<%BANNER%>

Emotion and Imagery

Permanent Link: http://ufdc.ufl.edu/UFE0021124/00001

Material Information

Title: Emotion and Imagery A Psychophysiological Analysis of Imagery Ability and Narrative Engagement
Physical Description: 1 online resource (89 p.)
Language: english
Creator: Mcteague, Lisa M
Publisher: University of Florida
Place of Publication: Gainesville, Fla.
Publication Date: 2007

Subjects

Subjects / Keywords: autonomic, electrocortical, electromyography, emotion, imagery, p300, psychophysiology, qmi, startle
Clinical and Health Psychology -- Dissertations, Academic -- UF
Genre: Psychology thesis, Ph.D.
bibliography   ( marcgt )
theses   ( marcgt )
government publication (state, provincial, terriorial, dependent)   ( marcgt )
born-digital   ( sobekcm )
Electronic Thesis or Dissertation

Notes

Abstract: Lang (1979) proposed that emotional imagery involves the mobilization of physiological systems involved in actual experience and as such, despite the difficulties imposed by the inherent subjectivity of mental events, could be reliably indexed as the physiological output associated with imagination. The validity of this premise has been consistently demonstrated in the emotional modulation of physiological systems indicative of: affective communication through facial action (facial EMG), autonomic mobilization (heart rate and skin conductance), and action readiness (startle responses). Study 1 of this investigation aimed to address whether a measure utilized in other paradigms to index attention allocation (probe P300) could viably reflect differences in attentional engagement during narrative imagery. While replicating and extending findings from previous studies on the modulation of facial EMG, autonomic measures, and reflexes as a function of pleasantness and arousal, probe P300 responses reliably varied with emotional arousal, suggesting that narrative imagery is a potent affective foreground that grabs natural selective attention and prioritizes the processing of the motivationally significant image at the expense of the intruding probe. Study 2 of this investigation examined whether ability to achieve vivid, mental imagery is a crucial individual difference that should be accounted for in efforts to delineate the foundations and processes of emotional imagery. Betts (1909) defined imagery ability in terms of the subjective experience of images conjured in different sensory modalities and accordingly devised the Questionnaire Upon Mental Imagery (QMI). Previous evidence on the association of the QMI and physiological reactivity during imagery is equivocal. To address whether physiological outcomes would improve the predictive validity of Betts? conceptualization of imagery ability and hence capture important variance in affective modulation during narrative imagery, a series of text prompts were adapted from the QMI to be presented as 'sensorimotor' imagery prompts in the context of an experimental physiological procedure-- a kind of 'psychophysiological' QMI. Neither self-reported imagery ability as assessed on the QMI or in the context of the physiological paradigm was systematically related to reactivity during narrative imagery. Findings are discussed in terms of potential alternative sources of variation in responding, highlighting individual differences in fearfulness and anxiety.
General Note: In the series University of Florida Digital Collections.
General Note: Includes vita.
Bibliography: Includes bibliographical references.
Source of Description: Description based on online resource; title from PDF title page.
Source of Description: This bibliographic record is available under the Creative Commons CC0 public domain dedication. The University of Florida Libraries, as creator of this bibliographic record, has waived all rights to it worldwide under copyright law, including all related and neighboring rights, to the extent allowed by law.
Statement of Responsibility: by Lisa M Mcteague.
Thesis: Thesis (Ph.D.)--University of Florida, 2007.
Local: Adviser: Lang, Peter J.
Local: Co-adviser: Bradley, Margaret M.
Electronic Access: RESTRICTED TO UF STUDENTS, STAFF, FACULTY, AND ON-CAMPUS USE UNTIL 2008-08-31

Record Information

Source Institution: UFRGP
Rights Management: Applicable rights reserved.
Classification: lcc - LD1780 2007
System ID: UFE0021124:00001

Permanent Link: http://ufdc.ufl.edu/UFE0021124/00001

Material Information

Title: Emotion and Imagery A Psychophysiological Analysis of Imagery Ability and Narrative Engagement
Physical Description: 1 online resource (89 p.)
Language: english
Creator: Mcteague, Lisa M
Publisher: University of Florida
Place of Publication: Gainesville, Fla.
Publication Date: 2007

Subjects

Subjects / Keywords: autonomic, electrocortical, electromyography, emotion, imagery, p300, psychophysiology, qmi, startle
Clinical and Health Psychology -- Dissertations, Academic -- UF
Genre: Psychology thesis, Ph.D.
bibliography   ( marcgt )
theses   ( marcgt )
government publication (state, provincial, terriorial, dependent)   ( marcgt )
born-digital   ( sobekcm )
Electronic Thesis or Dissertation

Notes

Abstract: Lang (1979) proposed that emotional imagery involves the mobilization of physiological systems involved in actual experience and as such, despite the difficulties imposed by the inherent subjectivity of mental events, could be reliably indexed as the physiological output associated with imagination. The validity of this premise has been consistently demonstrated in the emotional modulation of physiological systems indicative of: affective communication through facial action (facial EMG), autonomic mobilization (heart rate and skin conductance), and action readiness (startle responses). Study 1 of this investigation aimed to address whether a measure utilized in other paradigms to index attention allocation (probe P300) could viably reflect differences in attentional engagement during narrative imagery. While replicating and extending findings from previous studies on the modulation of facial EMG, autonomic measures, and reflexes as a function of pleasantness and arousal, probe P300 responses reliably varied with emotional arousal, suggesting that narrative imagery is a potent affective foreground that grabs natural selective attention and prioritizes the processing of the motivationally significant image at the expense of the intruding probe. Study 2 of this investigation examined whether ability to achieve vivid, mental imagery is a crucial individual difference that should be accounted for in efforts to delineate the foundations and processes of emotional imagery. Betts (1909) defined imagery ability in terms of the subjective experience of images conjured in different sensory modalities and accordingly devised the Questionnaire Upon Mental Imagery (QMI). Previous evidence on the association of the QMI and physiological reactivity during imagery is equivocal. To address whether physiological outcomes would improve the predictive validity of Betts? conceptualization of imagery ability and hence capture important variance in affective modulation during narrative imagery, a series of text prompts were adapted from the QMI to be presented as 'sensorimotor' imagery prompts in the context of an experimental physiological procedure-- a kind of 'psychophysiological' QMI. Neither self-reported imagery ability as assessed on the QMI or in the context of the physiological paradigm was systematically related to reactivity during narrative imagery. Findings are discussed in terms of potential alternative sources of variation in responding, highlighting individual differences in fearfulness and anxiety.
General Note: In the series University of Florida Digital Collections.
General Note: Includes vita.
Bibliography: Includes bibliographical references.
Source of Description: Description based on online resource; title from PDF title page.
Source of Description: This bibliographic record is available under the Creative Commons CC0 public domain dedication. The University of Florida Libraries, as creator of this bibliographic record, has waived all rights to it worldwide under copyright law, including all related and neighboring rights, to the extent allowed by law.
Statement of Responsibility: by Lisa M Mcteague.
Thesis: Thesis (Ph.D.)--University of Florida, 2007.
Local: Adviser: Lang, Peter J.
Local: Co-adviser: Bradley, Margaret M.
Electronic Access: RESTRICTED TO UF STUDENTS, STAFF, FACULTY, AND ON-CAMPUS USE UNTIL 2008-08-31

Record Information

Source Institution: UFRGP
Rights Management: Applicable rights reserved.
Classification: lcc - LD1780 2007
System ID: UFE0021124:00001


This item has the following downloads:


Full Text

PAGE 1

1 EMOTION AND IMAGERY: A PSYCHOPHYSIOLOGICAL ANALYSIS OF IMAGERY ABILITY AND NARRATIVE ENGAGEMENT By LISA M. MCTEAGUE A DISSERTATION PRESENTED TO THE GRADUATE SCHOOL OF THE UNIVERSITY OF FLOR IDA IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF DOCTOR OF PHILOSOPHY UNIVERSITY OF FLORIDA 2007

PAGE 2

2 2007 Lisa M. McTeague

PAGE 3

3 To my parents, Kevin and Patricia McTeague

PAGE 4

4 ACKNOWLEDGMENTS Foremost I thank Peter J. Lang and Margaret M. Bradley for their expertise, guidance, and patience in mentoring not only th is project but also the many e ndeavors throughout my graduate career. I also thank the two members in addi tion to Drs. Lang and Bradley on my doctoral committee, Ronald H. Rozensky and Jeffrey R. Fitzsimmons, for their valuable and perspicacious input on my dissert ation project. I furthermore ex tend my profound gratitude to the members of the NIMH Center for the Study of Emotion and Attention for their enduring contributions, collaboration, and support: Andr eas Loew, Marie-Claude Laplante, Bethany Wangelin, Vincent Costa, Joshua Shumen, Andreas Keil, Dean Sabatinelli, Francesso Versace, Dorothea Roebuck, Vera Ferrari, Aliss on Bittiker, and Robert Sivinski. This research was generously supported by gr ants from the National Institute of Mental Health (F31MH069048-01 and P50MH072850). To my parents, Kevin and Patricia McTea gue, whose consistent support and emphasis on the value of education undergir d my intellectual pursuits, I ex tend inestimable appreciation. I also offer heartfelt thanks to Jennifer and Shauna McTeague, Eleni Dimoulas, and Michael Antonucci.

PAGE 5

5 TABLE OF CONTENTS page ACKNOWLEDGMENTS...............................................................................................................4 LIST OF TABLES................................................................................................................. ..........7 LIST OF FIGURES................................................................................................................ .........8 ABSTRACT....................................................................................................................... ............10 CHAPTER 1 INTRODUCTION..................................................................................................................12 Mental Imagery in Clinical Fear and Anxiety........................................................................12 Bioinformational Model of Emotion......................................................................................13 Imagery as a Window to Emotional Experience.............................................................14 Emotional Imagery and Physiological Reactivity...........................................................15 Affect communication through facial action............................................................15 Autonomic mobilization...........................................................................................16 Action readiness: The probe startle response...........................................................17 Extending the Assessment of Narrative Imagery: Electrocortical Activity as a Potential Index of Attentional Engagement.......................................................................................19 Imagery Ability: An Important Predictor of Emotional Reactivity?......................................21 Narrative Imagery: Hypothe ses and Expected Findings........................................................23 Sensorimotor Imagery and Subjective Im agery Ability: Exploratory Analyses....................24 Associations between Imagery Ability and Narrative Imagery: Exploratory Analyses.........25 2 EXPERIMENT 1: NARRATIVE IMAGERY.......................................................................26 Method......................................................................................................................... ...........26 Participants................................................................................................................... ...26 Materials and Design.......................................................................................................26 Procedure...................................................................................................................... ...27 Physiological Response Measurement............................................................................28 Data Reduction................................................................................................................30 Data Analysis.................................................................................................................. .31 Results........................................................................................................................ .............32 Participant Characteristics...............................................................................................32 Narrative Imagery: Evaluative Judgments......................................................................32 Affect communication through facial action............................................................34 Autonomic mobilization...........................................................................................35 Action readiness: The probe startle response...........................................................36 Attentional engagement: P300 to the probe.............................................................37 Discussion: Narrative Imagery...............................................................................................38 Affect Communication th rough Facial Action................................................................38

PAGE 6

6 Autonomic Mobilization.................................................................................................39 Action Readiness: The Pr obe Startle Response...............................................................40 Attentional Engagement: P300 to the Probe...................................................................40 Conclusions.................................................................................................................... .41 3 EXPERIMENT 2: SENSORIMOTOR IMAGERY...............................................................56 Method......................................................................................................................... ...........56 Participants................................................................................................................... ...56 Materials and Design.......................................................................................................56 Procedure...................................................................................................................... ...57 Data Reduction................................................................................................................58 Data Analysis.................................................................................................................. .58 Results........................................................................................................................ .............59 Participant Characteristics...............................................................................................59 Physiological Effects of Prompt Intensity and Sensory Modality...................................59 Subjective Report of Imagery Vividne ss During Sensorimotor Imagery.......................60 Subjective Report of Imager y Vividness: QMI Scores...................................................61 Good and Poor Imagers: QMI Score as a Predictor of Sensorimotor Imagery........61 Good and Poor Imagers: QMI Score as a Predictor of Narrative Imagery..............62 Good and Poor Imagers: Psychophys iological QMI as a Predictor of Narrative Imagery........................................................................................................62 Consistent and Inconsistent Imagers: Consistency of Reactivity During Sensorimotor Imagery as a Pr edictor of Narrative Imagery........................................63 Discussion: Sensorimotor Imagery.........................................................................................64 4 GENERAL DISCUSSION.....................................................................................................75 Imagery Ability and Reactivity to Sensory and Emotional Imagery: Meaningful Associations?.................................................................................................................. ....75 Imagery Ability, Psychopathology, and Emotional Reactivity..............................................76 Startle Reflex Modulation in Sensorimotor Imagery: Implications for Understanding Startle Reflex Facilitatio n during Narrative Imagery..........................................................77 LIST OF REFERENCES............................................................................................................. ..80 BIOGRAPHICAL SKETCH.........................................................................................................87

PAGE 7

7 LIST OF TABLES Table page 2-1 Text of scripts for psychophysiological ex amination of affective narrative imagery.......42 2-2 Narrative imagery: Mean response (standa rd deviation) across measures averaged by valence category............................................................................................................... ..44 2-3 Narrative imagery: Mean response (standa rd deviation) across measures averaged by content category............................................................................................................... ..45 3-1 Text of scripts for psychophysiological experimental test of imagery ability...................67 3-2 Sensorimotor imagery: Mean response (standard deviation) across measures averaged by intensity.........................................................................................................68 3-3 Effects of sensorimotor responder st atus and valence category on physiology and evaluative ratings of narrative imagery..............................................................................69

PAGE 8

8 LIST OF FIGURES Figure page 2-1 Trial structure for narrative imagery..................................................................................46 2-2 Mean ratings of pleasure and arousal as prompted by pleasant, neutral, and unpleasant scripts measured with the Self-Assessment Manikin......................................47 2-3 Mean ratings of pleasure and arousal by narrative imagery content category as measured with the Self-Assessment Manikin....................................................................47 2-4 Average change in corrugator EMG activity by valence of imagined prompt..................48 2-5 Average 12-second change in corrugator EMG activity during imagery by category of imagined prompt............................................................................................................48 2-6 Average change in zygomatic EMG ac tivity by valence of imagined prompt..................49 2-7 Average change in orbicularis EMG activity by valence of imagined prompt..................50 2-8 Average change in orbicularis EMG activity during neutra l scenes and those emotional scenes that during narrativ e imagery evoked significant increases compared to neutral............................................................................................................51 2-9 Average change in heart rate by valence of imagined prompt...........................................52 2-10 Mean blink magnitude to the st artle probe during narrative imagery................................53 2-11 Mean peak skin conductance response to the startle probe during narrative imagery by valence of imagined prompt..........................................................................................54 2-12 Event-related averages to the startle pr obe at electrodes Cz and Pz 100 ms preto 400 ms post-probe by valence of imagined prompt...........................................................55 2-13 Scalp potential at 11 cen tro-parietal sensors 260-340 ms after the startle probe by valence of imagined prompt...............................................................................................55 3-1 Trial structure for sensorimotor imagery...........................................................................70 3-2 Average physiological change during low and moderate intensity sensorimotor imagery........................................................................................................................ ......71 3-3 Average orbicularis EMG change duri ng during sensorimotor imagery by sensory modality....................................................................................................................... ......72 3-4 Average corrugator EMG change during during sensorimotor imagery by sensory modality....................................................................................................................... ......72

PAGE 9

9 3-5 Average experimental ratings of imagery vividness by sensory modality........................73 3-6 Imagery vividness as rated on the Qu estionnaire Upon Mental Imagery (QMI) averaged by sensory modality............................................................................................73 3-7 Mean blink magnitude to the startle pr obe during narrative imagery for individuals classified as inconsiste nt and consistent physio logical responders during sensorimotor imagery.........................................................................................................74

PAGE 10

10 Abstract of Dissertation Pres ented to the Graduate School of the University of Florida in Partial Fulfillment of the Requirements for the Degree of Doctor of Philosophy EMOTION AND IMAGERY: A PSYCHOPHYSIOLOGICAL ANALYSIS OF IMAGERY ABILITY AND NARRATIVE ENGAGEMENT By Lisa M. McTeague August 2007 Chair: Peter J. Lang Major: Psychology Lang (1979) proposed that emotional imager y involves the mobilization of physiological systems involved in actual experience and as such, despite the difficulties imposed by the inherent subjectivity of mental events, could be reliably indexed as the physiological output associated with imagination. The validity of this premise has been consistently demonstrated in the emotional modulation of physiological sy stems indicative of: affective communication through facial action (facial EMG), autonomic m obilization (heart rate and skin conductance), and action readiness (startle responses). Study 1 of this investigation aimed to address whether a measure utilized in other paradigms to index attention allocation (probe P300) could viably reflect differences in attenti onal engagement during narrative imagery. While replicating and extending findings from previous studies on the m odulation of facial EMG, autonomic measures, and reflexes as a function of pleasantness and arou sal, probe P300 responses reliably varied with emotional arousal, suggesting that narrative imagery is a potent affective foreground that grabs natural selective attention and prioritizes the processing of the motivationally significant image at the expense of the intruding probe.

