1 ANTICIPATION, AFFECT, AND ATTENTION: CENTRAL AND PERIPHERAL PROCESSES By CHRISTOPHER SEGE A THESIS PRESENTED TO THE GRADUATE SCHOOL OF THE UNIVERSITY OF FLORIDA IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE O F M ASTER OF SCIENCE UNIVERSITY OF FLORIDA 2012
2 2012 Christopher Sege
3 To those who have helped me put into words that which I have known in thought
4 ACKNOWLEDGMENTS I thank the chair, Peter Lang, and members of my supervisory committee f or their mentorship, my research mentors and all the members of my lab for their support and assistance, and the participants in my study for their honest and open participation. I thank my parents, Suzanne and Ronald Sege, my brother, Jonathan Sege, and my friends for their loving encouragement, which motivates me always.
5 TABLE OF CONTENTS ACKNOWLEDGMENTS .................................................................................................. P age 4 LIST OF TABLES ............................................................................................................ 6 LIST OF FIGURES .......................................................................................................... 7 ABSTRACT... 8 CHAPTER 1 INTRODUCTION ................................................................................................... 9 Autonom ic and Somatic Arousal Reflexes ........................................................... 10 Event related Potentials ....................................................................................... 12 The Hearts Response During Anticipati on .......................................................... 14 Research Aims .................................................................................................... 15 2 METHOD ............................................................................................................. 17 Participants .......................................................................................................... 17 Design and Materials ........................................................................................... 17 Physiological Recording and Data Reduction ...................................................... 19 Data Analytic Strategy ......................................................................................... 21 3 RESULTS ............................................................................................................ 23 Pleasantness/ Unpleasantness Ratings .............................................................. 23 Startle Response ................................................................................................. 23 Heart Rate Response .......................................................................................... 24 CNV ..................................................................................................................... 25 4 DISCUSSION ...................................................................................................... 30 Interpretation and Significance ............................................................................ 30 Future Research .................................................................................................. 36 Limitations ........................................................................................................... 37 LIST OF REFERENCES ............................................................................................... 39 BIOGRAPHICAL SKETCH ............................................................................................ 43
6 LIST OF TABLES Table P age 3 1 Post experimental ratings of pleasantness/ unpleasantness .............................. 26 3 2 Magnitude of startle blink response .................................................................... 26 3 3 Mean heart rate deceleration and acceleration .................................................. 26 3 4 Mean continuous EEG dur ing early and late CNV components ......................... 26
7 LIST OF FIGURES Figure page 3 1 Magnitude of startle reflex response during anticipation ..................................... 27 3 2 Heart rate change during anticipation and perception. ....................................... 28 3 3 Event related potential during the 6second anticipatory interval. ...................... 29
8 Abstract of Thesis Presented to the Graduate School of the University of Florida in Partial Fulfillment of the Requirements for the Degree of Master of Science ANTICIPATION, AFFECT, AND ATTENTION: CENTRAL AND PERIPHERAL PROCESSES By Christopher Sege May 2012 Chair: Peter Lang Major: Psychology Human emoti on is built on animal reflexes and neural circuits that serve basic evolutionary functions. In the context of anticipation of salient upcoming stimuli, these reflexes show increasing reactivity as the anticipated stimulus approaches, and serve to prepare t he organism to respond appropriately to the stimulus when it appears. The current study sought to characterize physiological reactivity in humans during the anticipation of emotional pictures across several modalities. Participants saw colored cues for 6 s econds that indicated what kind of picture (romance, violence, or mundane) would be seen next. Participants startle responses to discrete probe stimuli, heart rate responses, and electroencephalography were measured while they passively viewed cue stimuli and pictures. Analysis of physiology during cues revealed a general enhancement of responding for the anticipation of romance and violence pictures; startle potentiation, negative voltage frontal EEG, and heart rate deceleration all increased as picture o nset approached. Implications for theory of emotion and future research on anticipation are discussed.
