in their HR deceleratory responses to emotional or neutral
stimuli. A similar pattern of findings was also observed for HR
acceleratory responses, as reflected in a trend for a Sex x VF
interaction (£ = .077). Male Ss again tended to produce greater HR
acceleratory responses to LVF stimuli than to RVF stimuli.

94
In contrast to the hemispheric asymmetries in arousal responses
observed in males, females did not show significant differences in HR
arousal responses to stimuli across left and right visual field.
However, females were differentially impacted by the emotional valence
of the stimuli. This was revealed as a Sex x Stimulus interaction in
analyses of HR deceleratory and HR acceleratory responses. Males did
not show this differential pattern of responding to the emotional
valence of the stimuli. Specifically, females had significantly
greater heart rate deceleratory responses for happy, disgusting, and
neutral trials compared to angry and fearful trials. Similarly,
females Ss had significantly greater HR acceleratory responses to
angry and fearful stimuli relative to happy and neutral stimuli.
Taken together, findings from both HR deceleratory and
acceleratory responses, that were obtained in Experiment II, suggest
that greater HR arousal responses occurred when stimuli were presented
to the right hemisphere (LVF) than when presented to the left
hemisphere (RVF). This laterality effect occurred only for male Ss
and was not dependent on the emotional/nonemotional content of the
stimuli. That is, emotional and neutral stimuli directed to the right
hemisphere resulted in comparable HR arousal effects. This finding is
not consistent with the view that the right hemisphere is specifically
dominant for mediating arousal responses only to emotional stimuli.
Nor is it consistent with valence or preparatory models of emotional
processing. Rather, this finding suggests that the right hemisphere
of males is dominant for mediating arousal/activation, regardless of

95
the emotional/nonemotional content of the stimuli and this right
hemisphere dominance is not specific to emotional stimuli.
Critical Issues
Several issues are raised by these findings. The first concerns
why there were no differences in arousal responses to emotional versus
neutral stimuli. Other investigators have found that emotional
stimuli induced greater arousal responses than do neutral stimuli
(Hare & 31evings, 1975; Klorman et al., 1977; Klorman & Ryan, 1980).
One possibility accounting for the lack of differences in arousal
responses to emotional versus neutral stimuli is that the emotional
stimuli exerted some sort of "priming" effect, such that even the
neutral stimuli were treated as "emotional." Kinsbourne (1970; 1973)
has suggested that adoption of a cognitive set can serve to
asymmetrically arouse or prime the hemispheres for processing. It may
have been the case in this study that emotional stimuli exerted a
priming effect in which neutral stimuli were treated as emotional
stimuli.
A second issue concerns why the laterality effect in HR responses
were observed only in male Ss. In a major review of the literature on
sex differences, McGlone (1980) concludes that overall males possess a
significantly greater degree of observed asymmetry relative to female
subjects. While this conclusion has been criticized on a number of
grounds, the weight of the evidence does suggest that, at least for
language functions, the hemispheric representation of males may differ
from that of females. That is, on measures of severity of aphasia and

96
recovery from aphasia following hemispheric lesions, it appears that
males are more left hemisphere lateraiized for language, whereas
females have more bilateral representation of speech and language
functions (McGlone, 1980).
With respect to arousal asymmetries, studies have not
systematically addressed sex differences. Prior investigations which
have looked at the right hemisphere's role in production of arousal
responses in normal, neurologically intact Ss have not investigated
the effect of sex on production of laterality effects in HR and SCR
(Hugdal et al., 1983; Hugdahl et al., 1982; Walker & Sandman, 1982).
Heart rate findings from the present investigation suggest that the
greater role of the right hemisphere in production of arousal
responses may exist for male Ss to a greater extent than for female
Ss; findings which are consonant with neuropsychological
investigations of laterality effects across sex.
In addition to studies which have looked at the generation of
autonomic arousal (i.e., HR, SCR), a related area of research has also
investigated cerebral asymmetries in heart beat perception and
detection. Davidson, Horowitz, Schwartz, and Goodman (1981) measured
RT differences between R-wave occurrence and key press latencies.
These authors report that finger taps by the left hand (right
hemisphere) had shorter mean latencies from heartbeat than did taps
from the right hand (left hemisphere), suggesting that the right
hemisphere is more sensitive or more efficient than the left
hemisphere in response tracking to cardiac afferent feedback.

97
Similarly, Hantas, Katkin, and Reed (1984) using male subjects
only demonstrated that right hemisphere preferent individuals (as
indexed by conjugate lateral eye movements) performed significantly
better than left hemisphere preferent individuals on a task of heart
beat detection. Furthermore, these findings were significant both
before and after heart rate discrimination training. Montgomery and
Jones (1984) reported similar findings in male subjects, with right
hemisphere preferent subjects performing significantly better on a
heart beat perception task relative to left hemisphere preferent
subjects. In addition, they also found higher emotionality scores for
right hemisphere preferent individuals relative to left hemisphere
preferent individuals.
It is of interest to note that these studies which find
significant laterality effects on tasks of heart beat perception and
detection have used primarily male subjects. This occurrence may, in
part account, for significant laterality effects observed on these
tasks as male subjects often reveal a significantly greater degree of
cerebral lateralization than female subjects (McGlone, 1980).
A consistent finding in the cardiac awareness literature is that
male subjects are superior to female subjects in detecting heart beats
and heart rate across a variety of psychophysiological paradigms
(Pennebaker & Hoover, 1984; Whitehead, Drescher, Heiman, & Blackwell,
1977). More recent research (Rouse, Jones, 4 Jones, 1988), however,
has suggested that such sex differences in cardiac awareness are
accounted for primarily by body fat composition {% fat) and general
fitness differences across male and female subjects. By controlling

98
for level of body fat, Rouse et al. (1988) did not find the gender
effect previously documented in the literature. Furthermore, sex
differences were found only when level of body fat differed between
males and females.
In the present investigation, a main effect of Sex was observed
in Experiment II for maximum HR deceleratory responses. More
specifically, male subjects had significantly greater HR decelerations
than female subjects. It may be the case that body fat composition
and general fitness also plays some role in magnitude of cardiac
decelerations although this relationship is unclear at present.
To briefly summarize, findings from Experiment II provide support
for the greater role of the right hemisphere in mediating arousal
responses, at least for male subjects. That is, males showed greater
HR deceleratory and acceleratory responses to stimuli when they were
presented to the LVF (right hemisphere) tnan to the RVF (left
hemisphere). This concerns the basis for these sex differences.
The evolutionary significance of lateralized heart rate responses
in male subjects can be viewed from the more general perspective of
the evolutionary significance of greater cerebral lateralization in
males. Flor-Henry (cited in McGlone, 1930) provides an interesting
argument in this regard. He notes that dopaminergic pathways in
rodents are known to be asymmetrical and related to directional
preferences and turning behaviors with the direction of these
behaviors contralateral to the hemisphere with greatest dopamine
concentrations (Glick, Jerussi, & Zimmer berg, 1977). Turning
behaviors in rodents, cats, and dogs are associated with fighting and

99
sexual display. Furthermore, he notes that the lateralized control of
song in the left brain of canary and chaffinch is related to mate
attraction and territoriality, i.e., sex and spatial analysis
(Nottebohm, 1977; Webster, 1977).
Mach (1959) suggests that this neural asymmetry serves to
increase the efficiency of spatial analysis in an environment where no
systematic left/right bias exists. Furthermore, he argues the
evolutionary advantage of this increased efficiency is related to mate
attraction and in species where the male actively seeks the female,
this greater lateralization for male organisms is of adaptive
significance and survival value.
As previously discussed, female subjects did not show significant
differences in HR deceleratory responses or HR acceleratory responses
across left and right visual fields as found in male subjects.
However, female subjects were differentially impacted by the emotional
valence of the stimuli (revealed as a significant interaction of Sex x
Stimulus Type for analysis of both HR deceleratory responses and HR
acceleratory responses). Male subjects, in contrast, did not show
this differential heart rate responding across stimulus categories.
Analysis of deceleratory HR responses for female subjects revealed
significantly greater decelerations for happy, disgusting, and neutral
trials relative to fearful trials. Similarly, analysis of
acceleratory HR responses for female subjects revealed significantly
greater acceleratory responses for angry and fearful slides compared
with neutral slides. These results are consistent with those of other

100
investigators who report HR accelerations in response to fearful
stimuli (Hare & Blevings, 1975; Klorman & Ryan, 1980; Vrana et al.,
1986). In addition, these results also parallel the findings of Ekman
et al. (1983) who reported HR increases in response to production of
emotional facial expressions of anger, fear, and sadness and HR
decelerations in response to disgust, surprise, and happiness.
While previous research has suggested that different patterns of
autonomic arousal are associated with different types of emotional
states, in the present study differences in autonomic responses across
different emotional categories were observed only for female
subjects. One critical question concerns why this "autonomic
patterning" to emotional stimuli should occur in females but not
males. It may be the case that female subjects were more amenable to
fully cooperating with task demands of emotional imagery. This
suggestion is supported by prior investigations which have found
significant effects for mood manipulation and emotional imagery
instruction for female but not male subjects (Delp & Sackeim, 1987;
McKeever & Dixon, 1981).
Alternatively, sex differences in autonomic patterning observed
in the present study may be related to differences in imagery ability
between males and females. Prior investigations have revealed that
differences in autonomic control and HR conditioning are related to
subject's imagery ability. Carroll, Baker, and Preston (1979) have
reported that ability to increase HR through voluntary imaging was
significantly correlated with reported vividness of the subject's
images. Ikeda and Hirai (1976) reported that ability to control SCR