PAGE 11

11 Study 2 of this investigation examined whether ability to achieve vivid, mental imagery is a crucial individual difference that should be account ed for in efforts to delineate the foundations and processes of emotional imagery. Betts (1909 ) defined imagery abili ty in terms of the subjective experience of images conjured in different sensory modalities and accordingly devised the Questionnaire Upon Mental Imagery (QMI). Previous evid ence on the associ ation of the QMI and physiological reactivity during imagery is equivocal. To address whether physiological outcomes would improve the predictive validity of Betts conceptualization of imagery ability and hence capture important variance in affec tive modulation during narrative imagery, a series of text prompts were adapted from the QMI to be presented as sensorimotor imagery prompts in the context of an experimental physiol ogical procedurea kind of psychophysiological QMI. Neither self-reported imagery ability as assessed on the QMI or in the context of the physiological paradigm was systematically re lated to reactivity during narrative imagery. Findings are discussed in terms of potential alterna tive sources of vari ation in responding, highlighting individual differences in fearfulness and anxiety.

PAGE 12

12 CHAPTER 1 INTRODUCTION From the perspective of natu ral science, human emotions include three measurable response classes: verbal reports of experien ce, overt actions, and associated physiological mobilization. Several theorists have suggest ed (Bradley, 2000; Dickinson & Dearing, 1979; Konorski, 1967) that primitive survival reflexes are the foundation for emotions physiological mobilization and action. That is, humans and other animals approach pleasant things that sustain life (appetitive motivation) and fi ght or flee in the face of threat s to their continued existence (defensive motivation). Humans, however, seldom react as directly as do less complex species. With the development of the cerebral cortex, emer ged a greater capacity for inhibition and delay, and for assessing alternatives and outcomes. Nevertheless, the primitive reactance is yet adumbrated in muscles and glands, supported by neur al circuits deep within the brain and widely shared among species. For this reason, emotions (fear and anger; joy and desire) are action dispositions (Frijda, 1986; Lang, Bradley, & Cut hbert, 1997) and as such are often most evident when we are overtly passive, but mobilized soma tically and autonomically for actions that may never actually take place. Mental Imagery in Clinical Fear and Anxiety One of the conditions in which this is eviden t is mental imagery of human experience, and thus, has been used extensively in treatment to reinstate clinically-releva nt emotional experiences (e.g., counter-conditioning, extin ction). Emotional imagery, in fact is an essential component in conventional behavioral methodologies implemente d to treat the entire anxiety spectrum (e.g., specific phobia (Craske Antony, & Barlow, 2006), pos ttraumatic stress disorder (Foa, Hembree, & Rothbaum, 2007), generalized anxiety disord er (Zinbarg, Craske, & Barlow, 2006)). Lang, Melamed, & Hart (1970) demonstrated the physiol ogical activation inhere nt during clinically-

PAGE 13

13 pertinent image processing by r ecording a desensitizat ion intervention implemented in two samples of phobics. The authors found in the first experiment that fearful imagery evoked elevated autonomic arousal that reduced with repeated processing. Furthermore, participants who showed greater autonomic mobilization to fear scenes also showed gr eater reduction over the course of the intervention and ultimately indi cated more subjective de sensitization. In a followup study the authors discovered that the fear hier archies developed with each participant for desensitization, yielded autonomic gradients that varied with s ubjective fearfulness and imagery vividness. Essentially phobics showed increased autonomic reactivity coincident with verbal report of fear and image clarity and furtherm ore those individuals demonstrating the most correspondence between verbal evaluation and efferent physio logy experienced the greatest therapeutic gains. Taken, together these da ta suggest that physiol ogical reactiv ity during emotional imagery provides useful information for delineating the phenomenology of fear and anxiety, importantly including e nhanced diagnostic and prognos tic impressions. The current investigation is an effort to further develop im agery-based experimental procedures intended for translation to the clinical environment, focu sing on both expanding the breadth of physiological measurement and emotional content processed as well as considering the role of imagery vividness in reactivity. Bioinformational Model of Emotion Relying on a propositional network model, La ng (1979) posited that emotions result from the activation of associated neur al networks, which code inform ation according to both sensory and semantic properties. Deviating from earlier models that focused on sensory intake, Lang additionally emphasized the role of relevant re sponse units in the dist ributed structure and function of an emotional network. These respon se units code for outcomes such as overt behavior, efferent physiology, and emotional language, including both expressive and evaluative

PAGE 14

14 responses (Lang, Cuthbert, & Bradley, 1998). This th eory implies that the parameters of input stimulation will necessarily influence the degree of activation of select network units, the spread of activation across multiple nodes or units, a nd ultimately, determine the strength of the response. Additional predictive parameters are the specific demands of the eliciting context, which may limit the behavioral repertoire (e.g., escape, avoidance, freezing). In summary, an emotional episode is determined by cues that activate stimulus, meaning, and response representations in memory and the consequent output as demonstrated in behavior, physiology, and language. Imagery as a Window to Emotional Experience Lang (1977) suggested that emotional imager y might be a mental phenomenon that would allow a window for systematic investigati on into affective acti on dispositions. Langs conceptualization makes the distinction that simply tapping into one modality such as imagining a motor movement (e.g., Neuper & Pfurtscheller, 2001) or imagining a sensory percept (e.g., Kosslyn, Ganis, & Thompson, 2001) is insufficient for emotional experience. Rather, emotional imagery involves the convergence of the neural networks and physiology involved in an actual emotional experience (i.e., sensory experience, physiological changes, semantic appraisals, behavioral dispositions). Although psychophysiological re search on imagery has a long history (Jacobson, 1931) involving many investigators, the most sustained research program has been carried forward by Lang and colleagues, endeavoring to test the bioinformational a nd biphasic models of emotion (e.g., Cuthbert et al., 2003; Lang, Levin, Miller, & Kozak, 1983; Miller et al., 1987; Vrana & Lang, 1990; Weerts & Lang, 1978). Over the course of three decades, emotional imagery has been utilized to characterize the contributions of varied physiological systems to affective profiles in community and college sample controls as well as anxiety and mood disorder patients.

PAGE 15

15 For a single trial in the typical imagery design, a participant memori zes or listens to an emotional or neutral scene described with a combination of stimulus, meaning, and response elements (e.g., The large snake darts forward, fangs protruding, striking my leg in a flash of pain). Then following a tone cue, the participant imagines being personally involved in the ongoing scene until an additional tone signals the offset of imagery. In accordance with the bioinformational model, the differences in efferent physiology to fearful versus neutral imagery are accentuated when physiological response prope rties (e.g., your heart pounded) are included in imagery text prompts focusing on stimulus properties (Lang, Ko zak, Miller, Levin, & Mc Lean, 1980; Miller et al., 1987). Prompts depicting scenes of higher arousal, eith er through inclusion of robust response properties, exciting contex tual descriptions, a nd/or intense perceptu al experiences, have reliably shown enhanced affective recr uitment (e.g., Cuthbert et al., 2003). Emotional Imagery and Physiological Reactivity The extent of affective motivation during em otional imagery has been shown to activate facial muscles involved in affect communicati on (e.g., corrugator, zygomatic, and orbicularis EMG); to mobilize the autonomic nervous system (e.g., heart rate, skin conductance); and to ready the somatic system for action (e.g., potenti ation of the startle reflex). The imagery experiments presented here assess all of these aspects of emotion activation. The following is a brief review of previous research that examin ed these changes in physio logical reactivity during emotional processing. Affect communication th rough facial action Corrugator supercilii. The corrugator supercilii muscles ar e responsible for contraction of the eyebrows, and when the motor response is large enough, produce an identifiable frown (Tassinary & Cacioppo, 2000). Corr ugator EMG activity varies linear ly with hedonic valence, with the largest responses el icited during unpleasant imaginin g, intermediate tension during

PAGE 16

16 neutral and the least activity for pleasant imag ery. In fact, the corruga tor muscle often shows relaxation below baseline dur ing pleasant processing (La ng, Greenwald, Bradley, & Hamm, 1993; Larsen, Norris & Cacioppo, 2003; McTea gue, Bradley, & Lang, 2002; McTeague, Dimoulas, Strauss, Bradley, & Lang, 2003). Zygomatic major. The zygomaticus major muscles pull the corners of the mouth into a smile (Larsen et al., 2003) and are most active du ring highly pleasant proce ssing (e.g., pictures or imagery of nurturance or food) Interestingly, the most highl y pleasant stimuli (e.g., cuddling babies) are not the most arousing (erotica) and, hence, zygomatic EMG change covaries more closely with valence than ar ousal. Zygomatic EMG change also emerges during highly unpleasant stimuli such as mutilations and contamination, often associated with subjective reactions of disgust. Orbicularis oculi. The orbicularis oculi muscles are re sponsible for the blink response to abrupt startle probes. Additionally continuous measurement of the muscle throughout the imagery period has shown strong, sustained linear cova riation with arousal in that neutral scenes elicit the least change from baseline, followed by low and then high arousal scenes irrespective of valence (Bradley, Cuthbert, & Lang, 1995; McTeague et al., 2006). Autonomic mobilization Heart rate. Changes in heart rate can reflect activa tion in either the parasympathetic or sympathetic nervous systems (e.g., Brownley, Hurwitz, & Schneiderman, 2000). During a six second picture-processing period, a triphasic waveform results with an initial de celeration, then acceleration, and secondary deceleration. Av eraged cardiac waveforms show relative deceleration for unpleasant pictures and a relati ve acceleration for pleasant pictures (Hamm, Greenwald, Bradley, & Lang, 1993). In contrast heart rate activity during imagery has consistently shown acceleration (v agal release and sympathetic activation) during both pleasant

PAGE 17

17 and unpleasant imagining re lative to neutral with greater acceleration duri ng scenes rated higher in arousal (e.g., Cook, Hawk, Davis, & Stev enson, 1991; Patrick, Cuthbert, & Lang, 1994; Witvliet & Vrana, 2000). For example, imagery of sports victory and att acking animals reliably prompt heart rate acceleration (e.g., Bradley et al., 1995; McTeague et al., 2002). Emphasizing the action-oriented process of narrative imag ery, Lang (1977, 1979) sugges ted that the relative heart rate acceleration during imagery might be due to the activation of a perceptual-motor memory that cues somatic response units reminis cent of actual experience. In contrast, picture viewing is a passive, observationa l task and averaged waveforms suggest an overall orienting deceleration (Bradley, 2000) with preferentially sustained orien ting for stimuli conferring threat, starkly opposite to the pronounced heart rate acceleration during narrativ e imagery of similar content. Skin conductance level. Eccrine sweat glands are innerv ated by the sympathetic nervous system and, as such, increased skin conductance due to sweating is considered a measure of sympathetic activity of the autonomic nervous system (e.g., Dawson, Schell & Filion, 2000). During narrative imagery (e.g., Cook et al., 1991; Cuthbert et al., 2003 ; Miller, Patrick, & Levenston, 2002) protracted changes in skin con ductance level (SCL) vary linearly with rated emotional arousal, independent of affective va lence; highly arousing pleasant and unpleasant relative to neutral imagining produces larger increases. The magnitude of skin conductance responses to highly pleasant and unpleas ant stimuli often does not differ. Action readiness: The probe startle response Whereas heart rate, skin conductance, corru gator, zygomatic, and orbicularis EMG are somatic muscle changes naturally evoked by pe rception and imagery, the startle probe is a separate stimulus administered during foregr ound affective processing. The magnitude of the reflexive eyeblink to a probe stim ulus is utilized as an index of motive system dominance (i.e.,

PAGE 18

18 appetitive or defensive motivation) as effere nts from the central nucleus of the amygdala influence startle magnitude via synapses on th e nucleus reticularis pontis caudalis (Davis & Lang, 2001). Similar to the elevated whole body startle observed in rats following shock sensitization or conditioning (Davis, 2000), humans s how reliably potentiated blink responses to acoustic startle probes, most discernibly in orbicularis EMG activity. Startle response potentiation has been hypothesized to reflect activation of the defensive motivational system, thereby accentuating the defensive reflex (shoc k; Greenwald, Bradley, Cuthbert, & Lang, 1998; Hamm et al., 1993; pictures; La ng, Bradley, & Cuthbert, 1990). Recent studies have shown that this pattern of affective modulation varies according to the type of cognitive processing invoked. During pict ure perception, the acoustic startle reflex is potentiated when viewing aversi ve compared to neutral pict ures and, conversely, inhibited during viewing of pleasant rela tive to neutral pictures (L ang, Bradley, & Cuthbert, 1990). However, during imagery augmented startle responding has been found for both high arousing, pleasant and unpleasant compared to neutral imagery (Bradley et al., 1995; Cook et al., 1991; McTeague et al., 2003; Miller et al., 2002; W itvliet & Vrana, 1995; 2000). Bradley et al. (1995) suggested that similar to the in terpretation of autonomic output dur ing imagery startle facilitation during arousing narrative scenarios reflects th e motor response dispositions engaged during imagined action. However, other investigators (M iller et al., 2002) have posited that augmented startle during imagery is a function of attenti on, independent of emoti on. In short the underlying mechanisms are still under deba te and further investigation is necessary to disaggregate the processes yielding response differences duri ng picture and imagery processing. The current investigation attempts to incr ease the existing database on st artle reflex modulation during imagery with the inclusion of a larger stimulus set more thoroughly represen tative of valence and

PAGE 19

19 arousal dimensions as well as including more simultaneous recordings of other physiological systems, which can potentially clarify em otional and attentional contributions. In addition to the blink reflex, the star tle probe prompts a punctate superimposed deflection in skin conductance level. Vrana and colleagues (Vrana, 1995; Witvliet & Vrana; 1995) have demonstrated that similar to star tle responding, larger magni tude responses occur secondary to acoustic startle probes delivered dur ing narrative imagery of both arousing pleasant and unpleasant scenes further verifying that ongoi ng affective processing modulates the startle reflex. Extending the Assessment of Narrative Imagery : Electrocortical Acti vity as a Potential Index of Attentional Engagement One of the primary goals of this investig ation was the examination of event-related potentials to the startle probes elicited during imagery. Event-related potentials (ERPs) are the aspects of the electrococortical potential that are specifically time-lock ed to events and are regarded as manifestations of br ain activities that occur in prepar ation for, or in response to discrete events. ERPs have subcomponents with time-varying fields resulting from summation of electromagnetic activity generated by neural populations in differe nt parts of the brain (Fabiani, Gratton, & Coles, 2000). Component s are operationalized as pa rt of a waveform with a circumscribed scalp distribution and a theoretical ly circumscribed relationship to manipulated experimental variables (Handy, 2005). Of note, f eatures of the waveform such as peaks and troughs can result from the summation of seve ral contributing sources and therefore do not represent functionally homogeneous neural or cognitive processes. Nonetheless, ERP components are usually defined in terms of bot h proposed functional si gnificance and underlying neural sources.

PAGE 20

20 The P300 is a component of the ERP that peaks approximately 300 ms after stimulus onset, is maximum at parietal lo cations, and has been reliably show n to increase with the extent of directed attention toward a task. It has b een proposed to reflect attention allocation and context-updating of the environment (Donchin & Coles, 1988). For example, during pictureviewing, the onset of pleasant and unpleasant co mpared to neutral slides results in more positivity (e.g., Amrhein, Muhlberger, Pauli, & Weidemann, 2004; Keil et al., 2007) suggesting increased attention to motivationally-relevant pi ctures. However, when a secondary, unrelated interrupting stimulus such as an acoustic star tle probe is delivered during the same pictureviewing task, the resultant P300 amplitude is attenuated during both defensive and appetitive perception relative to neutral content (Cut hbert, Schupp, Bradley, McManis, & Lang, 1998; Keil et al., 2007; Schupp, Cuthbert, Bradley, Birbaume r, & Lang, 1997), consistent with a hypothesis of sustained attention: W ith greater resource allocation to the in teresting, arousing picture, less is available for processing the irrelevant, acous tic probe stimulus, attenuating the P300. Of particular relevance to the current i nvestigation, Schupp et al., (1997) found that not only were probe P300 responses dur ing affective pictures less positive, but also this difference from neutral pictures persisted af ter picture offset when participan ts were instructed to sustain mental images of the pictures. Thus mental imag ery of the emotional cont ent appeared to yield effects similar to those during actual perception. Although no data has been published concerni ng the influence of affective narrative imagery on the P300 to a startle probe it is expect ed based on picture-proc essing results that as affective arousal and hence motivational releva nce increases during emotional imagery, probe P300 amplitude will show the same attentional d ecrease indicating engagement in the imagery task.