9 CHAPTER 1 INTRODUCTION The view of emotion motivating this study is founded on a well established animal model. Emotional reactions reflect the activat ion of defensive and appetitive motive circuits in the brain, centered around the amygdala, which mediate a range of autonomic and somatic reflexes that evolved to ensure the survival of organisms and their progeny (see Davis, 1997 and Lang & Davis, 2006 f or reviews). Thus, in the context of a danger cue (e.g., a predator), reactions include freezing (Fanselow, 1994), fear bradycardia (Campbell, Wood, & McBride, 1997), and an enhanced startle response (Davis, 1998), all mediated by projections from the am ygdala to efferent cortical structures in the circuit (Kapp, Whalen, Supple, & Pascoe, 1992; Davis, 1998; Lang & Davis, 2006). If the threat is imminent, however, an alarm reaction develops, which involves a shift from heart rate deceleration to acceleration and from freezing to behavioral activation (Masterson & Crawford, 1982; Faneslow, 1994). Interestingly, responding is similar in the predator as it is in the prey; as a potential food source nears, patterns of responding shift from a dampening of reac tivity, which facilitates sensory intake, to sympathetic activation (and metabolic arousal), which allows for a rapid strike. Thus, overlapping limbic structures, including the amygdala, mediate motive response in both appetitive and defensive arousal, wit h similar physical reactions recruited to serve different goals. Research in humans has suggested that exposure to emotional stimuli in different contexts modulates many of the same physiological reactions defined by the animal model, and that this modulat ion is mediated by similar corticolimbic circuitry. Research using emotional imagery paradigms has consistently shown evidence of greater
10 potentiation of the blink startle reflex, heart rate acceleration, skin conductance increases, and increases in activ ation of facial electromyography when participants are asked to imagine emotional, compared to neutral, scenes ( Miller, Levin, Kozak, Cook, McLean, & Lang, 1987 ; Witvliet & Vrana, 1995; Miller, Patrick, & Levenston, 2002). Viewing emotional pictures also m odulates responding in these modalities, in a manner that is similar to the animal spotting a salient stimulus at a distance (Lang, Bradley, & Cuthbert, 1997; Bradley, Codispoti, Cuthbert, & Lang, 2001). Importantly, functional magnetic resonance imaging ( fMRI) research has shown that similar areas of the limbic architecture, including the amygdala and the insula, are activated whether humans are imagining emotional events (Sabatinelli, Lang, Bradley, & Flaisch, 2006; Costa, Lang, Sabatinelli, Versace, & Br adley, 2010), viewing emotional pictures (Sabatinelli, Lang, Keil, & Bradley, 2006), or put into anxiety provoking situations (e.g., Phelps, OConnor, Gatenby, Gore, Grillon, & Davis, 2001). Data such as these have lent credence to a suggestion that human emotions represent action dispositions, which serve functional goals in evolutionary history (e.g., sensory intake, selection of approach/ avoidance behavior, facilitation of contingency learning). These data further suggest that emotional action disposi tions are built on the same basic building blocks as are animal responses during situations motivating threat or approach (Frijda, 1987; Lang, 1994). Autonomic and Somatic Arousal Reflexes: Gradients of Approach and Avoidance When animals are placed in a c ontext of potential reward or threat, strength of responding (e.g., approach or avoidance) increases in a graded fashion with proximity to the goal (Miller, 1951). This response gradient is stronger for avoidance situations than it is for approach, given t he often more immediate survival relevance of events that
11 threaten life (Miller, 1951; Kahneman & Tversky, 2000). A recent study investigated an analogous situation in humans in which the temporal proximity of monetary gains or losses was systematically va ried (Lw, Lang, Smith, & Bradley, 2008). This study examined physiological reactions (including skin conductance, heart rate, and startle probe response) as participants anticipated making a rapid motor response (key press) to attain a reward or avoid a loss. As predicted by the animal model, skin conductance increased and heart rate decelerated with the approach of the stimulus signaling a response. Interestingly, startle potentiation diminished over the anticipatory period, possibly reflecting general m otor inhibition that facilitated attention to the critical final stimulus signaling need for a response. Taken together, these findings elucidate both the functional and graded nature of anticipation in humans; changes in physiology reflect an evolving coping action