101
through biofeedback training was associated with the frequency of
subject's images. In addition, Kuzendorf (1982) also found that
subject's ability to produce temperature differences across the hands
was significantly correlated with prevalence of imagery. Lastly,
Arabian and Furedy (1983) report that HR deceleratory conditioning
(conditioned response) in good imagers was more similar to HR
deceleratory unconditioned responses than HR deceleratory conditioning
(conditioned responses) for poor imagers.
In the present investigation, imagery ability was not assessed.
However, it is possible that the current findings which suggest that
female subjects showed differntial autonomic responding across various
emotional stimuli might be related to potential differences in imagery
ability between female and male subjects. For example, there is some
evidence in the literature to suggest that female subjects, in
general, are better imagers than male subjects (White, Sheehan, &
Ashton, 1977). This finding has led some investigators to use only
female subjects in investigations of imagery and autonomic responding
(Jones & Johnson, 1978; 1980).
Regardless of the basis for the sex differences, the differential
pattern of HR responses to the various emotional stimuli by female
subjects in the present study is similar in pattern to that reported
by other investigators (Ekman et al., 1983; Hare & Blevings, 1975;
Klorman & Ryan, 1980). Findings from the present investigation
suggest that HR decelerations were significantly greater for both
happy and neutral slides relative to angry and fearful slides.

102
Similarly, angry and fearful slides elicited significantly greater HR
accelerations than neutral slides.
A question which arises concerns the basis for the differential
HR patterning across different emotion categories. Recently, several
authors have systematically examined the relationship between
emotional imagery and production of cardiac responses. Jones and
Johnson (1978; 1980) have argued that two major factors contribute to
the particular patterning of HR responses. These include (a) the
emotional content of the image (pleasant vs unpleasant); and (b) the
level of activity inherent in the image (high activity vs low
activity). By systematically manipulating these two factors, Jones
and Johnson (1980) found that high activity images produced
significantly greater HR accelerations than low activity images.
These findings are congruent with reports from other investigators who
have noted that instructional manipulations on imagery tasks can
produce significant effects on autonomic responses (Lang, 1977;
Melamed, 1969). Similarly, by requesting active engagement in
stimulus processing, Bauer and Craighead (1979) also revealed
significant effects on psychophysiological responses.
Jones and Johnson (1980) also report that negative-low activity
images resulted in greater HR accelerations relative to positive-low
activity images. Such findings suggest that the emotional valence of
the image was also a significant contributing factor to differential
HR responses. While the effect of valence of affective stimulus on HR
responding had been previously investigated, these authors provide

103
support for the role of this factor when differences in activity level
of imagery are accounted for.
It may be the case that a priori differences in activity level
across emotional categories could, in part, account for differential
patterns of HR acceleration across emotional categories. Lang (1979)
has suggested that emotional imagery results in patterns of autonomic
activity very similar to those found in the actual emotional
situation. In this view, those emotions which inherently contain a
greater motor component (i.e., flight or fight type of responding) as
in the case of anger or fear, would result in significantly greater HR
acceleratory responses than those emotions with less inherent motor
demand; predictions which parallel the findings in the present
investigation.
It may also oe the case that these changes in HR activity across
various emotional categories may serve to differentially prepare the
organism for action. In accounting for differences in acceleratory
and deceleratory HR responses, Obrist and colleagues (197^) 'nave
emphasized the relationship between motor requirements and cardiac
activity. In addition, these authors suggest that conditions may
exist whereby increases in HR are observed without overt changes in
somatic activity. In support of this view, Freyschuss (1970) reported
cardiac accelerations under conditions where subjects were instructed
to tense or move an arm despite their inability to do so because of
experimentally induced paralysis. Based on these findings, it is
suggested that cardiac activity is not solely coupled with direct,
overt somatic activity but rather that cardiac activity is coupled

with real as well as intended somatic activity (as in the case of
imagery).
One additional question concerns the fact that findings from
Experiment II provided support for the greater role of the right
hemisphere in production of arousal responses but results from
Experiment I failed to provide support for a greater role of the right
hemisphere in production of arousal responses. Differences across the
two experiments may help to clarify some of this apparent
discrepancy. First, the most plausible explanation for these
discrepant findings is that in Experiment I subjects were required to
make a left-right decision based on the emotional/nonemotional nature
of the warning stimulus. As previously discussed, it is likely that
this left-right discrimination (a predominantly left hemisphere
ability) inherent in the task demands of Experiment I is related to
the finding of faster RTs for male Ss when stimuli were presented to
the RVF (left hemisphere).
Secondly, exposure durations in the two experiments differed
quite significantly. Experiment I utilized relatively short exposure
durations while Experiment II utilized quite lengthy exposure
durations. The overall complexity of the stimuli together with the
relatively short exposure durations used in Experiment I versus
Experiment II may have contributed to these divergent findings.
Similarly, the task demands of the two experiments were quite
different. Requirements inherent in Experiment II in which subjects
were asked to generate very personal and likely very meaningful
episodes from their own lives may have contributed to the overall

105
effectiveness of this procedure in demonstrating differential and
lateralized heart rate responses in female and male Ss, respectively.
Conclusions
Findings from Experiment I, in which RTs were made to midline
neutral stimuli that were preceded by lateralized stimuli of different
emotional valences, failed to support any of the laterality models of
emotion. No overall superiority for RTs to LVF (right hemisphere)
versus RVF (left hemisphere) trials was found. In addition, no
evidence was found for hemispheric specific emotional valence
effects. Likewise, no evidence was present for the view of the
hemispheric differences in preparatory versus nonpreparatory
emotions. Rather, faster RTs were observed for male subjects when WS
were presented to RVF (left hemisphere). This finding is not
consistent with any of the proposed hemispheric models of emotional
processing. The most plausible interpretation of these findings is
that subjects were required to make a left-right decision based on the
emotional/nonemotional nature of the warning stimulus and that this
left hemisphere mediated task requirement accounts for the faster RTs
of male subjects to RVF (left hemisphere) presentations of WS.
Findings from Experiment II, in which measures of autonomic
arousal (HR and SCR) were obtained to laterally presented
emotional/nonemotional stimuli, were more congruent with models of
hemispheric differences in processing emotional stimuli. Males showed
lateralized effects of HR arousal responses which supports greater
right hemisphere involvement in production of arousal responses.

106
However, this effect was present regardless of the emotional/
nonemotional content of the stimulus. This finding is not consistent
with the view that the right hemisphere is specifically dominant for
mediating arousal responses only to emotional stimuli. Nor is it
consistent with valence or preparatory models of emotional
processing. Rather, this finding suggests that the right hemisphere
of males is dominant for mediating arousal responses, regardless of
the emotional/nonemotional content of the stimulus and this right
hemisphere dominance is not specific to emotional stimuli.
However, SCRs did suggest some findings which are consistent with
emotion specific hemispheric arousal effects. This was revealed in a
trend for a Sex x VF interaction (£ = .087) effect, in which male
subjects had greater SCRs to happy stimuli when they were presented to
the RVF (left hemisphere); greater SCRs occurred when angry stimuli
were presented to LVF (right hemisphere) than to RVF (left
hemisphere). These findings are also consistent with the preparatory
model of emotional processing.
The question is raised why emotion specific hemispheric arousal
effects are found for SCR but not for HR responses. Prior
investigations have revealed that SCR are more related to intensity of
emotional stimuli whereas HR responses were more related to valence of
emotional stimuli (Greenwald, Lang, & Cooke, 1988). It may be that
intensity differences existed across the emotional categories and this
might account for the discrepancy in findings. However, this is not
likely the case as intensity ratings across the different emotional
categories were similar. No plausible explanation is available for

107
the lack of emotion specific effects in HR responses together with the
possible emotion specific effects in SCRs. The other consideration is
that the SCR findings represent only a trend and may not truly effect
an emotion specific hemispheric effect.
The findings of female Ss are not consistent with any hemispheric
emotionality models. In contrast to laterality effects observed for
male Ss, female Ss appeared to show a qualitatively different pattern
of HR responding which differentiated their performance from male
Ss. This was revealed in a significant Sex x Stimulus Type
interaction for analysis of HR deceleratory responses and HR
acceleratory responses. This pattern of findings demonstrated that
female but not male Ss snowed differential deceleratory as well as
acceleratory HR responses across the different conditions of stimulus
type with significantly greater acceleratory responses to angry and
fearful slides compared iith neutral slides and significantly greater
deceleratory responses to happy, disgusting, and neutral slides
relative to fearful slides.
These findings also provide further support for the view that
different patterns of autonomic arousal may be associated with
different types of emotional states. In addition, they also suggest
that possible differences in imagery ability across male and female Ss
may have, in part, mediated this differential pattern of responding.
These findings point to the relative importance of considering
both sex and imagery ability of Ss in further investigations of
emotional processing and autonomic responding. Sex differences in
degree of lateral asymmetry of arousal responses suggest that the

108
greater role of the right hemisphere in production of arousal
responses exists to a significantly greater extent in male Ss versus
female Ss; findings which are congruent with findings from
neuropsychological investigations which report a greater degree of
lateral asymmetries for male Ss on a variety of tasks. Similarly, sex
differences in patterning of HR responsivity across different
emotional categories suggests that female but not male Ss reveal
differences in magnitude of HR acceleratory and HR deceleratory
responses across various emotional categories. The extent to which
this differential autonomic responding in male versus female Ss is due
to differences in imagery ability across these groups is an area of
future investigation.