PAGE 21

21 Imagery Ability: An Important Pred ictor of Emotional Reactivity? The second study in this investig ation addresses the issue of wh ether the ability to achieve vivid, mental imagery is a crucia l individual difference that should be accounted for in efforts to delineate the foundations and processes of em otional imagery. Betts (1909) defined imagery ability in terms of the subjective experience of images conjured in different sensory modalities. The Questionnaire Upon Mental Imagery (Betts 1909; revised Sheehan, 1967) was accordingly devised and is now a commonly us ed subjective measure of imager y ability or fidelity. On the 35-item self-report form participants are asked to generate images of stimuli from seven different sensory modalities (visual, auditory, olfact ory, cutaneous, kinesthetic, gustatory, and interoceptive/organic) and to rate each image in terms of vividness in comparison to the clarity of actual experience. Although c onstruct validity is undetermined due to the fundamental subjectivity of mental imagery, the QMI has demonstrated accepta ble reliablity (White, Sheehan, & Ashton, 1977). In previous imagery studies subjective im agery ability as assessed on the QMI has sometimes been related (e.g., Arabian & Furedy, 1983; Lang et al., 1970; Le vin, 1978; Miller et al., 1987) and sometimes not (e.g., McTeague et al., 2002, Witvliet & Vrana, 1995) to physiological reactivity. In terms of demonstrating a reliable asso ciation between imagery clarity and physiological reactivity, Wh ite (1978) found that while imagining food, good as opposed to poor imagers showed strong correspondence betw een salivation level and food preference. Marks (1972) found that during imagery of a sta tic object good and poor imagers did not differ in eye movements. However, when instructed to imagine an action-oriented scene, no change in responsivity was evident in poor imagers wher eas good imagers evinced increased oculomotor activity, which incremented further when scene pr ompts depicted eye movement. In the context of Pavlovian conditioning, Arabian and Fure dy (1983) discovered that when good imagers

PAGE 22

22 imagined a previous unconditioned stimulus (i.e ., 45 degree below horizontal tilt) sustained heart rate deceleration was evident for ten seconds, th e duration of the actual tilt condition; poor imagers showed equivalent heart rate decelera tion but less prolonged and hence less consistent with the in vivo exposure. Specific to narrative im agery, Lang et al. (1970) found that heart rate increase during fearful imagery was positively correlated with ratings of imagery clarity. Similarly, Van Diest and colleagues (2001) f ound that suggestive of respiratory action mobilization, high arousal compared to low arousa l scenes yielded drops in end-tidal fractional carbon dioxide concentration, most pronounced in better imagers. Similarly, Levin (1978) discovered greater heart rate elevations among good compared to poor imagers during fearful imagery. Miller et al. (1987) assessed good and poor imagers before and after training that encouraged somatovisceral recruitment during imag ery. Both at baseline, but even more so after training good imagers showed stronger affective modulation during imagery as indexed in skin conductance, heart rate, and ocul omotor activity. In short, subj ective imagery ability has been shown to discriminate individuals based on the consistency of their efferent physiology during imagery to implied action and experience. Imagery ability has, however, also shown in consistent associations to physiological reactivity during imagining. Thus, although, Levi n (1982) found that good compared to poor imagers showed more robust heart rate increases during imagery of feared scenarios, the pattern for electrodermal responding was just the oppositepoor imagers showed larger skin conductance increases and no group differences em erged in sternomastoid muscle tension. Cook, Melamed, Cuthbert, McNeil, & Lang (1988) found that imagery ability interacted with specific script content in the prediction of visceral activity. In particular, during dental imagery good imagers showed larger heart rate increases whereas no group difference was observed for speech

PAGE 23

23 imagery or for electrodermal response to either content. More recently, McTeague et al. (2002) found no associations between imagery ability and reactivity during emotional imagery as indexed in autonomic (heart rate, skin conductance) facial EMG (corrugator and orbicularis) or blink magnitude to the acoustic startle probe. The second experiment reported here is an effo rt to further explore the view that imagery ability is a significant moderator variable that should be considered in evaluating imagery processing of emotional content. To this end a se ries of text prompts were adapted from the QMI to be presented as sensorimotor imagery prompts in the context of an experimental physiological procedurea kind of psychophysi ological Questionnaire Upon Mental Imagery. Guided by Betts conceptualization but tran slated into the outcomes of psychophysiology, individuals who demonstrated greater physiological activity during imagery as a function of the experiential intensity (e.g., walk ing vs. running) of the depicted scene were operationalized as physiologically vivid imagers. Reactivity durin g the experimental adaptation of the QMI was correlated with subjective imagery vividness, and in turn both measures were related withinsubject to patterns of em otional reactivity during more elabor ate narrative imagery. In short, this design was implemented to address whether physiological outcomes would improve the predictive validity of Betts conceptualization of imagery ability. Narrative Imagery: Hypotheses and Expected Findings Participants imagined narrative scenarios that differed in pleasure and arousal while the previously described physiological measures we re recorded. In additi on evaluative ratings (i.e., pleasure and arousal) were collected for each imagined scene. Specific imagery scenarios corresponded to seven content categories: social reinforcement, high arousal pleasant, neutral, social threat, pani c, contamination, and attack/danger. Attentional Engagement. Probe P300 responses were an ticipated to decrease during emotionally arousing images reflecting gr eater engagement in the imagery task.

PAGE 24

24 Autonomic Mobilization. Increases in heart rate and skin conductance level were expected to be highest for the most arous ing categories irrespective of valence (i.e., attack/danger and high arousal pleasant). Affect Communication through Facial Action. Corrugator EMG change was expected to show the greatest tension to the most unpleas ant scenes (e.g., disgust) and relaxation below baseline to the most pleasant (e.g., nurturan ce/affiliation) scenes. Zygomatic EMG change was also anticipated to be largest for the mo st pleasant contents and to show moderate activity (in conjunction with orbi cularis EMG) to highly aversive contents such as disgust, reflecting grimacing. Orbicularis EMG was also expected to show increases for both pleasant and unpleasant contents, and fore most for higher arousing contents (e.g., attack/danger). Action Readiness: The Probe Startle Response. Responses to the acoustic startle probe as indexed in both blink magnitude and skin c onductance were expected to be greatest to probes delivered during imagery of the most ar ousing categories irrespective of valence. Sensorimotor Imagery and Subjective Im agery Ability: Exploratory Analyses Participants completed imagined simple sensory-motor experiences while the previously described physiological measures were r ecorded. In addition vividness ratings were collected for each imagined sensation. Depi cted sensory experiences corresponded to the seven modalities represented on the QMI: kinesthetic, interoceptive, visual, acoustic, olfactory, gustatory, and cutaneous. Each m odality was represented by prompts depicting low and moderate sensory intensity. Particip ants also completed the self-report QMI assessment of imagery vividness/ability. Physiological Effects of Prompt Intensity. Due to the stronger associated response elements, the moderate compared to low intens ity sensory-motor prompts were expected to elicit more robust physiological re activity. Specifically, as indi ces of intensity or arousal, larger responses were expected in blink ma gnitude to the startle probe, orbicularis EMG tension, skin conductance le vel, and heart rate. Evaluative and Physiological Effects of Prompt Modality. In the absence of data to guide hypotheses about relative po tency of images in specif ic sensory modalities (e.g., olfactory vs. visual), exploratory tests were performed to assess modality differences in both subjective clarity (i.e., experimental vividness rati ngs and QMI) and physiological reactivity. Association of Subjective Imagery Ab ility and Physiological Reactivity. o The hypothesis was tested that elevated physiological reactivity to more intense prompts might coincide with higher e xperimental ratings of vividness. o In light of prior inconsistent findings concerning the relation between QMI and objective measures of imagery, the te ntative hypothesis was proposed that individuals reporting high s ubjective clarity of their mental images on the QMI

PAGE 25

25 would more reliably demonstrate in creased physiological reactivity during moderate compared to low intensity imag ery of simple sensory experiences. Associations between Imagery Ability and Narrative Imagery: Exploratory Analyses In the present study, the same participant co mpleted assessments of both sensory-motor imagery ability and extent of emotional reactivity during narrative imagery. This withinsubjects design enabled direct comparison of physiological engagement in the two conditions. QMI ratings were additionally ex amined in relation to narrative imagery. Subjective Imagery Ability as a Predictor of Emotional Imagery. On the basis of previous studies (e.g., Levin, 1982; Miller et al., 1987), if subjective imagery vividness covaries with affective react ivity, it was expected to be most evident in visceral responding, particularly unpleasant co mpared to neutral conditions. Psychophysiological Imagery Ability as a Predictor of Emotional Imagery. o In light of the similarities in the re sponding modalities, imagery ability assessed physiologically as opposed to subjectively wa s expected to be more predictive of reactivity during narrative imagery. o If, in fact, sensory imagery predicts em otional imagery, individuals demonstrating greater differentiation between low a nd moderate sensory-motor imagery responses were expected during narrative imagery to show bot h greater affective differentiation among contents (i.e., more reliable valence effects) and larger magnitude responses.

PAGE 26

26 CHAPTER 2 EXPERIMENT 1: NARRATIVE IMAGERY Method Participants Fifty-seven students (30 women; 27 men) from the University of Florida introductory psychology class participated in tw o studies in exchange for cour se credit. Study 1 entailed the completion of an experimental psychophysiological assessment of emotional narrative imagery. Materials and Design Text depicting twenty-four emotional and neutra l experiences were devised (Table 2-1) to comprise three superordinate valence categorie s (pleasant, neutral, unpleasant) with seven different content categories. Th e content categories included: panic, contamination, social fear, attack/danger, neutral, social rein forcement, high arousal pleasant Each of the seven content categories included three different text exemplars, except for neutral, which included six exemplars for added reliability. The scenes were 20 words long and reflected action and participation in the ongoing scene as opposed to bystander observation. Sentences were written to reveal the affective tone within the first three to four words. Words explicitly denoting emotional states (e.g., fear, sadness, anger) or semantic interpretations of the scene (e.g., dangerous, unpredictable) were excluded. An additional two scenes were written to serve as stimuli for demonstration trials. The sentences were digitized into 10 second a udio files. A female w ith instructions to use minimal prosody recorded all sentences. Sentences were recorded in stereo at a sampling rate and size of 44.1 kHz and 16-bit. The audio script s, along with tones of uniform intensity, were played via Neurobehavioral Systems Presen tation program commands from a SoundBlaster AWE 64 Gold sound card, connected to a 12 Vo lt power supply and gated by a Coulbourn

PAGE 27

27 Audio-Mixer Amplifier (S82-24; Coulbourn Instrume nts, Allentown, PA). All auditory stimuli were presented to the participant over ma tched Telephonics TDH-49 headphones (Telephonics Corporation, Huntington, NY). Trials were presented in 8 orde rs across subjects so that no more than 2 stimuli of the same hedonic valence, and no 2 stimuli in the same content category were presented consecutively. Specific exemplars were pseudo-randomized into th e category sequence sti pulated by each order. A single trial consisted of an initial 3-s econd baseline, followed by the onset of a 12second auditory script describi ng an ongoing emotional or neutra l scene, followed by 12 seconds of imagery, and finally a tone-cued return to relaxation (Figure 2-1). After a brief recovery period a series of numbers were presented over the earphones and the partic ipant was instructed to press a button when a given target number was heard. Startle probes were presented 4 and 10 seconds into each imagery period and 3 sec onds into the intertrial interval on 50% of trials. The acoustic startle stimulus consisted of a 50-ms presentation, 95 dB(A) burst of white noise with instantaneous rise time generated by a Coulbourn S81-02 white noise generator and presented over matched Telephonics TDH-49 headphones. Procedure The participant was seated in a recliner in a sound-atttenuated, dimly lit room. Following completion of consent procedures sensor placem ent and introduction to the protocol commenced. The participant first read the upcoming imagery prompts and rated each in terms of pleasantness and arousal using the Self-Assessment Ma nikin (SAM; Lang, 1980; Bradley & Lang, 1994). At the completion of ratings, the participant was instructed to carefully attend to the acoustic scripts when presented. At stimulus of fset the subject was to vividly imagine being actively involved, as a participant as opposed to an observer, in the situation suggested by the

PAGE 28

28 script. The participant was told to maintain this active imagining until hearing the next tone, and then to relax. The participant was notified that a series of numbers would be read and to press the joystick if the target number (identified at this time) was heard. (The identified number was varied across participants. This number tracking task was intended to interfere with residual or ongoing imagining that might otherwise occur). Th e participant was instructed to keep eyes closed throughout the entire sess ion and that brief noises (star tles) heard over the headphones could simply be ignored. Following these instru ctions a demonstration program was run to present two practice trials. Next the experimenter reviewed with the participant the imagery task in regards to the demonstration trials, confirming comprehension of the instructions, and then began data collection. Following completion of the experiment the pa rticipant was subsequently debriefed, paid credit, and thanked. Physiological Response Measurement As noted above, stimulus presentation, timing, and ratings data acquisition was accomplished using a PC-compatible computer running Neurobehavioral Systems Presentation software (Neurobehavioral Systems, 2005) Acquisition of blink magnitude, facial electromyography, and autonomic measures was accomplished using a PC-compatible computer running VPM software (Version 11.2, Cook, 2000). Physiological signals were continuously sampled at 20 Hz for the duration of the experi ment. Integrated electr omyographic potentials were recorded from the corrugator (left brow), z ygomatic (left cheek), and orbicularis (left eye) regions with Sensormedics miniature electrode s (Sensormedics, Yorba Linda, CA), using the placement recommended by Fridlund and Cacio ppo (1986). The raw corrugator and zygomatic EMG signals were amplified by 10000, and freque ncies below 13 Hz and above 1000 Hz were

PAGE 29

29 filtered, using a Coulbourn S75-01 bioamplifier. Th e raw signals were rectified and integrated using a Coulbourn S76-23A contour following in tegrator, with a time constant of 500 ms. Skin conductance electrodes were placed adja cently on the hypothenar eminence of the left palmar surface using Sensormedics standard elect rodes filled with 0.05-m NaCl Unibase paste. The signal was acquired with a Coulbourn S71-23 sk in conductance coupler and calibrated prior to each session to detect activity in the range from 0 Siemens. The electrocardiogram was recorded from the left and right forearms, using large Sensormedics electrodes filled with electrolyte paste. The signal was filtered using a Coulbourn S75-04 bioamplifier, and a Schmitt trigger interr upted the computer each time the R component of the cardiac waveform was detected. Usi ng VPM software (Cook, 2000) interbeat intervals were recorded to the nearest millisecond. The eyeblink component of the startle res ponse was measured by recording EMG activity over the orbicularis oculi muscle of the left eye. For continuous measurem ent of the orbicularis EMG, the raw signal was amplified (5000), and fr equencies below 28 Hz and above 500 Hz were filtered, using a Coulbourn S75-48 bioamplifier. Th e raw signal was rectified and integrated using a Coulbourn S76-23A contour following integr ator, with an actual time constant of 20 ms. To index the eyeblink to the startle reflex, activity in the orbicularis oculi muscle was sampled at 20 Hz with the exception of an increase in sa mpling rate to 1000 Hz for 50 ms prior to and 250 ms following startle probe onset. EEG was recorded from 129 electrodes using an Electrodesics Inc. (EGI) high-density EEG system and digitized at a rate of 250 Hz us ing Cz as a recording reference. Impedances were maintained below 50 kOhm as recommen ded for the Electrical Geodesics high-input impedance amplifiers. A subset of EGI electrodes located at the outer canthi and below the right

PAGE 30

30 eye were used to determine horizontal and vert ical electrooculogram (EOG). All channels were processed on-line with 0.1 Hz high-pass and 100 Hz low-pass filtering. Data Reduction Using VPM software (Cook, 2000) corrugator, z ygomatic, and orbicularis oculi EMG, and skin conductance were reduced off-line into half-second bins. Following the suggestions of Graham (1978) interbeat intervals were reduced to rates and then to weighted averages representing heart rate in beats per minute in half-second bins. Reactions in each measure were determined by subtracting activity in the 1 s prio r to script presentation from that occurring at each half-second following onset. For all measures, response averages were calculated for the 12-second imagery periods. For skin conductan ce level log transforma tion (log[SCL+1]) was performed to normalize the data. Further, ra nge correction was performed to control for individual differences in SCL (Lykken, Rose, & Luther, 1966). Specifica lly, within-subject the average for each trial was expressed as a proportion of the participants total range of variation across trials (range corrected SCL = (trial SCL minimum SCL)/(maximum SCL minimum SCL)). The eye blink data was reduced off-line us ing a program (Balaban, Losito, Simons, & Graham, 1986) that scores the maximum excursi on in analog to digital units from the level immediately preceding response onset. Trials with cl ear artifacts were reject ed, while trials with no responses were scored as zero magnitude blinks. For all trials the blink magnitude to each of the two startle probes delivered during an imager y epoch were standardized within-subject in relation to the mean and standard deviation of the probe response s elicited during the inter-trial interval. The standardized values provide a m easure of each participants responsivity during imagery compared to a rest condition.