disposition, as the individual increases readiness to attain a reward or to try to avoid a punishing loss (Lw et al., 2008). Another body of research has provided evidence that physiological reactivity is modulated in the anticipation of aversi ve stimuli even when participants do not have the possibility of response control. Evidence for modulation of autonomic and somatic reflexes has been found in participants anticipating aversive physical stimuli such as shocks (e.g., Grillon, Ameli, Woods, Merikangas, & Davis, 1991) and noise blasts (e.g., Skolnick & Davidson, 2002), and has been shown during the anticipation of arousing pictures as well. A previous study examined the anticipation of pleasant (erotic), neutral, and unpleasant fear relevant ( snake) pictures in a sample of highly snakefearful men (Sabatinelli, Bradley, & Lang, 2001). Importantly, this study reported that startle reflex magnitude was enhanced prior to viewing emotional pictures but not neutral pictures
12 (Sabatinelli et al., 2001 ). A subsequent study replicated the finding of enhanced startle during anticipation of unpleasant pictures in nonanxious individuals; however, only weak evidence of potentiation in advance of pleasant pictures was reported in this study (Nitschke, Larson Smoller, Navin, Pederson, Ruffalo, et al., 2002). An interesting distinction between studies investigating anticipation of pictures or other emotional stimuli and those investigating anticipation of an active, motivated response is the discrepancy between the general increasing inhibition of the startle response noted in activeresponse anticipation (Lw et al., 2008) and the overall potentiation of reflexes noted during anticipation of pictures (Sabatinelli et al., 2001; Nitschke et al., 2002). It is as yet unclear if startle potentiation during anticipation of passive events, such as picture viewing, can be characterized in terms of response gradients such as those suggested by animal (Miller, 1951) and humans (Kahneman & Tversky, 2000; Lw et al., 2008) research. If responding in anticipation of pictures occurs in a graded fashion, a shallower gradient in advance of pleasant stimuli may account for discrepant patterns of startle response modulation found during pleasant anticipation across studies (Sabat inelli et al., 2001; Nitschke et al., 2002). Event related Potentials: Brain Responses in Anticipation Electroencephalographic (EEG) responses evoked by exposure to environmental stimuli, called event related potentials (ERPs) when timelocked to stimulus presentations, are modulated in a manner similar to physiological reflexes in the face of salient stimuli. Of particular interest is a frontal negative ERP called the Contingent Negative Variation (CNV), which develops when a cue signals the upcoming presentation of a significant stimulus. The waveform is comprised of two components; a
13 faster early component that reflects an orienting response to the cue (Loveless & Sanford, 1974), and a later, voltagenegative component that increases monotonically over the interval preceding the expected stimulus (Loveless & Sanford, 1974; Rohrbaugh & Gaillard, 1983). The later component has been the subject of much study in emotional anticipation, and is thought to alternatively reflect both expectancy and motor preparati on (Rohrbaugh & Gaillard, 1983). Similarly to the startle reflex, the CNV has been found to be modulated during anticipation of emotional pictures. In one study, anticipation of an emotionally arousing pleasant picture resulted in a more robust late component of the CNV than did anticipation of a neutral picture, but only under conditions in which a response to the picture was required or when the picture was presented for a brief time (Simons, Ohman, & Lang, 1979). A more recent study examined the late CNV in advance of a wider variety of emotional and neutral pictures, and found enhanced late CNV responding prior to the presentation of higharousing emotional stimuli in the absence of a required motor response (Poli, Sarlo, Bortoletto Buodo, & Palomba, 2007). While earlier studies implied a need for increased salience over andabove the arousal value of an anticipated picture (Simons et al., 1979), this more recent report suggests that emotional salience is in itself enough to evoke an enhanced CNV (Poli et al., 2007). One factor possibly accounting for the discrepancy between these reports is the type of experimental stimuli used; while the more recent study employed a wide variety of pleasant and unpleasant pictures (Poli et al., 2007), the earlier study c ited above focused only on the anticipation of higharousing erotic pictures and neutral pictures (Simons et al., 1979). Given that anticipation of reward has been found to be less potent