REFERENCES
Ahern, G. L., & Schwartz, G. E. (1979)- Differential lateralization
for positive versus negative emotions. Neuropsychologia, 17,
693-697.
Ainsworth, M. D. S., Blehar, M., Waters, E., & Wall, E. (1978).
Patterns of attachment. Hillsdale, NJ: Erlbaum.
Alema, G., Rosadini, G., & Rossi, G. F. (1961). Psychic reactions
associated with intracarotid amytal injections and relation to
brain injury. Excerpta Medica, 37, 154-155.
Arabian, J. M., & Furedy, J. J. (1983). Individual differences in
imagery ability and pavlovian heart rate decelerative
conditioning. Psychophysiology 20, 325-331.
Ax, A. F. (1953). The physiological differentiation between fear and
anger in humans. Psychosomatic Medicine, 15, 433-442.
Babinski, J. (1914). Contribution a 1'tude des troubles mentaux dans
1'hemisplegie organique crbrale (anosognosie). Revue
Neurologique, 27, 845-848.
Bauer, R. M., *%. Craighead, W. E. ( 1979). Psychophysiological
responses to the imagination of fearful and neutral situations:
The effects of imagery instruction. Behavior Therapy, 10,
389-403.
Bayley, N. (1968). Behavioral correlates of mental growth: Birth to
36 years. American Psychologist, J_, 1-17.
Bayley, N. (1969). Bayley Scales of infant development^ Birtji to two
years. New York: Psychological Corporation.
Bear, D. M., & Fedio, P. (1977). Quantitative analysis of interictal
behavior in temporal lobe epilepsy. Archives of Neurology, 34,
454-467.
Benson, D. F., & Geschwind, N. (1971). Aphasia and related cortical
disturbances. In: A. B. Baker & L. H. Baker (Eds.), Clinical
Neurology. New York: Harper & Row.
Benton, A. (1968). Right-left discrimination. Pediatric Clinics of
North America, 15, 747~758.
109

110
Berg, M. R., & Harris, L. J. (1980). The effect of experimenter
location and subject anxiety on cerebral activation measured by
lateral eye movements. Neuropsychologia, 18, 89-9^.
Berger, H. (1933)- Uber das electroenkephalogram des menschen.
Archiv fur Psychiatrie und Nervenkrankheiten, 99, 555-574.
Bihrle, A. M., Brownell, H. H., & Powelson, J. A. (1986).
Comprehension of humorous and nonhumorous materials by left and
right brain damaged patients. Brain and Cognition, 5, 399-411.
Borod, J. C., 4 Caron, H. S. (1980). Facedness and emotion related to
lateral dominance, sex and expression type. Neuropsychologia,
J_8, 237-242.
Borod, J. C., Caron, H. S., 4 Koff, J. (1981). Asymmetry in positive
and negative facial expressions: Sex differences.
Neuropsychologia, 19, 819-824.
Borod, J. C., Kent, J., Koff, E., Martin, C., & Alpert, M. (1988a).
Facial asymmetry while posing positive and negative emotions:
Support for the right hemisphere hypothesis. Neuropsychologia,
26, 759-764.
Borod, J. C., Koff, E., Perlman, E., Lorch, M., & Nicholas, M.
(1986). The expression and perception of facial emotion in brain
iamaged patients. Neuropsychologia, 24, 169-180.
Borod, J. C., Koff, E., & White, B. (1983). Facial asymmetry in posed
and spontaneous expressions of emotion. Brain and Cognition, 2,
165-175.
Borod, J. C., Vingiano, W., & Cytryn, F. (1988b). The effects of
emotion and ocular dominance on lateral eye movement.
Neuropsychologia, 26, 213-220.
Bower, G. H. (1981). Mood and memory. American Psychologist, 38,
129-148.
Bowers, D., Bauer, R., Coslett, B., & Heilman, K. (1985). Processing
of faces by patients with unilateral hemispheric lesions. I.
Dissociation between judgements of facial affect and facial
identity. Brain and Cognition, _4, 258-272.
Bowers, D., & Heilman, K. (1984). Dissociation between the processing
of affective and non-affective faces: A case study. Journal of
Clinical Neuropsychology, _6, 367-374.
Bowlby, J. (1972). Separation: Volume II. Attachment and loss.
New York: Basic Books.

Briggs, G. G., & Nebes, R. D. (1975). Patterns of hand preferences in
a student population. Cortex, 11, 230-238.
Brody, N., Goodman, S. E., Halm, E., Krinzman, S., 4 Sebrechts, M. M.
(1987). Lateralized affective priming of lateralized affectively
valued target words. Neuropsychologia, 25, 935-946.
Brownell, H. H., Potter, H. H., Michelow, D., 4 Gardner, H. (1984).
Sensitivity to lexical denotation and connotation in brain
damaged patients: A double dissociation? Brain and Language,
22, 253-265.
Bruder, G. E. (1983). Cerebral laterality and psychopathology: A
review of dichotic listening studies. Schizophrenia Bulletin, _9,
143-151.
Brumback, R. A., 4 Staton, R. D. (1981). Depression-induced
neurological dysfunction. New England Journal of Medicine, 304,
642.
Bryden, P. (1978). Strategy effects in hemispheric asymmetry. In:
G. Underwood (Ed.), Strategies of information processing.
New York: Academic Press.
Bryden, P., 4 AlLard, F. (1976). Visual hemifield differences depend
on typeface. Brain and Language, _3, 191-200.
Bryden, P., 4 Ley, R. G. (1983). Right hemisphere involvement in the
perception and expression of emotion in normal humans. In:
K. M. HeiLman (Ed.), Neuropsychology of human emotion.
New York: Guillford Press.
Bryden, M. P., Ley, R. G., 4 Sugarman, J. H. (1982). A left ear
advantage for identifying the emotional quality of tonal
sequences. Neuropsychologia, 20, 8387.
Buchtei, H. A., Campari, F., DeRisio, C., 4 Rota, R. (1978).
Hemispheric differences in discriminative reaction time to facial
expressions. Italian Journal of Psychology, jj, 159-169.
Buck, R., 4 Duffy, R. J. (1980). Nonverbal communication of affect in
brain damaged patients. Cortex, 16, 351-362.
Cannon, W. B. (1927). The James-Lange theory of emotions: A critical
examination and an alternative theory. American Journal of
Psychology, 34, 106-124.
Carmon, A., 4 Nachson, I. (1973). Ear asymmetry in perception of
emotional nonverbal stimuli. Acta Psychologica, 37, 351-357.

112
Carroll, D., Baker, J., & Preston, M. (1979). Individual differences
in visual imaging and the voluntary control of heart rate.
British Journal of Psychology, 70, 39-49.
Charlesworth, W. 3. (1964). Instigation and maintainence of curiosity
behavior as a function of surprise versus novel and familiar
stimuli. Child Development, 35, 1169-1186.
Cicone, M., Wapner, W., & Gardner, H. (1980). Sensitivity to
emotional expression and situations in organic patients. Cortex,
J_6, 1 45-158.
Cimino, C., Verfaellie, M., Bowers, D., & Heilman, K. (1988).
Emotional autobiographical memory in right hemisphere damaged
patients. Unpublished manuscript.
Coulbourn, C. J., & Lishman, W. A. (1974;. Lateralization of function
and psychiatric illness. A left hemisphere deficit? In: J.
Gruzelier & P. Flor-Henry (Eds.), Hemispheric asymmetries of
function and psychopathology. Amsterdam: Elseview.
Cummings, J. L., & Mendez, M. P. (1984). Secondary mania with focal
cerebrovascular lesions. American Journal of Psychiatry, 141,
1084-1087.
Davidson, R. J., Horowitz, M. E., Schwartz, G. E., & Goodman, D. M.
(1981). Lateral differences in the latency between finger
tapping and the heart beat. Psychophysiology, 18, 36-41.
Davidson, R. J., & Schwartz, G. E. (1976). Patterns of cerebral
lateralization during cardiac biofeedback versus the self
regulation of emotion: Sex differences. Psychophysiology, 13.
62-68.
Davidson, R. J., Schwartz, G. E., Saron, C., Bennett, J., & Golemena,
D. J. (1979). Frontal vs. parietal EEG asymmetry during positive
and negative affect. Psychophysiology, 16, 202-203.
Davies, J. B. (1978). The psychology of music. London: Hutchinson.
DeKosky, S., Heilman, K. M., Bowers, D., & Valenstein, E. (1980).
Recognition and discrimination of emotional faces and pictures.
Brain and Language, 206-214.
d'Elia, G., & Perris, C. (1973). Cerebral functional dominance and
depression: An analysis of EEG amplitude in depressed
patients. Acta Psychiatrica Scandinavica, 49, 191-197.