PAGE 31

31 For assessing skin conductan ce responses (SCRs) to the acoustic startle probe during narrative imagery the procedure outlined by Bradley, Silakowski, & Lang (2007) was implemented. The skin conductance level at each half-second following the probe presentation during imagery was deviated from the average of the two half-seconds prior to probe presentation. The peak change that occurred fr om one to five seconds following the probe was scored as the skin conductance response. Hence, change scores reflected increase or decrease relative to the pre-startle baseline. Brain Electrical Source Analysis was used for EEG data reduction (BESA; MEGIS Software, Inc., Grfelfing, Munich). After transf ormation to the average reference, data was digitally filtered at 0.1 to 30 Hz, corrected for ocular artifacts and averaged by condition. Reasonable sensor groupings for ANOVAs were determined by inspection of topographical maps of scalp voltage. Eleven centro-parietal se nsors were selected for analysis (Sensors: 53, 54, 61, 62, 67, 68, 73, 78, 79, 80, 87). To extract the startle probe P300, artifact-free segments in the EEG were averaged and the P300 component to the startle probe was id entified as the mean amplitude between 260 and 340 ms after probe onset. The P300 amplitude was measured relative to the mean of the 100 ms prior to probe onset. Data Analysis Analyses of each ERP, physiological, and refl ex measure were conducted to assess the effects of stimulus content on imagery. In genera l, repeated measures analyses of variance (ANOVA) were employed using content as within subjects factors. Statistical analyses were accomplished using the SPSS package (Version 11.0, 2002) First, valence (p leasant, neutral, unpleasant) was the repeated-subje ct variate and linear and quadratic trend components as well as follow-up planned comparison (e.g., unpleasant v. neutral) were used to delineate valence response patterns. To explore specific category effects, content categories were then each

PAGE 32

32 contrasted with reactivity during neutral imager y (e.g., neutral v. social reinforcement). Due to signal to noise ratio limitations, probe P300 data were not analyzed by content categories. For all analyses that involved repeated measures with more than two levels, the multivariate test statistic (Wilks' lambda) was employed to avoid potential sphericity issues (Vasey & Thayer, 1987). Results Participant Characteristics Fifty-seven students (30 women; 27 men) from the University of Florida introductory psychology class participated in two studies in exchange for c ourse credit. The self-reported racial status of the participants was as follows : white (70%), black (12%), Hispanic (9%), Asian (4%), multiracial (2%), and other (3%). The part icipants reported first languages as follows: English (85%), Spanish (5%), Korean (2%), Cr eole (2%), both English and Spanish (3%) and other (2%). The mean age of participants wa s 19.45 years (SD = 2.30). Ei ghty-nine percent of participants were right-handed, 4% left-h anded, 5% ambidextrous, and 2% unreported. Narrative Imagery: Evaluative Judgments Listed in Table 2-2 are the means and standard deviations for ratings of pleasantness and arousal for the imagery scripts averaged accord ing to superordinate valence categories. As expected, valence categories affected ratings of pleasantness, F (2, 55) = 594.97, p < .001, with the previously defined pleasant scripts evoking the highest pleasantness ratings followed by neutral and lastly unpleasant scripts, linear trend F (1, 56) = 1208.62, p < .001 (Figure 2-2A). Further as expected, compared to neutral script s, ratings of pleasure were greater for pleasant imagery, F (1, 56) = 265.96, p < .001, and less for unpleasant imagery, F (1, 56) = 663.85, p < .001.

PAGE 33

33 The a priori valence categories reliab ly affected ratings of arousal, F (2, 55) = 216.13, p < .001 (Figure 2-2B). Although not di fferent from one another, F (1, 56) = 0.12, ns, arousal ratings were greater for both pleasant, F (1, 56) = 437.08, p < .001, and unpleasant, F (1, 56) = 237.98, p < .001, compared to neutral imagery. To address evaluative differences within va lence category, the s ubordinate pleasant and unpleasant content categories were next analyzed w ithin-valence, as well as in relation to ratings for neutral contents. The ratings averaged by c ontent are presented in Table 2-3. Concerning the pleasant categories, both social reinforcement, F (1, 56) = 265.14 p < 001, and pleasant arousing, F (1, 56) = 186.58 p < .001, were rated more pleasant than neutral scripts, while not differing from one another, F (1, 56) = 0.12, ns Similarly in terms of arousal, both social reinforcement, F (1, 56) = 335.04, p < .001, and pleasant arousing, F (1, 56) = 416.64, p < .001, were rated more pleasant than neutral scripts. In contrast to commensurate levels of pleasantness, pleasant arousing contents were rated as more arousing than social reinforcement contents, F (1, 56) = 24.05, p < .001. The content categories in order of mean valence are illustrated in Figure 23A and in order of mean arousal in Figure 2-3B In considering the ratings of specific unpleasan t contents, contaminati on scenes were rated the least pleasant followed by attack/danger, panic, and lastly social threat, all of which were less pleasant than neutral scenes, all comparisons to neutral, p < .001. Contamination, attack, and panic scenes were rated as si milarly unpleasant, whereas social threat was rated the least unpleasant content, all comparisons to social threat, p < .001. All four unpleasant contents were rated as more arousing than neutral, all comparisons to neutral, p < .001. Attack/danger scripts were rated as the most arousing followed by pani c, contamination and social threat. Among the

PAGE 34

34 unpleasant categories all conditions excep t contamination and social threat, F (1, 56) = 0.19, ns were reliably different in rated arousal, all comparisons, p < .001. Affect communication th rough facial action Corrugator EMG activity. Figure 2-4A illustrates the half-second changes in corrugator EMG activity during the course of listen, imagery and recovery averaged by valence. Focusing on the average of the 12-second imagery period (Fi gure 2-4B) the valence of the scene affected corrugator activity, F (2, 49) = 10.39, p < .001, with greater tension el icited during imagining of unpleasant, compared to neutral, F (1, 50) = 5.60, p < .05, and pleasant scenes, F (1, 50) = 21.07, p < .001 (Table 2-2). In addition, pleasant imagery evoked a rela tive relaxation compared to neutral, F (1, 56) = 11.22, p < .01. Regarding pleasant contents, both social reinforcement, F (1, 56) = 8.07, p < .01, and pleasant arousing scripts, F (1, 56) = 6.37, p < .05, elicited a reduction in corrugator EMG activity compared to neutral imagery (Table 2-3; Figure 2-5). The two pleasa nt contents resulted in similar lower levels of corrugator activity, F (1, 56) = 1.67, ns Of the four unpleasant contents, the two rated most unpleasant, attack, F (1, 56) = 6.22, p < .05, and contamination, F (1, 56) = 6.68, p < .05, demonstrated increased corrugator activity compared to neutral. Posthoc analyses aimed at further discriminating among the unpleasant contents revealed that similarly, processing of attack, F (1, 56) = 3.72, p < .05, and contamination, F (1, 56) = 5.52, p < .05, imagery was marked by increased activity comp ared to the least unpleasant c ontent, social threat imagery. Zygomatic EMG activity. Figure 2-6A illustrates the ha lf-second changes in zygomatic EMG activity during the course of listen, imagery and recovery averaged by valence, reflecting that this measure is primarily sensitive to plea sant contents. Focusing on the average of the 12second imagery period (Figure 2-6B) no main effect of valence emerged, F (2, 51) = 1.79, ns (Table 2-2 ). However, planned follow-up tests revealed a significant increase during pleasant

PAGE 35

35 compared to unpleasant imagery, F (1, 52) = 3.64, p < .05*, and a trend for increased activity compared to neutral imagery, trend F (1, 52) = 2.39, p = .06*. Finer analyses revealed that social reinforcement elicited gr eater zygomatic activity than all other conditions, all comparisons, p < .05, and, furthermore, was the only condition to differ from neutral, F (1, 52) = 3.64, p < .05. Orbicularis EMG activity. Figure 2-7A illustrates the half-second changes in orbicularis EMG activity during the course of listen, imagery and recovery averaged by valence. A main effect of valence did not emerge during the imagery period, F (2, 47) = 2.04, ns, although a quadratic trend was evident F (1, 48) = 4.18, p < .05, (Figure 2-7B) Planned comparisons demonstrated that both pleasant, F (1, 48) = 3.32, p < .05*, and unpleasant, F (1, 49) = 3.26, p < .05*, imagery evoked increased orbicularis tension compared to neutral imagery. Regarding the specific content differences from neutral, imagery of contamination, F (1, 48) = 3.92, p < .05*, as well as social reinforcement, F (1, 48) = 3.18, p < .05*, scenes elicited significantly more or bicularis activity, F (1, 48) = 3.09, p < .05* (Figure 2-8). A trend also emerged for elevated activity during pleasant arousing imagery, F (1, 48) = 2.46, p = .06*. Autonomic mobilization Heart rate. Figure 2-9A illustrates the waveforms by valence obtained from averaging heart rate changes over half-second periods during listening, imagery, and recovery. A pronounced increase in heart rate wass evident during listening to the scripts, followed by a gradual reduction in h eart rate during the subsequent imagery and recovery periods. During imagery, stimulus valence affected heart rate, F (2, 49) = 8.68, p < .05, in that unpleasant, F (1, One-tailed test

PAGE 36

36 50) = 17.64, p < .01, and pleasant, F (1, 50) = 4.0, p < .01*, processing elicited more sustained heart rate change than the ne utral condition (Figure 2-9B). Content specific analyses revealed that of the pleasant contents social reinforcement, F (1, 50) = 5.92, p < .05, resulted in more sustained heart rate increase compared to neutral. All of the unpleasant contents showed increa sed heart rate change compar ed to neutral (social threat F (1, 50) = 11.34, p < .01; panic, F (1, 50) = 12.40, p < .01; contamination, F (1, 50) = 13.17, p < .05; attack; F (1, 50) = 2.94, p < .05*). Skin conductance level. During imagery, stimulus valence did not affect change in skin conductance level, F (2, 50) = 1.17, ns, and in follow-up pairwise comparisons, neither emotional content differed from neutral. Specific emotional contents similarly failed to evoke a reliable increase in skin conductance compared to neutral imagery. Post hoc analyses revealed that only 30 of the 52 (58%) participants ev inced more reactivity to emoti onal than neutral contents. For this subset of the partic ipants, valence influenced skin conductance level, F (2,28) = 6.82, p < .01, with pleasant, F (1, 29) = 8.45, p < .01, and unpleasant, F (1, 29) = 5.53, p < .05, contents eliciting more prolonged sympathetic activity than neutral imagery. Furthermore, among these sympathetic responders, contamination, F (1, 29) = 5.59, p < .05, panic, F (1, 29) = 3.59, p < .05*, and social reinforcement, F (1, 29) = 12.74, p < .01, imagery resulted in more sustained skin conductance responding than neutral imagery. Action readiness: The probe startle response Startle reflex. The blink magnitude in response to th e startle probe differed as a function of imagery valence, main effect, F (2,47) = 5.91, p < .01, with reliably augmented reflexes resulting during unpleasant comp ared to neutral imagining, F (1, 48) = 9.05, p < .01 (Figure 2* One-tailed test

PAGE 37

37 10A). A trend emerged for elevated responding du ring pleasant compared to neutral imagining, F (1, 48) = 1.97, p = 08*. Probe times (i.e., 4 seconds or 10 seconds) did not affect response magnitude (main effect of time, ns ) nor did valence modulati on differ by probe position (valence x time effect, ns ). Content specific analyses reveal ed that in comparison to neutral imagining attack, F (1, 48) = 4.16, p < .0 5 panic, F (1, 48) = 4.87, p < .0 5 and contamination, F (1, 48) = 8.96, p < .0 5 scenes resulted in startle potenti ation (Figure 2-10 B). A trend also emerged for pleasant arousing imagery to prompt larger probe responses than neutral imagery, F (1, 48) = 2.20, p = .07*. Skin conductance response. Figure 2-11 illustrates the mean responses in skin conductance secondary to the star tle probes averaged by valence. A main effect of valence emerged, F (2, 48) = 3.40, p < .05, primarily owing to a grea ter skin conductance response subsequent to the probes delivered during unpleasant compared to neutral imagery, F (1, 49) = 6.77, p < .05. Probe times (i.e., 4 seconds or 10 seconds) did not affect response magnitude (main effect of time, ns ) nor did valence modulation differ by probe position (valence x time effect, ns ). Thus the autonomic reaction to the probe wa s similar to the blink response, but varied in content emphasis. Content specific analyses revealed that contamination, F (1, 49) = 3.69, p < .05*, and social threat, F (1, 49) = 4.79, p < .05, imagining resulted in larg er post-probe skin conductance responses than neutral imagining. Attentional engagement: P300 to the probe The event-related averages at Cz and Pz by valence are shown in Figure 2-12 illustrating the effect of imagery engagement on later probe processing, F (2, 90) = 8.22, p < .001, with both One-tailed test

PAGE 38

38 pleasant, F (1, 45) = 11.49, p < .01, and unpleasant contents, F (1, 45) = 9.53, p < .01, compared to neutral evoking less positive deflections 260 to 340 ms after the startle probe. This pattern is the inverse of that recorded fo r the blink response indicating redu ced attentional al location to the probe during emotional processing. Discussion: Narrative Imagery Affect Communication through Facial Action Facial expressivity measures, corrugator and zygomatic EMG showed the expected covariation with valence, in that corrugator EMG demonstrated the largest increases for the most subjectively unpleasant (i.e., contamination) and th e greatest relaxation for the most subjectively pleasant (i.e., social reinforcement) imager y. Zygomatic EMG was elevated for pleasant imagery, and more specifically those scenes deno ting positive social inte raction (i.e., social reinforcement). Although the winning and victor y scenes depicted in the pleasant arousing category were rated commensurate to social reinfo rcement scenes in terms of pleasantness, the scenes describing direct positive reinforcement from superiors and friends elicited more facial motor activity, likely reflecting subovert smiling behavior. Orbicularis EMG activity showed commensur ate increases during both pleasant and unpleasant compared to neutral imagery, consistent with previous imager y studies. Interestingly, however content analyses revealed that orbicula ris activity covaried more closely with valence than arousal. In particular, the most pl easant (social reinforcement) and unpleasant (contamination) contents elicited the most increase similar to content specificity demonstrated during emotional picture viewi ng (Bradley, Codispoti, Cuthbe rt, & Lang, 2001). The coincident increases in zygomatic and orbi cularis EMG for highly pleasant materials, as observed here to social reinforcement scripts is typically cons idered a measure of the authentic or genuine Duchenne smile (Ekman, Davidson, & Friesen, 1990), in contrast to the unfelt, social smile

PAGE 39

39 recruiting only zygomaticus major (Bradley et al., 2001). Further, the marked increase in orbicularis during contamination imagery possi bly reflects grimacing (Bradley et al., 2001). Taken together the three facial action measur es showed strong correspondence to social communication behavior implied in the imagery scenes. Autonomic Mobilization Consistent with prior studies (e.g., Witvliet & Vrana, 2000) and as expected, pleasant and unpleasant narrative imagery compared to neutral prompted increased auto nomic reactivity as captured in heart rate. All unpleasan t scenes showed increases relati ve to neutral whereas social reinforcement, but not pleasant arousing script s showed increases. Somewhat unexpectedly, heart rate increase was not as closely coupled to rated arousal as dem onstrated in previous studies (e.g., Bradley et al., 1995). Rather, contamination, showed the greatest heart rate increase among unpleasant contents and social rein forcement among pleasant contents. Contrary to expectations changes in skin conductance level did not systematically vary with emotion. However, although these two output systems often show similar sensitivity to emotional arousal, particularly during discrete, ex ternal stimulation such as affective picture presentation (e.g., Bradley et al., 2001), the weak effects in electrodermal responding seen here are not surprising and have been demonstrated du ring prior investigations of narrative imagery (Bauer & Craighead, 1979; Carro ll, Marzillier, & Merian, 1982; Lang et al., 1980; Lang et al., 1983; Vrana, 1993, 1994; Witvliet & Vrana 1995). In re ference to these intermittent null effects Lang et al. (1983) previously conj ectured that the purposeful disatt ention to external stimulation requisite in mental imagery, in part functions to attenuate sk in conductance r eactivity by tuning out the modality to which it is most sensitive. He cautioned that to consistently detect changes during imagery samples would need to be of su fficient size and/or vi sual media should be employed to prompt subsequent imaging. Intere stingly, in the current study the phasic skin