14 than anticipation of punishment, in accordance with the animal model (Miller, 1951; Kahneman & Tversky, 2000), it is possible that these reports are actually indirectly indicative of this distinction. Further work explicating the CNV in the anticipation of both pleasant and unpleasant stimuli is required to systematically examine approach/ avoidance differences in this modality. As yet, the effects of emotional anticipation on the early CNV component are also unknown. To date, research on the CNV in anticipation has focused on the late component, with the assumption that this component is likely to reflect emotional salience inasmuch as it is more closely linked to processing of the upcoming stimulus. However, it is possible that the early component, and by extension processing of a cuing stimulus, may signal emotional sali ence as well, by virtue of the contingent association between the cue stimulus and the upcoming stimulus. Modulation of response to anticipatory cues may thus represent a functional aspect of the anticipatory response cascade, just as much later preparator y responding does. The Hearts Response During Anticipation Studies on the CNV have often concurrently measured heart responding, as it is itself a wellvalidated index of cognitive and attentional processes (Lacey, 1967). Distinct heart rate components have been associated with different psychological processes; specifically, heart rate increases, or acceleration, have been associated with increased cognitive load or engagement (as in imagery, e.g., McNeil, Vrana, Melamed, Cuthbert, & Lang, 1993), while heart rate decreases, or decelerations, have been associated with orienting to environmental stimuli (as in fear bradycardia, e.g., Campbell et al., 1997). This distinction has been made in the emotional literature as well, with
15 heart rate deceleration found after onset of emotional pictures (Bradley et al., 2007), while emotional imagery is marked by heart rate acceleration (e.g., McNeil et al., 1993). In anticipation, heart rate shows a reliable triphasic pattern following a cue stimulus; an initial deceler atory component, reflecting orienting to the cue stimulus, is followed by a marked acceleration and then by a second sharp deceleration, which again is thought to represent preparation for the upcoming stimulus (Graham, 1967; Simons et al., 1979; Poli et al., 2007). Given that, similar to the ERP, heart rate response reflects both orienting to a cue and expectancy of an anticipated stimulus, measuring heart rate allows an opportunity to provide convergent evidence of both enhanced orienting to the cue stimulus and increasing preparatory responding to the anticipated stimulus. Research Aims The current study explored motivational reactivity during anticipation of emotionally arousing pictures, emphasizing examination of early reactivity and the later developm ent of gradients of preparatory activation. To this end, several response modalities were measured within subject, specifically blink startle response, heart rate response, and cortical EEG. In the design presented, participants viewed cues of varying co lor that systematically cued the content of an upcoming picture. The interval preceding the picture in this study was designed to be of sufficient length to examine early and late components of both the CNV and heart rate response, and to allow presentati on of startle probes at several points later in anticipation. Overall, anticipation of arousing emotional pictures in this study was hypothesized to mimic anticipatory activity in animals approaching reward or avoiding threat. Specific hypotheses guiding t his research were as follows: first, based on findings of potentiated
16 startle in anticipation of pictures (Sabatinelli et al., 2001, Nitschke et al., 2002), it was expected that startle potentiation for emotional anticipation would increase monotonically w ith the imminence of an upcoming picture. It was further believed that this increase would be greater in anticipation of unpleasant than pleasant stimuli, in accordance with the animal model (Miller, 1951). In addition, it was expected that timing dependent emotional modulation in both heart rate and frontal ERP would arise during the later half of the anticipation interval, in the form of anticipatory heart rate deceleration and an increasingly negative slow wave ERP response. Finally, early responses in heart rate and ERP were of interest in this study; these responses were expected to reflect a general orienting to the cuing stimulus, in the form of an initial heart rate deceleration and a marked early negative waveform in frontal ERP. This research will determine if these initial responses vary significantly with the salience of an anticipated stimulus.