113
d'Elia, G., & Perris, C. (1974). Cerebral functional dominance and
memory functions: An analysis of EEG integrated amplitude in
depressive psychotics. Acta Psychiatrica Scandinavica
(Supplement), 255, 143-157.
Delp, M. J., & Sackeim, H. A. (1987). Effects of mood on lacrimal
flow: Sex differences and asymmetry. Psychophysiology, 24,
550-556.
Denny-Brown, D., Meyer, J. S., & Horenstein, S. (1952). The
significance of perceptual rivalry resulting from parietal
Lesions. Brain, 75, 434-471 .
DeWitt, G. W. (1978). Laterality, personality and the perception of
emotional stimuli. Dissertation Abstracts International,
38(8-B), 3850.
Dimond, S. J., Farrington, L., 4 Johnson, P. (1976). Differing
emotional response from right and left hemispheres. Nature, 261,
690-692.
Dopson, W. G., Beckwith, B. F., Tucker, D. M., & Bullard-Bates, P. C.
(1984). Asymmetry of facial expression in spontaneous emotion.
Cortex, 20, 243-252.
Ehrlichman, H., 4 Weinberger, A. (1978). Lateral eye movements and
hemispheric asymmetry: A critical review. Psychological
Bulletin, 85, 1080-1101.
Ekman, P. (1972). Universal and cultural differences in facial
expressions of emotion. In J. K. Cole (Ed.), Nebraska symposium
on motivation. Lincoln: University of Nebraska Press.
Ekman, P., Hager, J. C., & Friesen, W. V. (1981). The asymmetry of
emotional and deliberate facial actions. Psychophysiology, 18,
101-106.
Ekman, P., Levenson, R. W., & Friesen, W. V. (1983). Autonomic
nervous system activity distinguishes among emotions. Science,
221, 1208-1210.
Ellinwood, E. H. (1967). Amphetamine psychosis: 1. Description of
the individuals and process. Journal of Nervous and Mental
Disease, 144, 273-283.
Fehr, F. S., & Stern, J. A. (1970). Peripheral psychological
variables and emotion: The James-Lange theory revisited.
Psychological Bulletin, 74, 411-424.
Flor-Henry, P. (1969). Psychosis and temporal lobe epilepsy: A
controlled investigation. Epilepsia, 10, 363-395.

114
Flor-Henry, P. (1976). Lateralized temporal limbic dysfunction and
psychopathology. Annals of the New York Academy of Sciences,
280, 777-795.
Flor-Henry, P. (1979). On certain aspects of the localization of the
cerebral systems regulating and determining emotion. Biological
Psychiatry, 14, 677-698.
Flor-Henry, P. (1983). Functional hemispheric asymmetry and
psychopathology. Integrative Psychiatry, J_, 46-52.
Flor-Henry, P., & Koles, Z. V. (1980). EEC studies in depression,
mania and normals: Evidence for partial shifts of laterality in
the affective psychoses. Advances in Biological Psychiatry, 4,
21-43.
Fox, N. A., & Davidson, R. J. (1984). Hemispheric substrates of
affect: A developmental model. In: N. J. Fox & R. J. Davidson
(Eds.), The psychobiology of affective development. Hillsdale,
NJ: Erlbaum.
Freeman, R. L., Galaburda, A. M., Cabal, R. D., & Geschwind, N.
(1935). The neurology of depression. Archives of Neurology, 42,
289-291.
Freid, I., Mateu, C., Ojemann, G., Wohns, J., b Fedio, P. (1982).
Organization of visuospatial function in human cortex. Evidence
from electrical stimulation. Brain, 105, 349-371.
French, J. E., Hernandez-Peon, R., & Livingston, R. (1955).
Projections from the cortex to cephalic brainstem (reticular
formation) in monkey. Brain 1 8, 74-95.
Freyschuss, V. (1970). Cardiovascular adjustments to somato-motor
activation: The elicitation of increments in heart rate, aortic
pressure and venomotor tone with the initiation of muscle
contraction. Acta Physiologica Scandinavica, 342, 1-63.
Gage, D. F., & Safer, M. A. (1979). Hemispheric differences in the
mood state-dependent effect for recognition of emotional faces.
Journal of Experimental Psychology: Learning, Memory and
Cognition, 11, 752-763.
Gainotti G. (1972). Emotional behavior and hemispheric side of the
lesion. Cortex, _8, 41-55.
Galin, D., Diamond, R., b Braff, D. (1977). Lateralization of
conversion symptoms: More frequent on the left. American
Journal of Psychiatry, 134, 578-580.

115
Gasparrini, W. G., Satz, P., Heilman, K. M., & Coolidge, F. L.
(1978). Hemispheric asymmetries of affective processing as
determined by the Minnesota Multiphasic Personality Inventory.
Journal of Neurology, Neurosurgery and Psychiatry, 41, 470-473.
Gertsmann, J. (1940). Syndrome of finger agnosia, disorientation for
right and left, agraphia and acalculia. Archives of Neurology
and Psychiatry, 44, 398-408.
Geschwind, N. (1969). Problems in the anatomical understanding of the
aphasia. In A. Benton (Ed.), Contributions to clinical
neuropsychology. Chicago: Aldine.
Geschwind, N. (1975). The apraxias: Neural mechanisms of disorders
of learned movement. American Scientist, 63, 188-195.
Glick, S. D., Jerussi, T. P., & Zimmerberg, B. (1977). Behavioral and
neuropharmacological correlates of nigrostriatal asymmetry in
rat. In: S. Harnad, R. Doty, L. Goldstein, J. Jaynes, & G.
Krauthamer (Eds.), Lateralization in the nervous system.
New York: Academic.
Glick, S. D., Meibnach, R. C., Cox, R. D., & Maayani, S. (1979).
Multiple and interrelated functional asymmetries in rat brain.
Life Sciences, 25, 395-400.
Goldstein, K. (1948). Language and language disturbances.
New York: Grue 8c Stratton.
Goldstein, K. (1952). The effect of brain damage on the
personality. Psychiatry, 1 5, 245-260.
Goldstein, S. G., Filskov, S., Weaver, L. A., 4 Ives, J. 0. (1977).
Neuropsychological effects of electroconvulsive therapy. Journal
of Clinical Psychology, 33, 798-806.
Graham, F. K. (1980). Representing cardiac activity in relation to
time. In: I. Martin and P. Venables (Eds.), Techniques in
psychophysiology. New York: Wiley.
Graves, R., Landis, T., & Goodglass, H. (1980). Laterality and sex
differences for visual recognition of emotional and non-emotional
words. Paper presented at the meeting of the Academcy of
Aphasia, Cape Cod, MA.
Graves, R., Landis, T., 8c Goodglass, H. (1981). Laterality and sex
differences for visual recognition of emotional and non-emotional
words. Neuropsychologia, 19, 95-102.

116
Greenwald, M., Lang, P., & Cooke, E. (1988). Response patterning
during affective visual stimulation: Evidence for a dimensional
model of emotion. Unpublished manuscript.
Gur, R. E., & Gur, R. C. (1975). Defense mechanisms, psychosomatic
symptomatology and conjugate lateral eye movements. Journal of
Consulting and Clinical Psychology, _4_3 416-420.
Haggard, M. P., & Parkinson, A. M. (1971). Stimulus and task factors
as determinants of ear advantages. Quarterly Journal of
Experimental Psychology, 23. 168-177.
Hansch, E. C., & Pirozzolo, F. J. (1980). Task relevant effects on
the assessment of cerebral specialization for facial emotion.
Brain and Language, 10, 51-59.
Hantas, M. N., Katkin, E. S., & Reed, S. D. (1984). Cerebral
lateralization and heartbeat discrimination. Psychophysiology,
_21_, 241-278.
Hare, R. D., & Blevings, G. (1975). Defensive responses to phobic
stimuli. Biological Psychology, _3, 1-13*
Hecaen, H. (1962). Clinical symptomatology in right and left
hemispheric lesions. In: V. B. Mountcastle (Ed.),
Interhemispheric relations and cerebral aominance. Baltimore:
Johns Hopkins University Press.
Heilman, K. M., Bowers, D., Speedie, L., & Coslett, B. (1984).
Comprehension of affective and nonaffective prosody. Neurology,
34, 917-921.
Heilman, K. M., Scholes, R., & Watson, R. T. (1975). Auditory
affective agnosia: Disturbed comprehension of affective
speech. Journal of Neurology, Neurosurgery and Psychiatry, 38,
69-72.
Heilman, K. M., Schwartz, H. D., & Watson, R. T. (1978). Hypoarousal
in patients with the neglect syndrome and emotional
indifference. Neurology, 28, 229-232.
Heilman, K. M., 4 Valenstein, E. (1972). Frontal lobe neglect.
Neurology, 22, 660-664.
Heilman, K. M., 4 Van Den Abel, T. (1979). Right hemispheric
dominance for mediating cerebral activation. Neuropsychologia,
17, 315-321.