PAGE 40

40 conductance responses to the acoustic probes showed evidence of reliable modulation by emotion indicating that theme-rele vant processing is in fact c oded in the electrodermal system and that abrupt external stimul ation provides a useful window thr ough which to assess activation. Action Readiness: The Probe Startle Response In addition to the reliable modulation of the skin conductance response, blink magnitude to the startle probe differed as a function of vale nce with reliably augmented reflexes resulting during unpleasant compared to neutral imagini ng. Although not an overwhelming effect, the data were also suggestive of augmented startle re sponding during pleasant im agery. In particular, probes delivered during attack, pani c, and contamination imagery pr ompted larger startles than neutral imagery. Further, the pa ttern of modulation was suggestiv e of increased blink magnitude during highly arousing pleasant imagery. Importantly, in both blink magnitude and skin conductance response to the startle stimulus probe position (middle or late in the imagery pe riod) did not influence overall magnitude or the pattern of affective modulation, suggesting su stained motivational pr ocessing throughout the entire imagery period. Attentional Engagement: P300 to the Probe A primary aim of this study was to assess wh ethersimilar to pict ure viewing paradigms in which probe P300 amplitude was reduced du ring more emotionally arousing pictures (e.g, Schupp et al., 1997)more emotionally arousing narrative imagery would attenuate the P300 response. Findings revealed that, in fact, aff ective modulation of the P300 was present during emotional imagery with both pleasant and unpleasa nt contents compared to neutral evoking less positive deflections after the startle probe. Reduced P300 responses to probes delivered amidst competing non-affective cognitive tasks have been consistent across studi es and interpreted as reflecting attentional resource a llocation to the foreground task (P olich, 2007). In the context of

PAGE 41

41 emotional picture viewing, diminished probe P 300 responding has been conceptualized as an index of natural selective attention or processes activated to s ubserve motivational mobilization (Lang et al., 1998; Schupp et al., 199 7). In the current investigati on, attentional processes have been preferentially engaged in the generation of emotionally arousing as compared to neutral imagining. These findings are particularly compel ling in that a task requiring effortful mental elaboration grabs attention simila r to prominent external stimuli resulting in less resources for processing the probe interruption. Conclusions In conclusion these findings were largely c onsistent with hypothe ses concerning patterns of expected modulation in the different response systems. However, one findingthe prominence of contamination in the defensive pr ofilewas interesting in that even measures typically more reflective of emo tional arousal as opposed to vale nce, showed the most reactivity to this category (e.g., SCR, blink magnitude, h eart rate, orbicularis EMG). Contamination was rated the most unpleasant content, and alt hough more arousing than neutral, it was rated significantly less arousing than pa nic and attack/danger imagery. Th e consistency of preferential processing of contamination as the most unpleasa nt, but moderately arou sing category prompts the question of whether future investigations might productively implement a combined measure of valence and arousal, an aff ective vector of sorts that would enable the simultaneous examination of these dimensions in determining physiological response. In total, these findings sugge st that probe P300 responding in conjunction with autonomic, somatic, and reflexive recording provide inform ative, multi-modal assessment of attentional allocation, expressive dispositions, and somato motor readiness during narrative imagery.

PAGE 42

42Table 2-1. Text of scripts for psychophysiological examination of affectiv e narrative imagery Valence category Content category Exemplar Sentence Unpleasant Panic Exemplar 1 Panic comes out of the blue. No warning. Your hearts racing. You cant get your breath. Thoughts whirl: Im going crazy? Am I going to die? Exemplar 2 People are all around you, pressing closer. Its hard to breathe. Youre flushed, sweaty, dizzyconfused. You realize its another att ack and this time, you think, I will die. Exemplar 3 Youve waited endlessly at the check-out counter. Trapped. Others crowd against you. Theres a sudden rushing in your head. You gasp for breath, chest tight, temples throbbing. Is it a heart attack? Contamination Exemplar 1 You are leaving the concert. (whe n) A drunk, smelling of smoke and alcohol, stumbles into you and throws up on your jacket. You retch as vomit drips onto your hand. Exemplar 2 You bite hungrily into the hamburger, and abruptly catch the putrid smell of spoiled meat. You spit out, and a greasy piece falls. Exemplar 3 You gag, seeing a roach moving slowly over the surface of the pizza. You knock the pie on the floor. Warm cheese spatters on your shoes. Social fear Exemplar 1 Its your turn to speak to the group. Theyre all looking at you. Your mouths dry and you cant get the words out. Your heart pounds in the silent room. Someone laughs. Exemplar 2 Everyones talking, laughing together at the party. Youre alonetense, sweaty. People glance at you and quickly look away. When asked your name: Throat dry, you croak an answer. Exemplar 3 Everyones staring at you, waiting for your presentation. Youve misplaced all your notes, graphicseverythings lost! What will you say? They see you shaking, sweatingmumbling stupidly. Attack/danger Exemplar 1 Youre alone in the alley in a bad part of the city. A street gang slowly surrounds you, knives out, laughing with menace. Your heart pounds as they close in. Exemplar 2 Without thinking, you stepped off the curb into traffic. Brakes screech. You look up, frozen, heart jumping in your chest. A truck is skidding, hurtling towards you. Exemplar 3 Its late at night in a poorly lit parkin g lot. You tense, clutching the keys. Your car stands alone in the distance, when footsteps sound behind you.

PAGE 43

43Table 2-1. Continued Valence category Content category Exemplar Sentence Neutral Neutral Exemplar 1 You are sitting at the kitchen table with yesterdays newspaper in front of you. You push back the chair when you hear the coffee maker slow to a stop. Exemplar 2 Its good to be able to do nothing and just stretch out on the couch. The television is on with the sound off. You can hear the low rumble of traffic in the distance. Exemplar 3 You run the comb through your hair, strai ghten your collar, smooth out the shirts wrinkles. Water is running in the sink. You turn it off and leave. Exemplar 4 You unfold the map, spread it out on the table, and with your finger trace a route south towards the beach. You refold the map, pick up your bag, and leave. Exemplar 5 Its a quiet day without much to do. Youre sitting around your place, resting, reading, and looking out the windowwhere leaves swirl gently in the wind. Exemplar 6 You are relaxing on a lawn chair, looking out into the garden. A childs tricycle is abandoned on the grass. You hear the low buzz of a lawn mower in the distance. Social reinforcement Exemplar 1 Youve successfully completed a difficult assignment. Your friends are enormously pleased for youwith pats on the back and welcome praise. Your face is warm. You cant help smiling. Exemplar 2 She really likes your gift. As soon she saw it, she screamed with joy: Thank you. Its just perfect, fantastic. Your heart beats with pleasure, when she leaps up and hugs you. Exemplar 3 The boss smiles and shakes your hand. Youll receive a very big raise in pay. Good work! he says. Your heart skips a beat. Some one shouts congratulations. You smile back. High arousal pleasant Exemplar 1 The registered letter says that You have just won ten million dollars! Its amazingYou bought the winning ticket in the lottery. You cry, scream, jump with joy! Exemplar 2 Its the last few minutes of the big game and its close. The crowd explodes in a deafening roar. You jump up, cheering. Your team has come from behind to win. Exemplar 3 You won a free pass to the whole carniva l. Like kids again, you all jump on the merry-goround, laughing as it turns, singing along with the music: What a wonderful day!

PAGE 44

44 Table 2-2. Narrative imagery: Mean response (standard de viation) across measures av eraged by valence category Pleasant Neutral Unpleasant Dependent Measure N M SD M SD M SD Valence main effect Affect communication Corrugator EMG (V) 51 -0.46 (1.25) 0.03 (0.96) 0.35 (1.48) F (2, 49) = 10.39, p < .001a,b,c Zygomaticus EMG (V) 53 0.12 (0.62) -0.03 (0.42) -0.05 (0.18) F (2, 51) = 1.78, nsa,b Orbicularis EMG (V) 50 0.63 (1.41) 0.28 (1.01) 0.55 (0.79) F (2, 47) = 2.04, nsc Autonomic mobilization Heart rate (bpm) 50 -0.70 (4.46) -2.02 (3.41) 0.15 (2.91) F (2, 49) = 8.68, p < .01a,c SCL (log (S + 1)) 52 -0.03 (0.08) -0.03 (0.06) -0.03 (0.05) F (2, 50) = 0.77, ns Action readiness Startle (T-score) 49 57.23 (27.05) 52.46 (6.38) 55.97 (11.01) F (2, 47) = 5.91, p < .01c SCR (S) 51 0.05 (0.08) 0.03 (0.08) 0.05 (0.12) F (2, 49) = 3.40, p < .05c Attentional engagement P300 (V) 48 4.24 (2.64) 5.51 (2.90) 4.63 (2.82) F (2, 90) = 8.22, p < .001a,c Evaluative ratings Pleasure (1-9) 57 8.20 (0.73) 6.18 (0.79) 2.21 (0.72) F (2, 55) =594.97, p < .001a,b,c Arousal (1-9) 57 7.07 (1.07 3.40 (1.24) 7.14 (1.10) F (2, 55) =216.13, p < .001a,c Note. = change; EMG = electromyographic; SCL = skin conductan ce level; SCR = skin conductanc e response; bpm = beats per minute; V = microvolt; S = miscrosiemen ; a = Comparison of pleasant versus ne utral conditions is significant at p < .05; b = Comparison of pleasant versus unpleasa nt conditions is significant at p < .05; c = Comparison of unpleasant versus neutral condition is significant at p < .05.

PAGE 45

45Table 2-3. Narrative imagery: Mean res ponse (standard deviation) across meas ures averaged by content category Pleasant Social Neutral Social Panic Attack Contamination arousing reinforcement threat M SD M SD M SD M SD M SD M SD M SD Affect communication Corrugator EMG (V) -0.31 (1.66) -0.61 (2.02) 0.04 (1.18) 0.04 (1.66) 0.28 (2.52) 0.44 (1.53) 0.67 (2.53) Zygomatic EMG (V) -0.01 (0.67) -0.24 (0.85) -0.03 (0.42) -0.01 (0.42) -0.07 (0.40) -0.03 (0.32) -0.07 (0.39) Orbicularis EMG (V) 0.64 (1.74) 0.62 (1.29) 0.28 (1.01) 0.37 (1.21) 0.48 (0.91) 0.45 (1.45) 0.89 (2.07) Autonomic mobilization Heart Rate (bpm) -1.21 (5.86) -0.20 (4.81) -2.02 (3.41) 0.39 (4.80) 0.52 (4.02) -0.96 (3.85) 0.66 (5.04) SCL (log (S + 1)) -0.05 (0.13) -0.02 (0.08) -0.03 (0.06) -0.02 (0.06) -0.02 (0.08) -0.05 (0.10) -0.03 (0.07) Action readiness Startle (T-score) 55.84 (36.09) 58.71 (14.26) 52.46 (6.38) 54.15 (9.67) 55.44 (18.76)* 55.35 (18.49) 58.86 (10.62) SCR (S) 0.05 (0.17) 0.04 (0.12) 0.03 (0.08) 0.06 (0.15) 0.06 (0.15) 0.04 (0.13) 0.05 (0.12) Evaluative ratings Pleasure (1-9) 8.22 (0.80)* 8.19 (0.63)* 6.18 (1.24) 2.67 (0.91)* 2.18 (0.62) 2.04 (0.77) 1.95 (0.89) Arousal (1-9) 7.42 (1.53)* 6.73 (1.12) 3.40 (1.61) 6.62 (1.24) 7.33 (0.76) 7.95 (0.82) 6.64 (1.04) Note. = change; EMG = electromyographic; SCL = skin conductance leve l; SCR = skin conductance response; bpm = beats per minute; V = microvolt; S = miscrosiemen; = Comparison versus neutral condition is significant at p < .05.

PAGE 46

46 Figure 2-1. Trial structure for na rrative imagery. A single trial consisted of an initial 3-second baseline, followed by the onset of a 12-s econd auditory script describing an ongoing emotional or neutral scene, followed by 12 seconds of imagery, and finally a tonecued return to relaxation. After a brief recovery period, a series of numbers were presented over the earphones and the partic ipant was instructed to press a button when a given target number was heard. Startle probes were presented 4 and 10 seconds into each imagery period and 3 seconds into the intertrial interval on 50% of trials.

PAGE 47

47 Figure 2-2. Mean ratings of pleasure and arous al as prompted by pleasant, neutral, and unpleasant scripts measured with the Se lf-Assessment Manikin. A) Pleasantness ratings. B) Arousal ratings. Figure 2-3. Mean ratings of pleasure and arous al by narrative imagery content category as measured with the Self-Assessment Maniki n. A) Pleasantness ratings. B) Arousal ratings.

PAGE 48

48 Figure 2-4. Average change in corrugator EMG activity by valence of imagined prompt. A) Smoothed, half-second averages during liste n, imagine, and recover. B) 12-second averages during imagine. Figure 2-5. Average 12-second change in corrug ator EMG activity during imagery by category of imagined prompt. Categories on the x-axis are arranged according to rated pleasantness.

PAGE 49

49 Figure 2-6. Average change in zygomatic EMG activity by valence of imagined prompt. A) Smoothed, half-second averages during liste n, imagine, and recover. B) 12-second averages during imagine.

PAGE 50

50 Figure 2-7. Average change in orbicularis EMG activity by valence of imagined prompt. A) Smoothed, half-second averages during liste n, imagine, and recover. B) 12-second averages during imagine.

PAGE 51

51 Figure 2-8. Average change in orbicularis EM G activity during neutra l scenes and those emotional scenes that during narrativ e imagery evoked significant increases compared to neutral. A). Smoothed, half-s econd averages of orbicularis EMG activity during listen, imagine, and recover prompt ed by neutral, pleasant arousing, social reinforcement, and contamination imager y. B) 12-second averages during imagine.

PAGE 52

52 Figure 2-9. Average change in heart rate by valence of imagined prompt. A) Half-second averages during listen, imagine, and recove r. B) 12-second averages during imagine.

PAGE 53

53 Figure 2-10. Mean blink magnitude to the startl e probe during narrative im agery. A) By valence category. B) By content category arranged on the x-axis in order of increasing mean response magnitude.

PAGE 54

54 Figure 2-11. Mean peak skin c onductance response to the startle probe during narrative imagery by valence of imagined prompt.

PAGE 55

55 Figure 2-12. Event-related averages to the startle probe at electr odes Cz and Pz 100 ms preto 400 ms post-probe by valence of imagined prompt. Figure 2-13. Scalp potential at 11 centro-parietal sensors 260-340 ms after the startle probe by valence of imagined prompt.

PAGE 56

56 CHAPTER 3 EXPERIMENT 2: SENSORIMOTOR IMAGERY Method Participants Fifty-seven students (30 women; 27 men) from the University of Florida introductory psychology class participated in tw o studies in exchange for course credit. At a preceding session the same participants in the narrative imag ery protocol completed both an experimental psychophysiological assessment and self-report measure of imager y vividness/ability Materials and Design Text depicting fourteen common sensory experiences were devised (Table 3-1) to parallel the items of the QMI (Betts, 1909; Sheehan, 1967). The QMI ( Cronbachs alpha = 0.77) is a 35item self-report measure of indivi dual differences in imagery ability. Items relate to imaginal vividness of stimuli in seven sensory modaliti es (visual, auditory, olfactory, cutaneous, kinesthetic, gustatory, and in teroceptive/organic). Instructions and sample items include Think of [perceiving] the fo llowing, considering carefully th e picture which comes before your minds eye. Classify the image suggested by the following question as indicated by the degrees of clearness and vi vidness specified on the Rating Scale (1 = As vivid as the actual experience; 9 = No image at all. You are only thinking about it). Sample rating items include: The sun as it is sinking below the horizon; The whistle of a locomotive; Sand; Running upsta irs; Salt. Scores are summed to obtain a total score, such that a low total score implies hi gh self-reported imag ery ability. For the psychophysiological test of imagery vividness tw o text prompts (one low and one moderate sensory-motor intensity) were generated to correspond to each of the seven sensory-motor modalities represented in the QMI, resulting in seven low and seven moderate intensity prompts. Text prompts of 1.9 to 4 s in duration were digitized in the same procedure outlined in Study 1. Script exemplars were presented such th at low and moderate sensory-motor intensity

PAGE 57

57 trials alternated. Two orders were presented to en sure that low and moderate sensory-motor trials were equally often presented at each position during the protocol, counterbalanced across participants. For example, one order would pr oceed for 14 trials: LowModerateLowModerate and so forth. The alternative order would be ModerateLowModerateLow and so forth. The specific exemplars were pseudo-randomly assigned to the two category positions. A single trial consisted of an initial 3-sec ond baseline, followed by the onset of a brief auditory script describing a common daily activ ity with sensory and motor prompts, followed by 12 seconds of imagery, and finally a tone-cued return to relaxation a nd a rating of imagery vividness. Startle probes were presented 8 seconds into each imagery period and 4 seconds into the intertrial interval on 50% of trials (Figure 3-1). Participants received a total of 23 startle probes. Stimulus presentation, timing, and physiol ogical and ratings data acquisition were accomplished with the same parameters and equipment outlined for Study 1. Procedure The participant sat in a recliner in a soundatttenuated, dimly lit room. Following consent procedures sensors were placed on the participan t. The participant then received instructions regarding the parameters for im agining the imagery prompts: This next test is designed to assess your mental imagery, how vividly you can call back to mind memories of common experiences. For example: Close your eyes, and relax. Now consider a red apple in your minds eye. Compos e a vivid image of the apple, its color, the shine of the skin. Is there a stem on top? No w imagine that you pick up the apple, feeling its heft, round shape and smooth skin, as you hold the apple in your hand. Now bite into the apple, taste the ripe fruit, smell its freshness. (Pause). Ill bet you had a good image in your minds ey e? How vivid was it? as you think of vivid memory images that you have had in the past. What rating would you give your apple image on a scale from 1 to 9with 1 mean ing no image at all, and 9 meaning it was perfectly vivid and clear, the same as if you were experiencing a real apple?