17 CHAPTER 2 METHOD Participants Participants in this study were 38 undergraduate students (19 female; mean age=20 yrs) from General Psychology courses at the University of Florida who participated for course credit. Participants were recruited through the universitys online research participation scheduling system. The study was approved by the University of Florida Institutional Review Board, and informed c onsent was attained from all participants prior to participation. Data for one participant were removed from all analyses due to excessive movement. Startle data were removed for three additional participants due to small number of startles, resulting in a dataset of 34 participants with complete startle data. Heart rate data were removed for two participants due to excessive noise, resulting in a dataset of 35 participants with complete heart rate data. Continuous EEG data for 4 participants were removed due to excessive trial loss, resulting in a dataset of 33 participants with complete EEG data. A set of 30 participants had complete data in all three response modalities. Design and Materials In this procedure, differently colored cues were used to indica te what kind of picture would be seen next. Color cues were red, blue, green, or yellow squares, (360 X 360 pixels) presented centrally on a computer monitor. Colors were chosen from a colorblindsafe color palette to ensure cues could be discriminated by all participants. Picture stimuli were selected from the International Affective Picture System (Lang, Bradley, & Cuthbert, 2008) and depicted 3 content areas that varied in hedonic valence:
18 romance (erotica and romantic couples); violence (attacks and mut ilations); and mundane scenes (e.g., a woman reading on her lawn). Ten pictures were chosen for each category (20 for each content area [romance, violence, mundane]). A fourth picture category, not included here, were 20 pictures of geometric shapes, creat ed using Adobe Photoshop CS 10 software ( Adobe Systems Incorporated). An additional picture from each IAPS and shape category was chosen for the four practice trials. All pictures were converted to grayscale and presented at 1024 X 768 pixels. Normat ive ratings of pleasantness/ unpleasantness for pictures in the romance category (M = 6.71, SD = 0.4) were significantly higher (i.e., more pleasant) than in the mundane (M = 5.36, SD = 0.4; t = 3.09, df = 57, p < .05) and violence (M = 1.86, SD = 0.4; t = 6.60, df = 57, p < .05) categories. Ratings for pictures in the violence category were significantly lower (i.e., more unpleasant) than pictures in the mundane category ( t = 5.26, df = 57, p < .05). Normative arousal ratings for romance (M = 6.53, SD = 0. 7) and violence categories (M = 6.82, SD = 0.3) were not significantly different from each other, but both were significantly higher (i.e., more arousing) than for mundane pictures (M = 3.47, SD = 0.5; romance, t = 4.23, df = 57, p < .05, violence, t = 5.0 2, df = 57, p < .05). Pictures in each content area were matched for brightness and rated complexity. Experimental stimulus presentation and timing was controlled using Presentation v11.3 software ( Neurobehavioral Systems, Inc.), while physiological signals were acquired using VPM v2.0 (Cook, 1994) software run on a separate computer. Stimuli were presented on a 39 cm high X 70 cm wide LCD monitor 1.5 m from the participant. The acoustic startle probe was a 96 dB, 50ms burst of white noise with instant aneous rise time, produced by a Coulbourn S81 02 noise generator, gated
19 through a Coulbourn S82 24 amplifier, and presented binaurally over matched Telephonics TDH49 headphones. Net Station ( Electrical Geodesic Systems) software was used to acquir e EEG data. A fixation cross remained in the center of the screen throughout the entire experiment. Each trial began with the presentation of a cue for 6 seconds, followed by the 3second presentation of a picture. A variable length inter trial interval (ITI; 12, 15, 18, or 21 seconds) followed picture offset. One startle probe was presented on every trial either during the anticipation interval, during the picture perception interval, or during the ITI. Probes presented during the anticipation interval were presented 3, 4, or 5 seconds after the onset of the cue. Probes presented during the perception period were presented 2 seconds after onset of the picture. Probes presented during the ITI were presented 7.5 seconds after picture offset. In all, 16 startle probes were presented at each of the three anticipation presentation times, during picture perception, and during the ITI. One probe was presented during each practice trial. Trials were arranged in 10 blocks of 8, such that 2 pictures from each of t he 4 categories appeared in each block. Four picture presentation orders were generated such that, across participants, every picture was seen equally often in the first or second half of a block and the first, second, third, or fourth portion of the exper iment. The color predicting each picture type was uniformly counterbalanced across participants. Physiological Recording and Data Reduction The eyeblink component of the startle reflex was recorded electromyographically from the orbicularis oculi muscle beneath the left eye using miniature Ag/AgCl electrodes placed 1.5 cm apart. The raw signal was amplified by 30,000 and bandpass
20 filtered at 90 250 Hz using a Coulbourn S75 01 amplifier. The amplified signal was then rectified and integrated using a Coul bourn S76 01 contour following integrator using a time constant of 20 ms, and sampled at 20 Hz beginning 3 s before each trial. Sampling increased to 1000 Hz 50 ms prior to the onset of the startle probe and continued for 250 ms. Sampling then continued at 20 Hz until 3 s after picture offset. Blinks were scored off line with an interactive computer program that scored onset latency and peak. Heart rate was measured continuously using Ag/AgCl electrodes placed ~ 3 cm below the elbow joint on the inside o f each forearm. Raw signal was amplified and bandpass filtered at 8 40 Hz using a Coulbourn S75 01 amplifier and fed into a Coulbourn bipolar comparator (S2106) and retriggerable oneshot (S52 12), which registered the occurrence of the R spike of e ach QRS complex of the heart rate waveform and converted it to a digital pulse. The time in milliseconds between each of these pulses was converted to beats per minute (BPM). EEG was measured continuously using a 129 channel sensor net. EEG was recorded wi th a sampling rate of 250 samples per second and online filtering from 0.0150 Hz. EEG was recorded continuously and segmented offline into 6 second anticipation and 3second picture viewing epochs. Data from each epoch were refiltered offline using a 30 Hz low pass filter, eyemovement and artifact corrected, and baselinecorrected relative to 100ms precue or picture onset. Offline processing of data was accomplished using ElectroMagnetic EncephaloGraphy Software (EMEGS; Junghofer & Peyk).