117
Heilman, K. M., Watson, R. T., 4 3owers, D. (1983). Affective
disorders associated with hemispheric disease. In: K. M.
Heilman (Ed.), Neuropsychology of human emotion. New York:
Guillford Press.
Heller, W., 4 Levy, J. (1981). Perception and expression of emotion
in right handers and left handers. Neuropsychologia, 19,
363-372.
Holloway, F. A., & Parsons, 0. A. (1969). Unilateral brain damage and
bilateral 3kin conductance levels in humans. Psychophysiology,
6, 138-148.
Hommes, 0. R., & Panhuysen, L. H. H. M. (1971). Depression and
cerebral dominance: A study of bilateral intracarotid amytal in
eleven depressed patients. Psychiatria, Neurologa,
Neurochirurgia, 74, 259-270.
Hugdahl, K., Franzon, M., Anderson, B., 4 Walldebo, G. (1983). Heart
rate responses to lateralized visual stimuli. Pavlovian Journal
of Biological Sciences, 18, 186-198.
Hugdahl, K., Wahlgren, C., 4 Wass, T. (1982). Habituation of the
electrodermal orienting reaction is dependent on the cerebral
hemisphere initially stimulated. Biological Psychology, 15,
49-62.
Ikeda, Y., 4 Hirai, H. (1976). Voluntary control of electrodermal
activity in relation to imagery ana internal perception scores.
Psychophysiology, 13, 330-333.
Iversen, S. D. (1977). Brain dopamine systems and behavior. In:
L. L. Iversen, S. D. Iversen, 4 S. H. Snyder (Eds.), Handbook of
psychopharmacology, Volume 8. Drugs, neurotransmitters and
behavior. New York: Plenum Press.
Izard, C. E. (1977). Visual attention in infancy: Processes, methods
and clinical applications. In: I. C. Uzgiris 4 T. Weizmann
(Eds.), The structuring of experience. New York: Plenum Press.
Izard, C. E., Hubner, R. R., Risser, D., McGuiness, G. C., 4
Dougherty, L. M. (1980). The young infant's ability to produce
discrete emotional expression. Developmental Psychology, 16,
132-140.
Johnson, 0., 4 Crockett, D. (1982). Changes in perceptual asymmetries
with clinical improvement of depression and schizophrenia.
Journal of Abnormal Psychology, 91, 45-54.

118
Jones, G. E., & Johnson, H. J. (1978). Physiological responding
during self-generated imagery of contextually complete stimuli.
Psychophysiology, 15, 439-446.
Jones, G. E., & Johnson, H. J. (1980). Heart rate and somatic
concomitants of mental imagery. Psychophysiology, 17, 339~347.
Karlin, R., Weinapple, M., Rochford, J., & Goldstein, L. (1979).
Quantitated EEG features of negative affective states: Report of
some hypnotic studies. Research Communications in Psychology,
Psychiatry and Behavior, _4, 397-413.
King, F. L., & Kimura, D. (1972). Left ear superiority in dichotic
perception of vocal nonverbal sounds. Canadian Journal of
Psychology, 26, 111-116.
Kinsbourne, M. (1970). The cerebral basis of lateral asymmetries in
attention. Acta Psychologia, 33. 163-201.
Kinsbourne, M. (1972). Eye and head-turning indicates cerebral
lateralization. Science, 176, 539-541.
Kinsbourne, M. (1973). The control of attention by interaction
between the cerebral hemispheres. In: S. Kornblum (Ed.),
Attention and performance, IV. New York: Academic.
Kinsbourne, M., & Bemporad, B. (1984). Lateralization of emotion: A
model and the evidence. In: N. A. Fox & R. J. Davidson (Eds.),
The psychobiology of affective development. Hillsdale, NJ:
Erlbaum.
Kirk, R. E. (1968). Experimental design: Procedures for the
behavioral sciences. Belmont, CA: Wadsworth Publishing.
Klorman, R., & Ryan, R. (1980). Heart rate, contingent negative
variation and evoked potentials during anticipation of affective
stimulation. Psychophysiology, 17, 513-523.
Klorman, R., Weissberg, R., & Weisenfeld, A. (1977). Individual
differences in fear and autonomic reactions to affective
stimulation. Psychophysiology, 4, 45-51.
Kolb, B., & Milner, B. (1981). Observations on spontaneous facial
expression after focal cerebral excisions and after intracarotid
injection of sodium amytal. Neuropsychologia, 19, 505-514.
Kolb, B., & Taylor, L. (1981). Affective behavior in patients with
localized cortical excisions: Role of lesion site and size.
Science, 214, 89-91.

119
Kronfol, Z., Hamsher, K. De S., Digre, K., & Wazir, R. (1978).
Depression and hemisphere functions: Changes associated with
unilateral ECT. British Journal of Psychiatry, 132, 560-567.
Kuzendorf, R. G. (1982). Individual differences in imagery and
autonomic control. Journal of Mental Imagery, _6, 47-60.
Lacey, J. I. (1967). Somatic response patterning and stress: Some
revisions of activation theory. In: M. H. Appley & R. Trumball
(Eds.), Psychological stress: Issues in research. New York:
Appleton-Century-Crofts.
Lacey, J. I. (1972). Some cardiovascular correlates of sensorimotor
behavior: Example of visceral afferent feedback? In: C. H.
Hockman (Ed.), Limbic system mechanisms and autonomic function.
Springfield, IL: Charles C. Thomas.
Lacroix, D. M., & Comper, P. (1979). Lateralization in the
electrodermal system as a function of cognitive/hemispheric
manipulations. Psychophysiology, 16, 116-124.
Lang, P. J. (1977). Imagery in therapy: An information processing
analysis of fear. Behavior Therapy, j), 862-886.
Lang, P. J. (1979). A bio-informational theory of emotional
imagery. Psychophysiology, 16, 495-512.
Lansing, R. W., Schwartz, E., & Lindsey, D. B. (1959). Reaction time
and EEG activation under alerted and nonalerted conditions.
Journal of Experiment Psychology, 58, 1-7.
Levy, J. (1969). Possible basis for the evolution of lateral
specialization of the human brain. Nature, 224, 614-615.
Ley, R. G., & Bryden, M. P. (1979). Hemispheric differences in
processing emotions and faces. Brain and Language, 7, 127-138.
Ley, R. G., & Bryden, M. P. (1982). Dissociation of right and left
hemispheric effects for recognizing emotional tone and verbal
content. Brain and Cognition, 1, 3-9.
Lindsley, D. (1970). The role of nonspecific reticulo-thalamo-
cortical systems in emotion. In: P. Black (Ed.), Physiological
correlates of emotion. New York: Academic Press.
Lishman, W. A. (1968). Brain damage in relation to psychiatric
disability after head injury. British Journal of Psychiatry,
114, 373-410.

120
Mach, E. (1959). The analysis of sensations, and the relation of the
physical to the psychological (S. Waterlow, Ed., and C. M.
Williams, Trans.). New York: Dover. (Original work published
in 1914)
Mahoney, A. M., & Sainsbury, R. S. (1987). Hemispheric asymmetry in
the perception of emotional sounds. Brain and Cognition, _6,
216-233.
McGlone, J. (1980). Sex differences in human brain asymmetry: A
critical survey. Behavioral and Brain Sciences, _3, 215-263.
McKeever, W. F., & Dixon, M. S. (1981). Right hemisphere superiority
of discriminating memorized from non-memorized faces: Affective
imagery, sex and perceived emotional effects. Brain and
Language, 12, 246-260.
Melamed, B. G. (1969). The habituation and psychophysiological
responses to tone, and to filmed fear stimuli under varying
conditions of instructional set. Unpublished doctoral
dissertation, University of Wisconsin.
Montgomery, W. A., & Jones, G. E. (1984). Laterality, emotionality
and heartbeat perception. Psychophysiology, 21, 459-465.
Morrow, L., Urtunski, P. B., Kim, Y., & Boiler, F. (1981). Arousal
responses to emotional stimuli and laterality of lesion.
Neuropsychologia, 19, 65-71.
Moruzzi, G., & Magoun, H. W. (1949). Brainstem reticular formation
and activation of the EEC. EEG and Clinical Neurophysiology, J_,
455-475.
Moskovitch, M. (1983). Local versus global solutions to problems of
hemispheric specialization. Behavioral and Brain Sciences, 6,
520-521.
Moskovitch, M. (1986). Afferent and efferent models of visual
perceptual asymmetries: Theoretical and empirical
implications. Neuropsychologia, 24, 91-114.
Moskovitch, M., & Olds, J. (1982). Asymmetries in spontaneous facial
expression and their possible relation to hemispheric
specialization. Neuropsychologia, 20, 71-81.
Myslobodsky, M. S., & Horesch, N. (1978). Bilateral electrodermal
activity in depressive patients. Biological Psychology, 6,
111-120.
Myslobodsky, M. S., & Rattok, J. (1977). Bilateral electrodermal
activity in waking man. Acta Psychologia, 41, 273-282.