PAGE 58

58 Now, continue relaxing with your eyes closed. You are going to hear a series of different common eventslike eating an applefor you to imagine in the same way, as vivid mental experiences. Think about each event, and hold your image and reactions in your mind for several seconds. When you hear the tone, rate th e image for its clarity and vividnessremember, 1 to 9. Rate how much it was as if you were actually seeing, feeling, or doing something. The participant was further instructed to keep his/her eyes closed during imagery and that brief noises heard over the headphon es could simply be ignored. Following the completion of the imagery protocol, the participant completed the Questionnaire Upon Mental Imager y (QMI) and then was debriefe d, granted credit, and thanked. Data Reduction Data reduction procedures were similar for the two studies with the exception that due to a programming error the first eight rather than 12 seconds of conti nuous data were recorded during sensorimotor imagery. Consequently averages fo r imagery periods during the sensorimotor task represent the half-second averages over eigh t seconds. Additionally, due to the previously described programming error, data is unavail able for SCRs during sensorimotor imagery. Data Analysis Analyses of each ERP, physiological, and refl ex measure were conducted to assess the effects of stimulus content on imagery. In genera l, repeated measures analyses of variance (ANOVA) were employed using content as within subjects factors. Statistical analyses were accomplished using the SPSS package (Version 11.0, 2002) Influences of intensity and sensory modality were entered as within-subjects repeated measures (2 (Low, Moderate) x 7 (Gustatory, Olfactory, Cutaneous, Interoceptive, Acoustic, Vi sual, Kinesthetic)) to assess main effects and interactions In separate analyses, using mixed-model rep eated measures ANOVA individuals identified as good and poor subjective imagers on th e QMI measure of imagery vividness were

PAGE 59

59 compared in terms of emotional reactivity duri ng narrative imagery. The same analysis was then repeated for individuals iden tified as good and poor imag ers defined by magnitude and consistency of physiological differentiation betw een low and moderate intensity sensorimotor imagery. For all analyses that involved repeated measures with more than two levels, the multivariate test statistic (Wilks' lambda) was employed to avoid potential sphericity issues (Vasey & Thayer, 1987). Results Participant Characteristics Same as Study 1. Physiological Effects of Prompt Intensity and Sensory Modality Increased physiological reactivity was reli ably demonstrated across modalities during moderate compared to low intensity prompt s as indexed in orbicularis EMG activity, F (1, 55) = 10.31, p < .01, corrugator EMG, F (1, 55) = 3.87, p < .05*, skin conductance level, F (1, 55) = 9.05, p < .01 and blink magnitude to the startle probe, F (1,55) = 4.22, p < .05 (see Figure 3-2; Table 3-2). No stimulus intensity di fferences emerged in measures of z ygomatic EMG activity, F (1, 55) = 1.54, ns, heart rate change, F (1, 55) = 0.50, ns, or electrocortical probe responses as indexed by P300, F (1, 47) = 2.63, ns Concerning differences among sensory modalities an additional main effect emerged in orbicularis EMG activity, F (6,51) = 2.36, p < .05, with the largest in crease during imagery of acoustic prompts followed by visual, olfact ory, cutaneous, kinesthetic, gustatory and interoceptive imagery (Figure 3-3). Posthoc anal yses revealed that im agery of interoceptive One-tailed test

PAGE 60

60 experience, resulted in less orbicularis activit y than all other sensori-motor imagining, all comparisons, p < .05, although still yielded a reli able increase from baseline, F (1, 56) = 5.49, p < .05. Corrugator EMG activity was also affected by sensory modality, F (6,51) = 5.00, p < .001. As illustrated in Figure 3-4, imagery of cutaneous experience prompted the most increase followed by olfactory, visual, and acoustic imag ery. Taken together, these contents elicited significantly increased activity compared to baseline, F (1, 56) = 11.10, p < .05. Relaxation below baseline was evident during gustatory, ki nesthetic, and interoceptive imagining, F (1, 56) = 3.81, p < .05*. An interaction between intensity and se nsory modality emerged only in terms of orbicularis EMG activity, F (6,50) = 4.33, p < .001. Follow-up analyses demonstrated that olfactory, F (1, 55) = 3.11, p < .05*, acoustic, F (1, 55) = 10.60, p < .01, and gustatory. F (1, 55) = 13.49, p < .01, imagining prompted reliable increases during moderate compared to low intensity prompts, whereas no such difference emerged for cutaneous, kinesthe tic, and interoceptive imagery. Subjective Report of Imagery Vividne ss During Sensorimotor Imagery Contrary to expectations, rate d vividness of sensorimotor imagery was unrelated to the intensity of the prompts, F (1, 55) = 0.64, ns Sensory modality reliably affected ratings, F (6, 50) = 17.59, p < .001. As represented in Figure 3-5, imagery of interoceptive expe rience was rated as most vivid followed by kinestheti c, visual, acoustic, gustatory, ol factory and finally cutaneous scenes. The relationship between vividness and intensity varied within sense ( interaction F (6, 50) = 9.20, p < .001) with visual and acoustic prompts yielding higher vividness ratings while the One-tailed test

PAGE 61

61 opposite pattern emerged for olfactory and kinest hetic prompts. Cutaneous, interoceptive and gustatory high and low intensity prompts resulted in commensurate levels of imaginal clarity. Subjective Report of Imagery Vividness: QMI Scores The sum of ratings of imagery vividness during the sensorimot or task was correlated with imagery vividness and clarity as in dexed in the QMI Total score ( r = 0.52, p < .001). Sensory modality similarly affected rated vividness, F (6, 49) = 12.81, p < .001, with interoceptive experiences again rated as most vivid followe d by kinesthetic, visual gustatory, cutaneous, acoustic and finally olfactory scenes (Figure 36). Taken together, the experimental vividness and QMI ratings suggest that interoceptive and kinesthetic experiences are imagined with the most subjective vividness wher eas olfactory and cutaneous expe riences are conjured with less clarity. Good and Poor Imagers: QMI Score as a Predictor of Sensorimotor Imagery Did individuals reporting high s ubjective clarity of their mental images on the QMI show more pronounced physiological differences between moderate and low intensity sensorimotor imagery? Participants were defined as good a nd poor imagers based on subjective reports of imagery vividness on the QMI. Specifically, indivi duals in the lower quartile of QMI vividness scores (i.e., higher vividness) were defined as good imagers (n = 13; QMI Mean score = 56.15; SD = 19.77) and those in the upper quartile as poor imagers (n = 13; QMI Mean score = 117.77; SD = 19.77). The upper and lower quartiles we re selected to maximize detection of potential individual differences and to approximate group means found in previous studies that selected on imagery ability (e.g., Arabian & Furedy, 1983; Miller et al., 1987) Next, a series of 2 x 2 repeated measures ANOVAs were performed assessing good versus poor imager status as the between subjects factor and sensory intensity as the within subjects factor. No significant main effects or interactions emerged in the followi ng physiological measures: orbicularis, zygomatic,

PAGE 62

62 and corrugator EMG, SCL, heart rate, and probe responses in blink magnitude and P300 eventrelated potentials. The only reliable difference emerged in terms of experimental vividness ratings in that, not surprisingl y, those individuals defined as good imagers on the QMI rated the imagined sensorimotor experiences as more vivid, F (1, 24) = 28.17, p < .001. Good and Poor Imagers: QMI Score as a Predictor of Narrative Imagery Next imagery vividness, again defined subjectiv ely was related to pa tterns of affective responding during narrative imagery. Fi rst, the analytical scheme a pplied to predict sensorimotor imagery (i.e., upper and lower quartiles on the QMI as criteria for poor and good imagers respectively) was implemented to predict physiological responding dur ing narrative imagery. Again the upper and lower quartile s were selected to maximize de tection of potential individual differences and facilitate compar ability to prior studies (e.g., Mill er et al., 1987). A series of 2 (imager status) x 3 (valence) repeated meas ures ANOVAs were performed on each of the physiological dependent measures. No significant main effects or interactions emerged in the following physiological measures: orbicularis, z ygomatic, and corrugator EMG, SCL, heart rate, and probe responses in blink magnitude and P300 event-related potentials. Good and Poor Imagers: Psychophysiologi cal QMI as a Predictor of Narrative Imagery Next imagery vividness was operationalized as more robust discrimination between low and moderate sensorimotor imagery. Specifically, for those measures predicted to discriminate intensity (i.e., SCL, heart rate, orbicularis EMG, and blink magnit ude to the startle probe), the difference scores between low and moderate im agery were subjected to unrotated principal components analysis to yield a linear composite. Pa rticipants were next divided according to the upper and lower quartile on this vector, with t hose individuals showing the highest loadings purported to be good physiological imagers and those with the lowest loadings as poor

PAGE 63

63 physiological imagers. Imager status was then re lated to patterns of affective responding during narrative imagery. A series of 2 (sensorimotor imager status) x 3 (valence) repeated measures ANOVAs were performed on each of the physiological dependent measures. Consistent and Inconsistent Imagers: Consistency of Reac tivity During Sensorimotor Imagery as a Predictor of Narrative Imagery In light of the failure to pr edict narrative responding from magnitude of reactivity during sensorimotor imagery, posthoc analyses we re next performed to assess whether consistency of physiological differentiation between low and mode rate intensity sensorimotor imagery across measures might better predict emotional re activity. Specifically individuals who showed increased reactivity in four of th e five measures predicted to di fferentiate intensity (i.e., blink magnitude, orbicularis EMG, corrugator EMG, heart rate, and skin conductance) were categorized as consistent responders across me asures (n = 15; 47% female). In contrast, individuals showing increased reac tivity during moderate compared to low reactivity in two or less physiological systems were cat egorized as inconsistent res ponders (n = 15; 53% female). A series of 2 (consistency stat us) x 3 (valence) rep eated measures ANOVAs were performed on each of the narrative imagery phys iological dependent measures. Table 3-3 lists the main effects and intera ctions for each of the dependent measures. Consistent responders were expected to demons trate more reliable affective modulation than inconsistent responders, resulting in a signi ficant valence by group interaction. However, inspection of Table 3-3 reveals that only blink magnitude to the startle probe was suggestive of such a relationship. This trend was followed up by analysis of the separate groups, which demonstrated that consistent responders showed reliable affective modulation (valence main effect, F (2, 11) = 4.59, p < .05) whereas inconsistent res ponders showed no differentiation

PAGE 64

64 among contents (Figure 3-7). A main effect of group also resulted from larger blink magnitude responses during unpleasant contents by consistent responders. Discussion: Sensorimotor Imagery Increased somatovisceral and reflex activity was demonstrated during moderate compared to low intensity sensorimotor prompts as indexed in orbicularis EMG, SCL, and blink magnitude to the startle probe. Modest but reliable incr ease was also evidenced in corrugator EMG. The finding that imagery of simple, context-free, sensor y percepts results in discernible activation in efferent physiology and reflexes th at covaries with experiential intensity suggests that more intense sensations may be coded in memory with stronger associated somatomotor response elements, which although not prompted explicitly by imagery scripts, are reliably activated during mental imagery. Interestingly, effects of sensory modality em erged only in corrugator and orbicularis EMG. Overall the two measures showed little similarity in terms of sensitivit y to specific imagined modalities. Although, for both measures interoceptiv e experience resulted in the least reactivity, and in the case of corrugator activity, the most relaxation. Perhaps the relative restriction of expressive facial activity reflected the primarily internally-oriented proc ess of interoception. In contrast the largest response in orbicularis EMG was evinced during acoustic imagery while corrugator tension was the larg est during cutaneous imagery, bot h more externally-oriented experiences. Most investigations of sensory im agery focus on a single modality, however, these preliminary results suggest that comparison of m odalities might further elucidate mechanisms of imagined experience. The reliable physiological differentiation of imagined sensation demonstrated here complements neuroimaging findings on imagined perception showing brai n activation patterns akin to in vivo experience. For example, Kosslyn and colleagues (e.g., Kosslyn et al., 2001;

PAGE 65

65 Kosslyn & Thompson, 2003; Kosslyn, Thompson, Kim, Rausch, & Alpert, 1995) have found that mental recreations or depictions of simple visual percepts activate ne ural structures similar to those recruited during percep tion, including striate and extrastr iate cortex, and more recently demonstrating even retinotopically organize d activation (Slotnick, Thompson & Kosslyn, 2005). This pattern of sense-specific cortical involvement during mental imagery has emerged for other modalities as well. Concerning kinesthetics, im agery of simple volitional motor movements activates somatotopically organi zed contralateral primary, premotor and supplementary motor areas (e.g., Binkofski et al., 2000; Ehrsson, Geyer, & Naito, 2003). Imagined gustation results in increased blood flow to both primary and second ary gustatory areas (Kobayashi et al., 2004; Kikuchi, Kubota, Nisijima, Washiya, & Kato, 2005) while imagined heari ng reliably activates secondary auditory cortex (e.g., Halpern, Zatore e, Bouffard, & Johnson, 2004; Kraemer, Macrae, Green, & Kelly, 2005; Zatorre & Halpern, 2005). Th ese demonstrated neur al correlates of sensation imagery suggest processes that mimic actual experience and, perh aps, result in the efferent outflow exhibited in the current study. These conclusions are, however, necessarily tentative as little wo rk exists in the efferent and re flex physiology of imagining sensory experiences such as taste, smell, touch, a nd interoception. Furthermor e, compared to the previously mentioned neuroimaging studies, the cu es employed in the present investigation were much more degraded. For example, in typical se nsory imagery studies desc riptions of the to-beimagined percepts are often prompted repeat edly during imagery (e.g., Kobayashi et al., 2004), described with greater specificity at the outset (Slotnick et al., 2005), pre-exposed (Kraemer et al., 2005), exposed as detailed depictions in a separate modality (Kikuchi et al., 2005) and/or practiced (e.g., Ehrsson et al., 2003). Although the differences in the presentation of imagery prompts necessarily limit the comparability of the current to former findings, they

PAGE 66

66 simultaneously suggest that the effects demonstr ated here would very likely be more robust under typical presentatio n modes. Essentially, these resu lts underscore the utility of physiological recording of imagery in a ssessing unobservable mental phenomena. Taken together, these findings supp ort efferent outflow as an e ffect of mental conjuring of simple sensation, which in turn covaries with the intensity of the sensory and motor experience denoted by promptspresumably owing to stronger associated response elements.

PAGE 67

67Table 3-1. Text of scripts for psychophysiologi cal experimental test of imagery ability Sensory/motor system Low intensity Moderate intensity Taste Tasting mashed potatoes. Biting into a raw lemon. Smell Smelling grass that has just been mowed. Smelling meat thats spoiled. Sight Looking at a blue enam eled bowl setting on a red table. Looking into the sun, squinting to see something clearly. Sound Hearing the sound of traffic in the distan ce. Hearing someones nails scrape across a blackboard. Cutaneous Smoothing wrinkles in a wool sweater with your hand. A pin sticking deep into your finger. Interoception/organic Relaxing in an easy ch air. Your heart pounding as you run upstairs. Kinesthetic Lying in bed just before falling asl eep. Pushing hard against a door thats stuck.