21 Data Analyt ic Strategy A series of repeatedmeasures multivariate ANOVAs (RMANOVA) and follow up pairedsample t test comparisons were used to investigate all hypotheses. All factors were treated as withinsubjects repeatedmeasures in all analyses. Post experimental ratings of pleasantness/ unpleasantness were analyzed with separate RMANOVAs for anticipation and perception of pictures, with Content (romance, mundane scene, violence) as the repeatedmeasures factor. The magnitude of each subjects startle responses to probes during anticipation and perception were standardized relative to the mean and standard deviation of startles elicited during ITIs and expressed as T scores ((z*10)+50). Responses during anticipation were first analyzed in a 3 (Content: romance, mun dane scene, violence) X 3 (Probe Time: 3, 4, or 5 seconds) RMANOVA. Follow up univariate RMANOVAs and pairedsamples t tests were conducted to determine content differences at each probe time. In addition to RMANOVA analyses, difference scores were calculated by subtracting magnitude of response during mundane scene cues from magnitude during romance and violence cues at each probe time. These difference scores were analyzed in a 2(Content: romance neutral, violent neutral) X 3(Probe Time) polynomial tr end analysis with follow up analyses for each content separately. Responses during each cue at each probe time were compared using pairedsamples t tests. Finally, responses during picture perception were analyzed in a 5level onefactor (Category: erotic, romance, mundane scene, mutilation, attack) RMANOVA. A well established method for analyzing heart rate data during anticipation was employed here (Gatchel & Lang, 1973; see also Headrick & Graham, 1969, and Lang & Hnatiow, 1962). Visual inspection of heart rate response waveforms during pleasant,
22 neutral, and unpleasant anticipation revealed the expected triphasic waveform, with an initial deceleratory component, a subsequent acceleratory component, and a second, sharper deceleration. For each partici pant, the most negative heart rate changefrom baseline during a window from 02 seconds post cue onset was taken as a measure of peak initial deceleration (D1). Maximum changefrom D1 during a window from 24 seconds post cue onset was taken as a measure of peak heart rate acceleration (A1) during anticipation. The largest negative (or least positive) changefrom A1from 4 6 seconds post cue onset was taken as a measure of second heart rate deceleration (D2) during anticipation. Finally, deceleration during picture viewing was indexed as the most negative (or least positive) changefrom A1 during the entire 3second picture viewing period. Separate 3level onefactor (Content) RMANOVAs were used to examine each component during anticipation and perception. ERPs at each sensor, averaged across all trials and participants, were used in a multivariate cluster analysis to determine spatial clusters of sensors. Visual inspection of a frontocentral cluster of sensors (depicted in Figure 3) indicated the expected two components of the CNV; an early component during the first second of anticipation, and a later slow component for the rest of the interval. Based on this visual inspection, mean ERPs during three different time windows were used in analyses; a 5501000 ms window (the early component), a 10003000ms window (the first half of the late component), and a 30006000ms window (the second half of the late component). A onefactor (Content) RMANOVA was used to investigate ERP amplitude separately during each time window.