121
Nebes, R. D. (1974). Hemispheric specialization in commissurotomized
man. Psychological Bulletin, 81, 1-14.
Nottebohm, F. (1977). Asymmetries in neural control of vocalization
in the canary. In: S. Harnad, R. Doty, L. Goldstein, J. Jaynes,
& G. Krauthamer (Eds.), Lateralization in the nervous system.
New York: Academic.
Obrist, P. A., Howard, J. L., Lawler, J. E., Galosy, R. A., Meyers, K.
A., & Gaebelein, C. J. (1974). The cardiac-somatic
interaction. In: P. A. Obrist, A. H. Black, J. Brener, & L. V.
DiCara (Eds.), Cardiovascular psychophysiology: Current issues
in response mechanisms, biofeedback and methodology. Chicago:
Aldine-Atherton.
Oke, A., Keller, R., Mefford, I., & Adams, R. (1978). Lateralization
of norephinephrine in human thalamus. Science, 200, 1411-1413.
Oke, A., Lewis, R., & Adams, R. N. (1980). Hemispheric asymmetry of
norepinephrine distribution in rat thalamus. Brain Research,
188, 269-272.
Papanicolaou, A. C., Moore, B., Deutsch, G., Levin, H., & Eisenberg,
H. (1988). Evidence for right hemisphere involvement in recovery
from aphasia. Archives of Neurology, 45, 1025-1029.
Patterson, K., & Bradshaw, J. L. (1975). Differential hemispheric
mediation of nonverbal visual stimuli. Journal of Experimental
Psychology: Human Perception and Performance, _1_, 246-252.
Paul, H. (1909). Principien der sprachgeschichte (4th ed.). Germany:
Niemeyer.
Pennebaker, J. W., & Hoover, C. W. (1984). Visceral perception versus
visceral detection: Disentangling methods and assumptions.
Biofeedback and Self-Regulation, .9, 339-352.
Perria, P., Rosadini, G., & Rossi, G. F. (1961). Determination of
side of cerebral dominance with amobarbital. Archives of
Neurology, _4, 175-181.
Perris, C. (1974). Average evoked responses (AER) in patients with
affective disorders. Acta Psychiatrica Scandinavica
(Supplement), 225, 89_98.
Perris, C. (1975). EEG techniques in the measurement of the severity
of depressive syndromes. Neuropsychobiology, 1, 16-25.

122
Rader, N., Bausano, M.f & Richards, J. E. (1980). On the nature of
visual-cliff avoidance response in human infants. Child
Development, 51, 61-68.
Reuter-Lorenz, P. A., & Davidson, R. (1981). Differential
contributions of the two cerebral hemispheres for perception of
happy and sad faces. Neuropsychologia, 19, 609-614.
Reuter-Lorenz, P. A., Givis, R. P., & Moskovitch, M. (1983).
Hemispheric specialization and the perception of emotion:
Evidence from right-handers and from inverted and non-inverted
left handers. Neuropsychologia, 21, 687-692.
Robinson, R. G. (1983). In M. Reivich (Ed.), Cerebrovascular disease,
13th Princeton Conference. New York: Raven Press.
Robinson, R. G., & Benson, D. F. (1981). Depression in aphasic
patients: Frequency, severity and clinical-pathological
correlations. Brain and Language, 14, 282-291.
Robinson, R. G., & Price, T. R. (1982). Post-stroke depressive
disorders: A foilow-up study of 103 patients. Stroke, 13.
63-641.
Rockford, J., Swartzburg, M., Chaudberg, S. M., & Goldstein, L.
(1976). Some quantitative EEG correlates of psychopathology.
Research Communications in Psychology, Psychiatry and Behavior,
T, 211-226.
Ross, E. D. (1981). The aprosodias: Functional-anatomic organization
of the affective components of language in the right
hemisphere. Annals of Neurology, 38, 561-589.
Ross, E. D., & Mesulam, M-M. (1979). Dominant language functions of
the right hemisphere? Prosody and emotional gesturing. Archives
of Neurology, 36, 144-148.
Rossi, G. F., & Rosadini, G. R. (1967). Experimental analyses of
cerebral dominance in man. In: D. H. Millikan & F. L. Darley
(Eds.), Brain mechanism underlying speech and language.
New York: Grue & Stratton.
Rouse, C. H., Jones, G. E., & Jones, K. R. (1988). The effect of body
composition and gender on cardiac awareness. Psychophysiology,
25, 400-407.
Rubin, D. A., & Rubin, R. T. (1980). Differences in asymmetry of
facial expressions between right- and left-handed children.
Neuropsychologia, 18, 373-377.

123
Sackeim, H. A., Greenberg, M. S., Weiman, A. L., Gur, R. C.,
Hungerbuhler, J. P., & Geschwind, N. (1982). Hemispheric
asymmetry in the expression of positive and negative emotions.
Archives of Neurology, 39, 210-218.
Sackeim, H. A., Gur, R. C., & Saucy, M. C. (1978). Emotions are
expressed more intensely on the left side of the face. Science,
202, 434-436.
Safer, M. A. (1981). Sex and hemisphere differences in access to
codes for processing emotional expressions and faces. Journal of
Experimental Psychology: General, 110, 86-100.
Safer, M. A., & Leventhal, M. (1977). Ear differences in evaluating
tones of voices and verbal content. Journal of Experimental
Psychology: Human Perception and Performance, _3, 75-82.
Sauguet, J., Benton, A., & Hecaen, H. (1971). Disturbances of the
body schema in relation to language impairment and hemispheric
locus of lesion. Journal of Neurology, Neurosurgery and
Psychiatry, 34, 496-501.
Schacter, S. (1957). Pain, fear and anger in hypertensive and
normotensives. Psychosomatic Medicine, 19, 17-29.
Schacter, S. (1964). The interaction of cognitive and physiological
determinants of emotional state. Advances in Experimental Social
Psychology, J_, 49-80.
Schacter, S. (1970). The interaction of cognition and physiological
determinants of emotional state. In: J. Berkowitz (Ed.),
Advances in experimental social psychology (Vol. 1). New York:
Academic Press.
Schacter, S., & Singer, J. (1962). Cognitive, social and
physiological determinants of emotional state. Psychological
Review, 69, 379-399.
Schneider, S. J. (1983). Multiple measures of hemispheric dysfunction
in schizophrenia and depression. Psychological Medicine, 13,
287-297.
Schwartz, G. E., Ahern, G. L., & Brown, S. L. (1979). Lateralized
facial muscle response to positive and negative emotional
stimuli. Psychophysiology 16, 561 -57 3 -
Schwartz, G. E., Davidson, R. J., & Maier, F. (1975). Right
hemisphere lateralization for emotion in human brain:
Interactions with cognition. Science, 190, 286-288.

124
Schwartz, G. E., Weinberger, D. A., & Singer, J. A. (1981).
Cardiovascular differentiation of happiness, sadness, anger and
fear following imagery and exercise. Psychosomatic Medicine, 43,
343-364.
Sequndo, J. P., Naguet, R., & Buser, P. ( 1955). Effects of cortical
stimulation on electrocortical activity in monkeys. Journal of
Neurophysiology, 18, 236-245.
Sergent, J. (1983). Role of the input in visual hemispheric
asymmetries. Psychological Bulletin, 93 481-512.
Sergent, J., & 3indra, D. (1981). Differential hemispheric processing
of faces: Methodological considerations and reinterpretation.
Psychological Bulletin, 89, 541-554.
Silberman, E. K., & Weingartner, H. (1986). Hemispheric
lateralization of functions related to emotion. Brain and
Cognition, 5, 322-353.
Silberman, E. K., Weingartner, H., & Post, R. M. (1983a). Thinking
disorder in depression: Logic and strategy in an abstract
reasoning task. Archives of General Psychiatry, 40, 775-780.
Silberman, E. K., Weingartner, H., Stillman, R., Chen, H. J., & Post,
R. M. (1983b). Altered lateralization of cognitive processes in
depressed women. American Journal of Psychiatry, 140, 1340-1344.
Smokier, I. A., & Shevrin, I. (1979). Cerebral lateralization and
personality style. Archives of General Psychiatry, 36, 949954.
Sokolov, E. N. (1963). Perception and the conditioned reflex.
Oxford: Pergamon Press.
Sroufe, L. A., & Wunsch, J. P. (1972). The development of laughter in
the first year of life. Child Development, 43, 1325-1344.
Steinberg, C. R., & Campos, J. J. (1983). The facial expression of
anger in seven-month-old infants. Child Development, 54,
178-184.
Strauss, E. (1983). Perception of emotional words. Neuropsychologia,
21_, 99-103.
Strauss, E., & Moscovitch, M. (1981). Perception of facial
expressions. Brain and Language, 13, 308-332.
Suberi M., & McKeever, W. F. (1977). Differential right hemisphere
memory storage of emotional and non-emotional faces.
Neuropsychologia, 15, 757-765.