PAGE 68

68Table 3-2. Sensorimotor imagery: Mean response (stand ard deviation) across measur es averaged by intensity Low Intensity Moderate Intensity Dependent Measure N M SD M SD Intensity main effect Sustained responses Corrugator EMG (V) 56 0.06 (0.80) 0.30 (0.89) F (1, 55) = 3.87, p < .05 Zygomaticus EMG (V) 56 0.01 (0.22) -0.07 (0.47) F (1, 55) = 1.54, ns Orbicularis EMG (V) 56 0.98 (1.45) 1.54 (1.74) F (1, 55) = 10.31, p < .01 Heart Rate (bpm) 56 -0.33 (2.51) -0.64 (2.86) F (1, 55) = 0.50, ns SCL (log (S + 1)) 56 -0.01 (0.02) 0.00 (0.02) F (1, 55) = 2.80, p = .05 Probe responses Startle (T-score) 46 50.56 (9.13) 53.88 (8.78) F (1, 55) = 4.22, p < .05 P300 (V) 47 2.93 (2.26) 3.49 (2.48) F (1, 46) = 2.63, ns Evaluative rating Vividness (1-9) 56 6.64 (0.91) 6.53 (0.96) F (1, 55) = 0.64, ns Note. = change; EMG = electromyographic; SCL = skin conductan ce level; SCR = skin conductanc e response; bpm = beats per minute; V = microvolt; S = miscrosiemen.

PAGE 69

69Table 3-3. Effects of sensorimotor responder status and valen ce category on physiology and evalua tive ratings of narrative imag ery Dependent measure N Valence main effect Group main effect Valence x group interaction Affective communication Corrugator EMG 27 F (2, 26) = 7.16, p < .01 F (1, 27) = 9.93, p < .01 a F (2, 26) = 3.90, ns Zygomaticus EMG 30 F (2, 27) = 2.19, ns F (1, 28) = 1.01, ns F (2, 27) = 1.99, ns Orbicularis EMG 27 F (2, 24) = 1.35, ns F (1, 25) = 0.04, ns F (2, 26) = 2.13, ns Autonomic mobilization Heart Rate 27 F (2, 25) = 11.35, p < .001 F (1, 26) = 0.21, ns F (2, 25) = 1.34, ns SCL 28 F (2, 26) = 0.16, ns F (2, 26) = 1.92, ns F (1, 27) = 0.46, ns Action readiness Startle 27 F (2, 24) = 4.56, p < .01 F (1, 25) = 2.91, p = .05 a F (2, 24) = 3.08, p = .06 SCR 26 F (2, 24) = 2.49, ns F (1, 25) = 1.07, ns F (2, 24) = 0.96, ns Attentional engagement P300 26 F (2, 24) = 1.04, ns F (1, 25) = 0.50, ns F (2, 24) = 0.65, ns Evaluative ratings Pleasure (1-9) 27 F (2, 27) = 271.26, p < .001 F (1, 28) = 0.09, ns F (2, 27) = 0.89, ns Arousal (1-9) 27 F (2, 27) = 107.45, p < .001 F (1, 28) = 0.02, p = .05 a F (2, 27) = 0.32, ns Note. = change; EMG = electromyographic; SCL = skin conductan ce level; SCR = skin conductanc e response; bpm = beats per minute; a = Consistent responders demonstrated significantly larger responses than in consistent responders at p < .05; b = Inconsistent responders demonstrated significantly larger responses than consistent responders at p < .05.

PAGE 70

70 Figure 3-1. Trial structure for sensorimotor imag ery. A single trial consisted of an initial 3second baseline, followed by the onset of a br ief auditory script describing a common daily activity with sensory-motor prompts, followed by 12 seconds of imagery, and finally a tone-cued return to relaxation and a rating of imagery vividness. Startle probes were presented 810 seconds into each imagery period and 4 seconds into the intertrial interval on 50% of trials.

PAGE 71

71 Figure 3-2. Average physiological change during low and moderate intensity sensorimotor imagery. A) Measured in orbicularis EMG (volt = microvolt change). B) Measured in corugator EMG (volt = microvolt change). C) Measured in SCL (siemen = microsiemen change). D) Indexed in blink magnitude to the startle probe.

PAGE 72

72 Figure 3-3. Average orbicularis EMG change dur ing during sensorimotor imagery by sensory modality. Sensory modality means on the x-ax is are ordered according to response magnitude. Figure 3-4. Average corrugator EMG change dur ing during sensorimotor imagery by sensory modality. Sensory modality means on the x-ax is are ordered according to response magnitude.

PAGE 73

73 Figure 3-5. Average experimental ratings of im agery vividness by sens ory modality. Increasing value indicates more subjective vividness. Sensory modality means on the x-axis are ordered according to increasing vividness. Figure 3-6. Imagery vividness as rated on the Questionnaire Upon Mental Imagery (QMI) averaged by sensory modality. Illustrated means are reverse-scored from the original values. Increasing value indicates more subj ective vividness. Sensory modality means on the x-axis are ordered according to increasing vividness.

PAGE 74

74 Figure 3-7. Mean blink magnitude to the startl e probe during narrative im agery for individuals classified as inconsistent (n = 15) and consistent (n = 15) physiological responders during sensorimotor imagery.

PAGE 75

75 CHAPTER 4 GENERAL DISCUSSION Imagery Ability and Reactivity to Sens ory and Emotional Imagery: Meaningful Associations? Imagery vividness as rated during both th e experimental sensorimotor task and Questionnaire Upon Mental Imagery (Betts, 1909; Sheehan, 1967) were positively correlated and in both sets of responses vividness varied by sensory modality. Across measures interoceptive and kinesthetic experiences were imagined with the most subject ive vividness whereas olfactory and cutaneous experiences were conjured with less clarity. In contrast to the high concordance between subjective measures of vividness (i.e., QMI and experimental ratings), neither measure predicted the exte nt of physiological reactivity during sensorimotor or emotional imagery. Similarly, physiologically vivid responders defined as large magnitude responders during sensorimotor imagery did not show more robust affective modulation during narrative imagery. Further, identification of indi viduals who showed consistent as opposed to inc onsistent differentiation between low and moderate intensity sensorimotor imagery across physiological measur es only slightly improved predictive power. Specifically, inconsistent responders failed to show valence-modulated startle responding in comparison to intact modulation among consis tent responders. However, no other group differences emerged and the gene ralizability of this is finding is clearly limited. Overall subjective imagery ability, reactivity during sensorimotor imagery, and reactivity during emotional imagery were not closely related in the current sample. These results do not directly address the intended aim of provid ing an alternative psychophysiological Questionnaire Upon Mental Imagery that w ould more precisely predict the physiology of emotional imagery. However, as noted at the outset, these findings are not wholly unexpected and lead to compelling que stions. If subjective imagery vividness, as

PAGE 76

76 demonstrated here is consistent across self-repo rt measures (i.e., QMI a nd experimental ratings) but does not predict physiologi cal reactivity duri ng imagery, how are respondents defining clarity? Should subjective sensory imagery be expected to predict engagement during more elaborate, multiply-determined images of em otional experience? Lang and colleagues (1980) emphasized that response propositions in emo tional images are not limited to perceptual information and, in fact, the data here sugge st that responders during percept imagery are not the same individuals showing robust affec tive modulation during narrative imagery. Imagery Ability, Psychopathology, and Emotional Reactivity While posing more questions th an answering, these results in directly promote confidence that the observed differences in patterns of emotional reactivity duri ng imagery among fearful and anxious patients are at tributable to characteristics central to the respective pathologies other than group differences in imagery ability. In a programmatic series of studies, Lang and colleagues have found that although nearly all anxiety patients repor t comparable fearfulness and symptoms of anxious arousal, not all anxiet y diagnoses show an accompanying physiology of defense (Lang, 1985). While defensive reactions ar e strong in specific p hobics when imagining encounters with fearful scenes, this has not always been true for other diagnoses, notably agoraphobia, panic disorder, and generalized anxiety disorder. The postulate that these differences might be due to imagery ability was raised secondary to Lang and colleagues (Cook et al., 1988) findings that heart rate increases to fearful imag ery were positively related to imagery ability among specific phobics but not for social phobics or agoraphobics. However, in more recent findings with patients the group differe nces have replicated in that those with circumscribed fearfulness show more pronounc ed defensive reactivity compared to the attenuated and unreliable responding of those w ith more diffuse and chronic apprehension and arousal (Cuthbert et al., 2003; Lang, McTeague, & Cuthbert, 2007; McNeil, Vrana, Melamed,

PAGE 77

77 Cuthbert, & Lang, 1993). Whereas the main effect of fearfulness versus a nxiety has persisted, the interaction with imager y ability has not (Cuthbert et al ., 2003; Lang, McTeague, & Cuthbert, 2007). Startle Reflex Modulation in Sensorimotor Imagery: Implications for Understanding Startle Reflex Facilitation during Narrative Imagery Although not an intended aim in the current investigation, the with in-subject data on blink magnitude to the startle pr obe and concurrent probe P300 dur ing imagery of simple sensory experiences as well as more elaborate narra tives inadverently addressed the ongoing debate concerning the processes underl ying facilitated startle refl exes during both pleasant and unpleasant imagery. Concerning atte ntional hypotheses, Panayiot ou & Vrana (1998) assessed the effects of externally versus in ternally-focused attention by exam ining startle reflex magnitude and heart rate changes during an initial period of acoustically presented digits (perception) and a subsequent rehearsal period (i nternal elaboration) and f ound both potentiated startle and accelerated heart rate during rehearsal compared to listen. Furthermore, both startle magnitude and heart rate acceleration increa sed with the length of digit span to rehearse suggesting influence of cognitive effort or load. The authors concluded th at these results, in conjunction with results on affectiv e startle modulation during imagery, i ndicate that when attention is not engaged to a specific sensory modality, increasi ng the effortful demand on processing resources increases the startle response. Miller and colleagues (2002) offered an alte rnative explanation that the parameters and associated processing demands of imagery versus picture processing yiel d the different startle patterns. Picture viewing is a passive, percep tual task requiring unimodal orienting to the external sensory environment. In contrast, imag ery is a cognitive task requiring purposeful elaboration and disengagement fr om the external sensory envir onment. During picture viewing,

PAGE 78

78 the acoustic startle probe is cr oss-modal and fewer resources are available for processing the stimulus, resulting in reflex attenuation. Convers ely, attention is focu sed internally during imagery and the startle probe functions as an interrupt thereby poten tiating responding. The authors further speculated that star tle is potentiated proportionally to the extent of engagement in imagery and the degree of proce ssing interruption and that this process is independent of valence-modulated startle. Bradley and colleagues (1995) further refined the comparisons of internal and external processing influences on the acoustic startle reflex by comparing reactivity during narrative imagery and imagery of an immediately pre ceding picture. Narrative imagery evoked the expected pattern of commensurate elevations in startle reactivity dur ing both pleasant and unpleasant contents compared to neutral. In co ntrast, imagery of visual percepts yielded increased startle during unpleasant compared to pleasant imagining, similar to the pattern evoked during viewing of the actual picture. These resu lts suggest that facil itated acoustic startle responding during emotional versus non-emotional imagery is not primarily attributable to interruption of internally-orien ted processing exaggerated by cro ss-modal stimulation. Rather, Bradley et al. (1995) suggested that similar to th e interpretation of autonomic and somatic output during imagery and further consistent with th e bioinformational model (Lang, 1979), patterns of startle facilitation during imagery of arousing narrative scenarios ma y be an aspect of the motor and response dispositions engaged during the re instatement of persona l experience and action. In the current study of simple sensory imag ery, startle probes elicited increased blink magnitude during moderate compared to low intensity imagery and no accompanying difference in P300 amplitude. In contrast, narrative imag ery prompted increased blink magnitude during emotional (most reliably unpleasant) processing co mpared to neutral, and coincident reductions

PAGE 79

79 in probe P300 responding. The disso ciation in these patterns bears on the postulates of Miller et al. (2002) that startle is potent iated as a function of the extent to which internally-directed attention is interrupted by th e intruding, external st artle stimulus. In th e present narrative imagery study the decreased probe P300 that resu lted during emotional sc enes (suggesting more attention to emotional imagery) and coincided w ith increased startle re flex responses are not inconsistent with the premise of Miller and colle agues (2002). However, the lack of differences in probe P300 responses during sensorimotor im agery (suggesting no differences in attentional engagement), in conjunction with the augmented startle during imagination of more intense sensory and motor experiences implicates, rath er, the recruitment of stronger somatomotor response elements. Although more research is needed to disentangle this i ssue, these preliminary data suggest that startle facilitation during imagery cannot fully be explained by degree of attentional engagement. Furthermore these results provide tentative suppor t for the hypothesis set forth by Bradley and colleagues (1995) that startle facilitation dur ing arousing narrative scenarios reflects readiness for action as part of broader motivational mobilization.

PAGE 80

80 LIST OF REFERENCES Amrhein, C., Muhlberger, A., Pauli, P., & Wied emann G. (2004). Modulation of event-related brain potentials during affective picture proc essing: a complement to startle reflex and skin conductance response? International Journal of Psychophysiology, 54 231. Arabian, J. M., & Furedy, J. F. (1983). Individual differences in imagery ability and Pavlovian heart rate decelerative conditioning. Psychophysiology, 20 325. Balaban, M., Losito, B., Simons, R. F., & Graham, F. K. (1986). Off-line latency and amplitude scoring of the human reflex eye blink with Fortran IV. Psychophysiology, 23 612. Bauer, R. M., & Craighead, W. E. (1979). Psych ophysiological responses to the imagination of fearful and neutral situations: Th e effects of imagery instructions. Behavior Therapy, 10, 389. Betts, G.H. (1909). The distribution and functi ons of mental imagery (Contributions to Education, No. 26). New York: Columb ia University, Teachers College. Binkofski. F., Amunts, K., Stephan, K. M., Posse S., Schormann, T., Freund, H. J., Zilles, K., & Seitz. R. J. (2000). Broca's region subs erves imagery of motion: A combined cytoarchitectonic and fMRI study. Human Brain Mapping, 11 273. Bradley, M. M. (2000). Emotion and motivation. In J. T. Cacioppo, L. G. Tassinary, & G. Bernston (Eds.), Handbook of psychophysiology (pp. 602). New York: Cambridge University Press. Bradley, M. M., Codispoti, M., Cuthbert, B. N., & Lang, P. J. (2001). Emotion and motivation I: Defensive and appetitive reac tions in picture processing. Emotion, 1 276. Bradley, M. M., Cuthbert, B. N., & Lang, P. J. (1995). Imagine that! Startle in action and perception. Psychophysiology 32 (Suppl. 1 ), s21. Bradley, M. M., & Lang, P. J. (1994). Measuring emotion: The self-assessment manikin and the semantic differential. Journal of Behavioral Therapy and Experimental Psychiatry, 25 49. Bradley, M. M., Silakowski, T. D., & Lang, P. J. (2007). Fear of pain and defensive activation Manuscript submitted for publication. Brownley, K. A., Hurwitz, B. E., & Schneider man, N. (2000). Cardiovascular psychophysiology. In J. T. Cacioppo, L. G. Tassinary, & G. Bernston (Eds.), Handbook of psychophysiology (pp. 224). New York: Cambridge University Press. Carroll, D., Marzillier, J. S., & Merian, S. (1982). Psychophysiological changes accompanying different types of arousing and relaxing imagery. Psychophysiology, 19, 75.

PAGE 81

81 Cook, E. W., III, Melamed, B. G., Cuthbert, B. N., McNeil, D. W., & Lang, P. J. (1988). Emotional imagery and the diffe rential diagnosis of anxiety. Journal of Consulting and Clinical Psychology, 56, 734. Cook, E.W., III, Hawk, L.W., Davis, T. L., & Stevenson, V. E (1991). Affective individual differences and startle reflex modulation Journal of Abnormal Psychology 100 5. Craske, M. G., Antony, M. M., & Barlow, D. H. (2006). Mastering your fears and phobias: Therapist guide (2nd ed.). New York, NY: Oxfo rd University Press. Cuthbert, B. N., Lang, P. J., Stra uss, C., Drobes, D., Patrick, C. J., & Bradley, M. M. (2003). The psychophysiology of anxiety disorder: Fear memory imagery. Psychophysiology, 40 407. Cuthbert, B. N., Schupp, H. T., Bradley, M. M., McManis, M. H., & Lang, P. J. (1998). Probing affective pictures: Attended startle and tone probes. Psychophysiology 35 344. Davis, M. (2000). The role of the amygdala in conditioned and unconditioned fear and anxiety. In J. P. Aggleton (Ed.), The Amygdala, Volume 2 (pp. 213). Oxford, United Kingdom: Oxford University Press. Davis, M., & Lang, P. J. (2001). Em otion: Integration of animal and human data and theory. In M. Gallagher & R. J. Nelson (Eds.), Comprehensive Handbook of Psychology, Volume 3: Biological Psychology (pp. 405). New York: Wiley. Dawson, M. E., Schell, A. M., & Filion, D. L. (2000). The electrodermal system. In J.T. Cacioppo, L. G. Tassinary, & G. Bernston (Eds.), Handbook of psychophysiology (pp. 200). New York: Cambridge University Press. Dickinson, A., & Dearing, M. F. (1979). Appeti tive-aversive interactions and inhibitory processes. In A. Dickinson & R. A. Boakes (Eds.), Mechanisms of learning and motivation (pp. 203). Hillsdale, NJ: Erlbaum. Donchin, E., & Coles, M. G. H. (1988). Is the P300 component a manifestation of cognitive updating? The Behavioral and Brain Sciences, 11 357. Ehrsson, H. H., Geyer, S., & Naito, E. (2003). Im agery of voluntary moveme nt of fingers, toes, and tongue activates corresponding body-pa rt-specific motor representations. Journal of Neurophysiology, 90 3304. Ekman, P., Davidson, R. J., & Friesen, W. V. (1990). The Duchenne smile: Emotional expression and brain physiology II. Journal of Personality & Social Psychology, 58, 342.