23 CHAPTER 3 RESULTS Pleasantness/ Unpleasantness Ratings Table 31 lists the means and standard deviations for post experimental ratings of pleasantness/ unpleasantness. Participants rated violent pictures as significantly more negative than mundan e pictures both when they were anticipating them ( t = 5.38, df = 35, p < .001) and looking at them ( t = 7.97, df = 35, p < .001). Conversely, romantic pictures were rated as significantly more pleasant than mundane pictures during anticipation ( t = 2.73, d f = 35, p < .001) and perception ( t = 2.63, df = 35 p < .001). Startle Response Figure 31 illustrates mean startle responses to probes presented anytime during anticipation (1A) and to each probe during anticipation (1B); Table 32 lists means, standard deviations, and outcomes of statistical tests for both anticipation and perception. During anticipation, the content of the upcoming picture reliably modulated startle reflex magnitude ( F (2, 32) = 7.28, p < .001). Follow up comparisons indicated that respo nse to probes was potentiated during anticipation of violence, relative to mundane anticipation ( t = 3.84, df = 33, p < .001), and during romance anticipation as well, at a trend level of significance ( t = 1.90, df = 33, p = .06). Although the Content X Probe Time interaction was not statistically reliable, a priori hypotheses, based on previous studies showing that startle response modulation is sensitive to latency from cue onset (e.g., Anthony & Davis, 1985), were tested. Startle responses were reliably modulated at the probe time most proximal to picture onset, with potentiation of the startle response while anticipating either romance ( t = 2.07, df = 33, p < .05) or violence ( t = 3.76, df = 33, p < .001), compared to mundane
24 events. Emotional modulat ion of the startle response was not significant at the earliest probe time and was marginal for the middle probe time ( t = 1.94, df = 33, p = .06) for violent cues only. Analyses of difference scores (emotion neutral) revealed that potentiation of response generally increased across anticipation, as indicated by a significant linear trend effect of Probe Time ( F (1, 33) = 5.27, p < .05). Follow up tests indicated that startle potentiation increased linearly while anticipating violence (linear trend, F (1, 3 3) = 6.52, p < .05) but not romance pictures. However, during romantic cues, startle potentiation was greater in response to the 5 second than the 4second probe, at a trend level of significance ( t = 1.85, df = 33, p = .07). As repeatedly observed in prev ious research (e.g., Bradley et al., 2001), startle magnitude was inhibited while viewing romantic as compared to mundane ( t = 2.56, df = 33, p < .05) and mutilation ( t = 1.85, df = 33, p = .07) pictures. However, the expected potentiation for attack and m utilation pictures was not observed (see Table 32). Heart Rate Response Figure 32 illustrates the heart rate responses during anticipation and perception; Table 33 lists means for heart rate components across participants, standard deviations, and outco mes of statistical tests. The heart rate response during anticipation consisted of a triphasic waveform, with an initial, small deceleration (D1) followed by a marked acceleration (A1) in the middle of the interval, and in turn by a sharp deceleration (D2) as picture onset approached. Analyses of D1 revealed a significant effect of Content ( F (2, 33) = 3.89, p < .05); deceleration was greater when anticipating romance ( t = 2.10, df = 34, p < .05) or violence ( t = 3.55, df = 34, p < .05) than when anticipatin g mundane scenes. There were no effects of Content on A1. However, the
25 Content main effect was once again significant for D2 ( F (2, 33) = 4.75, p < .05), with greater deceleration when anticipating onset of either romance ( t = 2.36, df = 34, p < .05) or vi olence ( t = 2.36, df = 34, p < .05), compared to mundane scenes. While viewing pictures, heart rate continued to decelerate more for higharousing than low arousing pictures, as indicated by a significant main effect of Content ( F (2, 33) = 9.35, p < .001) Both scenes of romance ( t = 3.73, df = 34, p < .001) and violence ( t = 3.99, df = 34, p < .001) prompted greater deceleration than did mundane scenes. CNV Multivariate cluster analysis of ERP response recorded at each sensor during anticipation identified a negative slow wave response in a frontocentral cluster of sensors. Figure 33 illustrates the ERP waveform across all 6 seconds of anticipation in this sensor cluster, and Table 34 lists means, standard deviations, and the outcome of statistical test s for each window of analysis. Examination of the early component of the CNV in frontocentral sensors revealed a more negative wave during anticipation of both romance ( t = 2.20, df = 32, p < .05) and violence ( t = 2.37, df = 32, p < .05) relative to mundane scene anticipation. Examination of the later CNV component revealed a more negative slow wave during anticipation of violence than the anticipation of mundane scenes in the first half ( t = 2.15, df = 32, p < .05) and second half ( t = 2.15, df = 32, p < .05) of this component. When anticipating pictures of romance, average ERP response was not significantly different from mundane scene cues during the first half of the late wave, but was more negative, at a trend level, during the second half ( t = 1.87, df = 32, p = .07).