125
Taylor, D. C. (1972). Mental status and temporal lobe epilepsy: A
correlative account of 100 patients treated surgically.
Epilepsia, 13, 727-765.
Terzian, H. (1964). Behavioral and EEC effects of intracarotid sodium
amytal injections. Acta Neurochirurgic, 12, 230-240.
Toone, B. K., Cooke, E., & Lader, M. H. (1981). Electrodermal
activity in the affective disorders and schizophrenia.
Psychological Medicine, 11, 497-508.
Tranel, D., Damasio, A., & Damasio, H. (1988). Intact recognition of
facial expression, gender and age in patients with impaired
recognition of face identity. Neurology, 38, 690-696.
Tucker, D. M. (1981). Lateral brain function, emotion and
conceptualization. Psychological Bulletin, 89, 19-46.
Tucker, D. M., Roth, R. S., Arneson, B. A., & Buckingham, V.
(1977a). Right hemisphere activation during stress.
Neuropsychologia, 15, 697-700.
Tucker, D. M., Stenslie, C. E., Roth, R. S., & Shearer, S. L.
(1981). Right frontal lobe activation and right hemisphere
performance decrement during a depressed mood. Archives of
General Psychiatry, 38, 169-174.
Tucker, D. M., Watson, R. T., & Heilman, K. H. (1977b). Affective
discrimination and evocation in patients with right parietal
disease. Neurology, 17, 947-950.
Verfaellie, M., Bowers, D., Heilman, K. M. (1987). Hemispheric
asymmetries in mediating intention, but not selective
attention. Neuropsychologia, 26, 521-531.
Verfaellie, M., & Heilman, K. M. (1987). Response preparation and
response inhibition after lesions of the medial frontal lobe.
Archives of Neurology, 44, 1265-1271.
Vrana, S. R., Cuthbert, B., & Lang, P. (1986). Fear imagery and text
processing. Psychophysiology, 23 247-253.
Wagner, H. N., Burns, D. H., Dannal3, R. F., Wondg, D. F., Langstrom,
B., Duelfer, T., Frost, J. J., Ravert, H. T., Links, J. M.,
Rosenbloom, S. B., Lucas, S. E., Kramer, A. V., & Kuhlar, M. J.
(1983). Imaging dopamine receptors in the human brain by
positron tomography. Science, 221, 1264-1266.
Walker, B., & Sandman, C. (1982). Visual evoked potentials change as
heart rate and carotid pressure change. Psychophysiology, 19,
520-527.

126
Watson, R. T.f Heilman, K. M., Cauthen, J. C., & King, F. A. (1973).
Neglect after cingulectomy. Neurology, 23, 10031007.
Webster, W. G. (1977). Territoriality and the evolution of brain
asymmetry. Annals of the New York Academy of Science, 299,
213-221 .
Weerts, T. C., & Roberts, R. (1976). The physiological effects of
imagining anger-provoking and fear-provoking scenes.
Psychophysiology, 13, 174.
Weingartner, H., & Silberman, E. K. (1982). Models of cognitive
impairment: Cognitive changes in depression. Psychopharmacology
Bulletin, J_8, 27-42.
Weintraub, S., Mesulam, M., & Kramer, L. (1981). Disturbance in
prosody: A right hemisphere contribution to language. Archives
of Neurology, 38, 742-744.
Weschler, A. F. (1973). The effect of organic brain disease on recall
of emotionally charged versus neutral narrative texts.
Neurology, 23, 130-135.
White, K., Sheehan, P. W., & Ashton, R. (1977). Imagery assessment:
A survey of self-report measures. Journal of Mental Imagery, J_,
145-170.
Whitehead, W. E., Drescher, V. M., Heiman, P., 4 Blackwell, B.
(1977). Relation of heart rate control to heart beat
perception. Biofeedback and Seif-Regulation, 2, 371-392.
Whitlock, F. A. (1982). Symptomatic affective disorders. New York:
Academic Press.
Winer, B. J. (1971). Statistical principles in experimental design
(2nd ed.). New York: McGraw-Hill.
Woods, D. J. (1977). Conjugate lateral eye movement, repression-
sensitization, and emotional style: Sex interactions. Journal
of Clinical Psychology, 33, 839-841.
Yokoyama, K., Jennings, R., Ackles, P., Hood, P., & Boiler, F.
(1987). Lack of heart rate changes during an attention demanding
task after right hemisphere lesions. Neurology, 37, 748-755.

BIOGRAPHICAL SKETCH
Cynthia Rodrigues Cimino was born May 11, 1958, in Marshfield,
Massachusetts. She earned her Bachelor of Science and Master of
Science degrees from the University of Florida. She will obtain her
Ph.D. in clinical and health psychology from the University of Florida
in December, 1988.

I certify that I have read this study and that in my opinion it
conforms to acceptable standards of scholarly presentation and is
fully adequate, in scope and quality, as a dissertation for the degree
of Doctor of Philosophy.
YJUM
Dawn Bowers
Associate Professor of Clinical and
Health Psychology
I certify that I have read this study and that in my opinion it
conforms to acceptable standards of scholarly presentation and is
fully adequate, in scope and quality, as a dissertation for the degree
of Doctor of Philosophy.
Hugh C. Qvis, Jr,
Professor of Clinical and Health
Psychology
I certify that I have read this study and that in my opinion it
conforms to acceptable standards of scholarly presentation and is
fully adequate, in scope and quality, as a dissertation for the degree
of Doctor of Philosophy.
^jJUtvu C5
Eileen B. Fennell
Professor of Clinical and Health
Psychology
I certify that I have read this study and that in my opinion it
conforms to acceptable standards of scholarly presentation and is
fully adequate, in scope and quality, as a dissertation for the degree
of Doctor of Philosophy.
M-
ussell M. Bauer
Associate Professor of Clinical and
Health Psychology

I certify that I have read this study and that in my opinion it
conforms to acceptable standards of scholarly presentation and is
fully adequate, in scope and quality, as a dissertation for the degree
of Doctor of Philosophy.
I certify that I have read this study and that in my opinion it
conforms to acceptable standards of scholarly presentation and is
fully adequate, in scope and quality, as a dissertation for the degree
of Doctor of Philosophy. x /
Professor of Neurology
This dissertation was submitted to the Graduate Faculty of the
Department of Clinical and Health Psychology in the College of Health
Related Professions and to the Graduate School and was accepted as
partial fulfillment of the requirements for the degree of Doctor of
Philosophy.
December, 1988
Dean, College of Health Related
Professions
Dean, Graduate School



25
have been consistently demonstrated in neurologic and psychiatric, as
well as normal, populations using a range of tasks including affective
facial judgements and production, emotional prosody judgements and
production, emotional memory, emotional verbal judgements, production
of affect, and humor appreciation.
Left Hemisphere Superiority for Positive Affect;
Right Hemisphere Superiority for Negative Affect
Early clinical reports of emotional/mood changes following brain
injury have also been interpreted as supporting a distinction between
the superiority of the left hemisphere in the processing of positive
affect and the superiority of the right hemisphere in the processing
of negative affect. Interpretation of these studies is based on the
notion of reciprocal inhibition which states that each hemisphere
exerts some degree of inhibition on the contralateral hemisphere
(Kinsbourne, 1973). In this view, damage to the left hemisphere would
result in disinhibition of the right hemisphere's negative affective
bias. Right hemisphere lesions would result in disinhibition of the
left hemisphere's positive affective bias.
Early observations suggested that RHD patients often appeared
indifferent or euphoric (Babinski, 1914; Denny-Brown et al., 1952).
In contrast, several authors reported that left hemisphere damage was
more associated with a "catastrophic" emotional response (Goldstein,
1952; Hecaen, 1962). These observations were later corroborated by
Gainotti (1972) in a large scale study of 160 patients. Based on
systematic investigation of patients' symptomatology, Gainotti


47
in differential patterns of arousal/activation, depending on the
hemisphere to which they were presented. According to the global
right hemisphere emotion model, emotional stimuli of any valence
directed to the right hemisphere should result in greater
arousal/activation responses than those directed to the left
hemisphere. According to valence models, negative emotional stimuli
(anger, fear, disgust) would induce greater arousal/activation
responses when directed to the right versus left hemisphere, whereas
the opposite should occur when positive stimuli (happiness) are
used. Finally, according to the preparatory/nonpreparatory model,
emotional stimuli having preparatory significance (anger, fear) should
result in greater arousal/activation responses when directed to the
right versus left hemisphere. The opposite should occur with
nonpreparatory emotional stimuli (happiness, disgust, neutral).
The focus of the present study was to further examine these
divergent views regarding the hemispheric processing of emotional
stimuli. The basic paradigm was one in which neutral and emotional
stimuli of different valence were laterally presented to either the
left or right hemisphere (using a tachistoscopic procedure). The
purposes of this study were to determine (a) the extent to which
laterally presented emotional/nonemotional stimuli might result in
differential patterns of behavioral activation (as assessed by
reaction time responses) as well as differential patterns of autonomic
arousal (as assessed by HR and SCR responses); (b) whether there were
hemispheric asymmetries in mediating arousal and/or activation in
response to these emotional/nonemotional stimuli; and (c) whether