PAGE 82

82 Fabiani, M., Gratton, G., & Coles M. (2000). Even t-related brain potentials In J. T. Cacioppo, L. G. Tassinary, & G. Berntson (Eds.), Handbook of Psychophysiology (pp. 602). New York: Cambridge University Press. Foa, E., Hembree, E., & Rothbaum. B. (2007). Prolonged exposure therapy for PTSD: Emotional processing of traumatic experiences: Therapist guide New York, NY: Oxford University Press. Fridlund, A. J., & Cacioppo, J. T. (1986). Guid elines for human electromyographic research. Psychophysiology, 23 567. Frijda, N. H. (1986). The Emotions. New York: Cambridge. Graham, F. K. (1978). Constraint s on measuring heart rate and period sequentially through real and cardiac time. Psychophysiology, 15 492. Greenwald, M. K., Bradley, M. M ., Cuthbert, B. N., & Lang, P. J. (1998). Startle potentiation: Shock sensitization, aversive learni ng, and affective picture modulation. Behavioral Neuroscience, 112 1069. Halpern, A. R. Zatorre, R. J., Bouffard, M., & Johnson, J. A. (2004). Be havioral and neural correlates of perceived and imagined musical timbre. Neuropsychologia, 42 1281. Hamm, A. O., Greenwald, M. K., Bradley, M. M., & Lang, P. J. (1993). Emotional learning, hedonic change, and the startle probe. Journal of Abnormal Psychology 102 453. Handy, T. C. (2005). Basic principles of ERP quantification. In T. C. Handy (Ed.), Event-related potentials: A methods handbook (pp. 3). Cambridge, MA: MIT Press. Jacobson, E. (1931). Electrical measurements of neuromuscular states during mental activities: Variation of specific muscles contracting during imagination. American Journal of Physiology, 96, 115. Keil. A., Bradley, M. M., Junghofer, M., Russmann, T., Lowenthal, W., & Lang, P. J. (2007). Cross-modal attention capture by affective stimuli: Evidence from event-related potentials. Cognitive, Affective, & Be havioral Neuroscience, 7, 18. Kikuchi, S., Kubota, F., Nisijima, K., Washiya, S., & Kato, S. (2005) Cerebral activation focusing on strong tasting food: A func tional magnetic resonance imaging study. Neuroreport, 16 281. Kobayashi, M., Takeda, M., Hattori, N., Fukunaga M., Sasabe, T., Inoue, N., et al. (2004). Functional imaging of gustatory percepti on and imagery: "Top-down" processing of gustatory signals. Neuroimage, 23 1271.

PAGE 83

83 Konorski, J. (1967). Integrative activity of the brain: An in terdisciplinary approach. Chicago: University of Chicago Press. Kosslyn, S. M., Ganis, G., & Thompson, W. L. (2001). Neural foundations of imagery. Nature Reviews Neuroscience, 2 635. Kosslyn, S. M., Thompson, W. L., Kim, I. J., Rausch, S. L., & Alpert, N.M (1995). Topographical representations of mental images in primary visual cortex. Nature, 378, 496. Kosslyn, S.M., & Thompson, W.L. (2003). When is ear ly visual cortex activated during visual mental imagery? Psychological Bulletin, 129, 723. Kraemer, D. J. M., Macrae, C. N., Green, A. E., & Kelley, W. M. (2005). Sound of silence activates auditory cortex. Nature, 434 158. Lang, P. J. (1977). Imagery in therapy: An information processing analysis of fear. Behavior Therapy, 8 862. Lang, P. J. (1979). A bio-informational theory of emotional imagery. Psychophysiology, 16 495. Lang, P. J. (1980). Behavioral tr eatment and bio-behavioral asse ssment: Computer applications. In J. B. Sidowski, J. H. Johnson, & T. A. Williams (Eds.), Technology in mental health care delivery systems (pp. 119-137). Norwood, NJ: Ablex Publishing. Lang, P. J. (1985). The cognitive psychophysiology of emotion: Fear and anxi ety. In A. H. Tuma & J. D. Maser (Eds.), Anxiety and the anxiety disorders (pp. 131). Hillsdale, NJ: Erlbaum. Lang, P. J., Bradley, M. M., & Cuthbert, B. N. (1990). Emotion, attention, and the startle reflex Psychological Review, 97, 377. Lang, P. J., Bradley, M. M., & Cuthbert, B. N. (1997). Motivated attention: Aff ect, activation and action. In P. J. Lang, R. F. Simons, & M. T. Balaban (Eds.), Attention and orienting: Sensory and motivational processes. Hillsdale, NJ: Lawrence Erlbaum Associates, Inc. Lang, P. J., Cuthbert, B. N., & Bradley, M. M. (1998). Measuring emotion in therapy: Imagery, activation, and feeling. Behavior Therapy, 29 655. Lang, P. J., Greenwald, M. K., Bradley, M. M ., & Hamm, A.O. (1993). Looking at pictures: Affective, facial, visceral and behavioral reactions. Psychophysiology, 30 261. Lang, P. J., Kozak, M. J., Miller, G. A., Levi n, D. N., & McLean, A., Jr. (1980). Emotional imagery: Conceptual structure and pattern of somato-visceral response. Psychophysiology, 17, 179.

PAGE 84

84 Lang, P. J., Levin, D. N., Miller, G. A., & Kozak, M. J. (1983). Fear imagery and the psychophysiology of emotion: The problem of affective response integration. Journal of Abnormal Psychology, 92, 276. Lang, P. J., McTeague, L. M., & Cuthbert, B. N. (2007). Fear, anxiety, depression, and the anxiety disorder spectrum: A psychophysiolo gical analysis. In T. Treat & T. Baker (Eds.), Psychological clinical science: Recen t advances in theory and practice. Integrative perspectives in honor of Richard M. McFall (pp. 167) Mahwah, NJ: Lawrence Erlbaum Associates. Lang, P. J., Melamed, B. G., & Hart, J. ( 1970). A psychophysiological analysis of fear modification using an automa ted desensitization procedure. Journal of Abnormal Psychology, 76 220. Larsen J. T., Norris, C. J., & Cacioppo, J. T. (2003). Effects of positive and negative affect on electromyographic activity over zygomatic us major and corrugator supercilii. Psychophysiology, 40 776. Levin, D. N. (1978). Imagery training and the psychophysiology of fear imagery in focal phobia and social anxiety Unpublished masters thesis University of Wisconsin. Levin, D. N. (1982). The psychophysiology of fear reduction: Role of response activation during emotional imagery. Dissertation Abstracts International, 43 (10), 3367-B. Lykken, D. T., Rose, R., & Luther, B. (1966). Correcting psychophysiological measures for individual differences in range. Psychological Bulletin, 66 481. Marks, D. F.. (1972). Individual differences in th e vividness of visual imag ery and their effect on function. In. P. W. Sheehan (Ed.), The function and nature of imagery New York: Academic Press. McNeil, D. W., Vrana, S. R., Melamed, B. G., Cuthbert, B. N., & Lang, P. J. (1993). Emotional imagery in simple and social phobia: Fear versus anxiety. Journal of Abnormal Psychology, 102, 212. McTeague, L. M., Bradley, M. M., & Lang, P. J. (2002). Creating a mental image: Is a picture worth a thousand words? Psychophysiology, 39 (Suppl. 1), S57 McTeague, L. M., Dimoulas, E., Strauss, C. C., Bradley, M. M. & Lang, P. J. (2003 ). Fear, anxiety, and physiology: Patte rns of emotional modulation. Psychophysiology, 40 (Suppl. 1), S62. McTeague, L. M., Shumen, J. R., Laplante, M. C., Bradley, M. M., Cuthbert, B. N., & Lang, P. J. (2006). Attacking animals, leering audien ces & the psychophysiol ogy of social phobia Psychophysiology, 43 (Suppl. 1), S66.

PAGE 85

85 Miller, G. A., Levin, D. N., Kozak, M. J., Cook, E. W., III, McLean, A., Jr., & Lang, P. J. (1987). Individual differences in imag ery and the psychophysiology of emotion. Cognition and Emotion, 1, 367. Miller, M. W., Patrick, C. J., & Levenston, G. K. (2002). Affective imagery and the startle response: Probing mechanisms of modul ation during pleasant scenes, personal experiences, and discrete negative emotions. Psychophysiology, 39, 519. Neuper, C., & Pfurtscheller, G. (2001). Event related dynamic of cortic al rhythms: Frequencyspecific features and functional correlates. International Journal of Psychophysiology, 43 41. Panayiotou, G., & Vrana, S. R. (1998). Effect of self-focused attention on the startle relex, heart rate, and memory performance among soci ally anxious and no nanxious individuals. Psychophysiology, 35 328. Patrick, C. J.; Cuthbert, B. N., & Lang, P. J. ( 1994). Emotion in the criminal psychopath: Fear image processing. Journal of Abnormal Psychology, 103, 523 Polich, J. (2007). Updating P300: An integrative theory of P3a and P3b. Clinical Neurophysiology Schupp, H. T., Cuthbert, B. N., Bradley, M. M., Birbaumer, & Lang, P. J. (1997). Probe P3 and blinks: Two measures of affective startle modulation. Psychophysiology 34 1 6. Sheehan, P. W. (1967). A shortened form of Betts' questionnaire upon mental imagery. Journal of Clinical Psychology, 223, 380. Slotnick, S. D., Thompson, W. L., & Kosslyn, S. M. (2005). Visual mental imagery induces retinotopically organized activa tion of early visual areas. Cerebral Cortex, 15 1570. Tassinary, L. G., & Cacioppo, J. T. (2000). The skeletomotor system: Surface electromyography. In J. T. Cacioppo, L. G. Tassinary, & G. Bernston (Eds.), Handbook of psychophysiology (pp. 163). New York: Cambridge University Press. Van Diest, I., Winters, W., Devriese, S., Vercam st, E., Han, J.N., Van de Woestijne, K.P., et al. (2001). Hyperventilation beyond fight/flight: Respiratory responses during emotional imagery. Psychophysiology, 38 961. Vasey, M. W., & Thayer, J. F. (1987). The continuing problem of false positives in repeated measures ANOVA in psychophysio logy: A multivariate solution. Psychophysiology, 24, 479. Vrana, S. R., (1993). The psychophysiology of disgust: Differentiating negative emotional contexts with facial EMG. Psychophysiology, 30 279.

PAGE 86

86 Vrana, S. R., (1994). Startle refl ex response during sensory modality specific disgust, anger, and neutral imagery. Journal of Psychophysiology, 8 211. Vrana, S. R. (1995). Emotional modulation of skin conductance and eyeblink responses to a startle probe. Psychophysiology 32 351. Vrana, S. R., & Lang, P. J. (1990). Fear imagery and the startle probe reflex. Journal of Abnormal Psychology 99 181. Weerts, T. C., & Lang, P. J. (1978). Psychophysio logy of fear imagery: Differences between focal phobia and social performance anxiety. Journal of Consulting and Clinical Psychology, 46, 1157. White, K. D.. (1978). Salivation: The significa nce of imagery in its voluntary control. Psychophysiology, 15 196. White, K. D., Sheehan, P. W., & Ashton, R. (1977). Imagery assessment: A survey of self-report measures. Journal of Mental Imagery, 1 145. Witvliet, C. V., & Vrana, S. R. (1995). Psychoph ysiological responses as indices of affective dimensions. Psychophysiology 32 436. Witvliet, C. V., & Vrana, S. R (2000). Emotiona l imagery, the visual startle, and covariation bias: An affective matching account. Biological Psychology 52 187. Zatorre, R. J. & Halpern, A. R. (2005). Mental concerts: Musical imager y and auditory cortex. Neuron, 47 9. Zinbarg, R.E., Craske, M. G., &, Barlow, D. H. (2006). Mastery of your anxiety and worry (MAW): Therapist guide (2nd ed.). New York, NY: Oxford University Press.

PAGE 87

87 BIOGRAPHICAL SKETCH Lisa M. McTeague is originally from Linc oln, NH and she received her Bachelor of Arts degree in psychology from Harvard College in Cambridge, MA in June 1998. She matriculated to the doctoral program in clinical psychology at the University of Florida to work with Peter J. Lang, Ph.D. and Margaret M. Bradley, Ph.D. in August 2000. She recently completed her predoctoral internship through the UF Department of Clini cal and Health Psychology, focusing on the multimodal assessment and treatment of anxiety and mood disorders. Lisas research pursuits have been motivat ed by a desire to bett er understand emotional processing and the physiological su bstrates and concomitants of fe ar and anxiety. For the last seven years under the mentorship of Drs. Lang and Bradley she has been involved in a series of translational research investigations at the NI MH Center for the Study of Emotion and Attention. These studies have centered on the neurosci ence and psychophysiology of emotion using multiple recording methodologies (e.g., startle reflex, autonomic measures, respiration, electromyography, high density EEG, fMRI, eye-tr acking) in both unselec ted and anxiety and mood disorder patient populations. Most recent ly she spearheaded a program of experiments comparing the extent of defensive reactivity acro ss the anxiety disorder spectrum. These studies, involving over 400 anxiety and mood disorder patients, were mo tivated by the presumption, on the basis of both current diagnostic nosology and anim al models of fear, that all anxiety patients show defense system hyper-reactivity. However, fi ndings to this point suggest a hierarchy of disorders defined by monotonically increasing co morbid psychopathology and negative affect, varying with the degree to which the principal complaint is circumscribed fear (i.e., specific phobia) versus generalized appreh ension (i.e., GAD/panic). On multip le self-report measures of anxiety, depression, and functional interfer ence, specific phobics report the least symptomatology followed by those with social ph obia, panic disorder, PTSD, and lastly, GAD.

PAGE 88

88 Paradoxically, this ascending hierar chy of self-reported di stress is directly associated with a reciprocal reduction in physiol ogical defensive reactivity. These findings echo a growing number of studies that prompt the question motivating much of Lisas current research: Are anxiety spectrum disorders better repres ented as discrete diagnoses or a continuum of severity? In addition to characterizi ng the comparative physiology a nd verbal report of anxiety disorder patients, Lisa has a pronounced interest in PTSD research that piqued in college while she worked as a domestic violence and rape crisis counselor. This curiosity spawned her senior honors thesis conducted under the mentorship of Richard J. McNally, Ph.D. at Harvard University. She developed a multivariate, lifespan model of PTSD in female Vietnam veterans applicable to the National Vietnam Veterans Readjustment Study (NVVRS) database. Eager for exposure to multimodal experimental investigat ion, upon graduation, Lisa began working as a research assistant with Brett Litz, Ph.D. and Mark Miller, Ph.D. at the National Center for PTSD at the Boston VA Medical Center on a parametr ic psychophysiological inve stigation of affective processing in Vietnam and Gulf War veterans. T ogether, these experiences were formative in influencing a lasting involvement in PTSD resear ch. Lisa recently completed an investigation supported by a National Research Service Awa rd (NRSA) Predoctoral Fellowship titled Traumatic Exposure, PTSD, and Physiological Reactivity exploring the reliability and task generalizability of both appetitive and de fensive processing deficits in PTSD. In the future, Lisa intends to inves tigate means by which psychophysiological and neuroscience techniques can be applied to enha nce characterization of symptom constellations, judgments of prognosis, treatment planning, and track ing of treatment progress. For example, if defensive (and/or appetitive) hypor eactivity demonstrated at asse ssment reflects a fundamental incapacity to engage in emotional processing, interv ention specific to this de ficit is warranted at

PAGE 89

89 the outset of treatment, prior to exposure-based or similar treatment method s. In short, Lisa is excited about the potential for psychophysiol ogy to augment conventional assessment and treatment methods. For the upcoming year Lisa w ill be pursuing these res earch interests as a postdoctoral research fellow at the NIMH Ce nter for the Study of Emotion and Attention.