120
Mach, E. (1959). The analysis of sensations, and the relation of the
physical to the psychological (S. Waterlow, Ed., and C. M.
Williams, Trans.). New York: Dover. (Original work published
in 1914)
Mahoney, A. M., & Sainsbury, R. S. (1987). Hemispheric asymmetry in
the perception of emotional sounds. Brain and Cognition, _6,
216-233.
McGlone, J. (1980). Sex differences in human brain asymmetry: A
critical survey. Behavioral and Brain Sciences, _3, 215-263.
McKeever, W. F., & Dixon, M. S. (1981). Right hemisphere superiority
of discriminating memorized from non-memorized faces: Affective
imagery, sex and perceived emotional effects. Brain and
Language, 12, 246-260.
Melamed, B. G. (1969). The habituation and psychophysiological
responses to tone, and to filmed fear stimuli under varying
conditions of instructional set. Unpublished doctoral
dissertation, University of Wisconsin.
Montgomery, W. A., & Jones, G. E. (1984). Laterality, emotionality
and heartbeat perception. Psychophysiology, 21, 459-465.
Morrow, L., Urtunski, P. B., Kim, Y., & Boiler, F. (1981). Arousal
responses to emotional stimuli and laterality of lesion.
Neuropsychologia, 19, 65-71.
Moruzzi, G., & Magoun, H. W. (1949). Brainstem reticular formation
and activation of the EEC. EEG and Clinical Neurophysiology, J_,
455-475.
Moskovitch, M. (1983). Local versus global solutions to problems of
hemispheric specialization. Behavioral and Brain Sciences, 6,
520-521.
Moskovitch, M. (1986). Afferent and efferent models of visual
perceptual asymmetries: Theoretical and empirical
implications. Neuropsychologia, 24, 91-114.
Moskovitch, M., & Olds, J. (1982). Asymmetries in spontaneous facial
expression and their possible relation to hemispheric
specialization. Neuropsychologia, 20, 71-81.
Myslobodsky, M. S., & Horesch, N. (1978). Bilateral electrodermal
activity in depressive patients. Biological Psychology, 6,
111-120.
Myslobodsky, M. S., & Rattok, J. (1977). Bilateral electrodermal
activity in waking man. Acta Psychologia, 41, 273-282.


37
The Model of Fox and Davidson
Fox and Davidson (1984) have proposed a developmental model of
neural substrates underlying affective response systems. Their model
is based on evolutionary considerations of the critical role of basic
approach and avoidance systems. These authors propose that approach
and avoidance comprise the two underlying behavioral dimensions upon
which all subsequent affective subsystems and responses have
evolved. In addition, they suggest that hemispheric specialization
constitutes the critical, neuroanatomical substrate of approach-
avoidance behavior. More specifically, these authors propose that the
left hemisphere is specialized for approach behaviors or positive
affects while the right hemisphere is more specialized for avoidance
behaviors or negative affects.
In support of tnis model, these authors relate the development of
hemispheric specialization and interhemispheric transfer to the
development of affective response systems. These authors argue that
all of the primary emotions emerge over the first year of life
(Bowlby, 1972; Charlesworth, 1 964; Izard, Hubner, Risser, McGuiness, &
Dougherty, 1980; Sroufe & Wunsch, 1972; Steinberg & Campos, 1983).
Subsequent to this, they propose that primary emotions are modified by
three processes: (a) the addition of new behaviors to the response
repertoire of the "affect program," (b) regulation in the form of
inhibition and appraisal, and (c) blending of primary emotions.
Interest and disgust are reliably elicited in the neonate (Izard,
1977), and Fox and Davidson (1984) cite evidence from EEG findings in
infants to suggest that these emotions are lateralized. They found


121
Nebes, R. D. (1974). Hemispheric specialization in commissurotomized
man. Psychological Bulletin, 81, 1-14.
Nottebohm, F. (1977). Asymmetries in neural control of vocalization
in the canary. In: S. Harnad, R. Doty, L. Goldstein, J. Jaynes,
& G. Krauthamer (Eds.), Lateralization in the nervous system.
New York: Academic.
Obrist, P. A., Howard, J. L., Lawler, J. E., Galosy, R. A., Meyers, K.
A., & Gaebelein, C. J. (1974). The cardiac-somatic
interaction. In: P. A. Obrist, A. H. Black, J. Brener, & L. V.
DiCara (Eds.), Cardiovascular psychophysiology: Current issues
in response mechanisms, biofeedback and methodology. Chicago:
Aldine-Atherton.
Oke, A., Keller, R., Mefford, I., & Adams, R. (1978). Lateralization
of norephinephrine in human thalamus. Science, 200, 1411-1413.
Oke, A., Lewis, R., & Adams, R. N. (1980). Hemispheric asymmetry of
norepinephrine distribution in rat thalamus. Brain Research,
188, 269-272.
Papanicolaou, A. C., Moore, B., Deutsch, G., Levin, H., & Eisenberg,
H. (1988). Evidence for right hemisphere involvement in recovery
from aphasia. Archives of Neurology, 45, 1025-1029.
Patterson, K., & Bradshaw, J. L. (1975). Differential hemispheric
mediation of nonverbal visual stimuli. Journal of Experimental
Psychology: Human Perception and Performance, _1_, 246-252.
Paul, H. (1909). Principien der sprachgeschichte (4th ed.). Germany:
Niemeyer.
Pennebaker, J. W., & Hoover, C. W. (1984). Visceral perception versus
visceral detection: Disentangling methods and assumptions.
Biofeedback and Self-Regulation, .9, 339-352.
Perria, P., Rosadini, G., & Rossi, G. F. (1961). Determination of
side of cerebral dominance with amobarbital. Archives of
Neurology, _4, 175-181.
Perris, C. (1974). Average evoked responses (AER) in patients with
affective disorders. Acta Psychiatrica Scandinavica
(Supplement), 225, 89_98.
Perris, C. (1975). EEG techniques in the measurement of the severity
of depressive syndromes. Neuropsychobiology, 1, 16-25.


Percent Correct ^ Percent Correct
69
0.95
0.85
0.75
Lgure 7.
100
90
80
70
Figure 8.
.938
1 1 1 1 1
Happy Angry Fearful Disgusting Neutral
Emotion
Experiment I--Percent Correct Analysis, Stimulus Type Main
Effect.
Trend
.957
i i i i i
Happy Angry Fearful Disgusting Neutral
Emotion
RVF
LVF
Experiment I--Percent Correct Analysis, Stimulus Type x
Visual Field Trend.


31
Mechanisms of Emotional Processing: The Role of Arousal
Schacter and colleagues (Schacter, 1964; Schacter & Singer, 1962)
proposed a theory of emotions termed the Cognitive-Arousal model.
According to this model, an emotional state is the product of an
interaction between arousal and cognition. An important assumption is
that both arousal and cognition are necessary components of emotion.
In this view, arousal is viewed as important in determining the felt
intensity of the emotion while the cognitive element is important in
determining the specific quality of the emotion.
Early support for the joint roles of arousal and cognition were
provided by Maranon (1924, cited in Fehr and Stern, 1970) who
artificially aroused subjects with administration of drugs that
stimulated the sympathetic nervous system. Maranon's subjects did not
report feeling emotions although some did report feeling "as if"
emotions. In contrast, if subjects were given a congitive set
(induction of an affective memory) they did report emotional reactions
when artifically induced arousal was present. Schacter (1970) has
also provided support for this notion in a study which looked at the
specific effects of pharmacologically induced arousal in neutral and
stressful situation. Schacter demonstrated that physiological arousal
alone (neutral condition) was not sufficient to evoke emotional
responses from subjects but that the combination of arousal and
availability of a cognitive label (stressful situation) was
necessary. Although Schacter's research has been heavily criticized,
especially on methodological grounds, this does not refute the


61
significant differences across LVF and RVF trials for female Ss [F_
(1, 14) = 3.71, £ = .0644)]. However, for male Ss, RVF (fi = 477.10)
were significantly faster than LVF (M = 485.49) trials [ F_ (1 14) =
5.23, £ = .0299)].
Findings also revealed a significant interaction of Sex x
Stimulus Type [F (4, 112) = 2.64, £ = .0376)], depicted in Figure 3.
Post-hoc simple effects testing revealed no significant differences
across Stimulus Type for female Ss [F (4, 56) = 1.26, £ = .294)].
However, male Ss showed a significant effect for Stimulus Type [F_
(4, 56) = 2.60, p = .0454)]. Duncan's post-hoc comparisons revealed
that for male Ss, RT responses to disgusting slides (_M = 506.28) were
significantly slower than happy (M = 468.29) or angry = 483.^7)
sLides at £ < .05.
A significant Hand x Visual Fieid x Stimulus Type interaction [F_
(4, 112) = 2.49, £ = .0473)] was also obtained and is depicted in
Figure 4. Post-hoc simple effects testing of RVF trials only revealed
no statistically significant differences across Hand and Stimulus Type
conditions [F_ (4, 116) = .97, £ = .4249)]. In LVF, however, post-hoc
simple effects tests revealed a significant Hand x Stimulus Type
interaction [F (4, 116) = 2.61, £ = .0390)]. Duncan's post-hoc
comparisons revealed that happy trials in LVF were significantly
faster when using the right (_M = 507.45) versus left (fl = 562.94) hand
at £ < .05.
Further inspection of the data revealed that the significant Sex
effects observed in the preceding analysis may have been influenced by
the markedly slowed performance of two female Ss. When compared to


Maximum Heart Rate Deceleration
(rom Baseline
78
Emotion
Figure 14. Experiment II--Maximum Deceleratory Heart Rate Analysis,
Sex x Stimulus Type Interaction.