26 Table 31. Post experimental ratings of pleasantness/ unpleasantness of anticipation and perception Mean (SD) Romance Mundane Violence F Anticipation 5.24 (1.2) a 4.52 (1.0) b 3.04 (1.7) c 39.31*** Perception 5.44 (1.0) a 4.76 (1.1) b 2.64 (1.1) c 48.98*** Note: subscripts indicate difference between levels at p < .001. *** p < .001. Table 32. Magnitude of startle blink response (T scored) to 3second, 4second, and 5 second probes in anticipation and perception Mean (SD) Pro be Time Romance Mundane Violence F Anticipation 3 s 50.58 (5.5) 51.10 (8.2) 52.23 (9.3) 0.50 4 s 50.21 (4.7) 50.74 (6.1) 53.08 (6.3) 2.58^ 5 s 50.95 (7.2) a 47.66 (6.4) b 52.86 (7.4) a 7.00** Mean (SD) Erotic Romance Mundane Mutilation Attack F Per ception 48.42 (7.0) 46.80 (8.2) a 50.66 (6.0) b 50.40 (7.5) b 49.82 (9.9) 1.68 Note: subscripts indicate significant differences between levels. ^ p < .10; ** p < .01. Table 33. Mean heart rate deceleration during D1, D2, and perception, and maximum heart rate acceleration (A1) during anticipation Mean (SD) Component Romance Mundane Violence F Anticipation D1 2.42 (1.5) a 1.87 (1.0) b 2.38 (1.2) a 3.89* A1 7.57 (2.9) 7.39 (2.9) 7.32 (2.8) 0.47 D2 10.48 (4.4) a 9.19 (3.8) b 10.18 (4.3) a 4.75* Per ception 13.82 (5.2) a 12.09 (4.9) b 13.61 (4.8) a 10.78*** Note: D1 = first deceleration (02s) in anticipation; A1 = acceleration (24s) during anticipation; D2 = second deceleration (24s) during anticipation. Subscripts indicate difference between lev els. p < .05; *** p < .001. Table 34. Mean continuous EEG during the early and late components of the CNV during anticipation Mean (SD) Romance Mundane Violence F Early (550 1000ms) 2.26 (2.1) a 0.90 (3.0) b 2.30 (3.4) a 3.89* Late, first half (1 000 3000ms) 0.63 (2.0) 0.32 (2.5) a 1.18 (2.7) b 3.24^ Late, second half (3000 6000ms) 1.03 (2.1) a 0.14 (2.7) b 1.41 (2.4) a 3.46* Note: subscripts indicate significant/ trend dif. between levels. ^ p < .10; p < .05.
27 Figure 31. Magnitude of startl e reflex response (T scored) A) during anticipation (ITI=50), B) to startle probes presented 3, 4, and 5 seconds following onset of the anticipatory cue for pleasant and unpleasant cues (deviated from reflexes elicited when anticipating everyday scenes).
28 Figure 32. A) Heart rate change during anticipation and perception. B) Heart rate change deviated from D1 during anticipation. C) Maximum heart rate deceleration 02 seconds following anticipatory cue onset. D) Maximum heart rate accelerati on 2 4 seconds following anticipatory cue onset, deviated from D1. E) Maximum heart rate deceleration 46 seconds following anticipatory cue onset and during picture presentation, deviated from A1.
29 Figure 33. A) Event related potential over frontal s ensors during the 6second anticipatory interval; ERPs from 16 seconds are averaged into 500ms bins. B) Mean frontal ERPs from 5501000ms following cue onset. C) Average ERPs during early (10003000ms) and late (30006000ms) windows of anticipation.
30 C HAPTER 4 DISCUSSION Interpretation and Significance In this study, p hysiological reactions across time were assessed during the antic ipation of emotional pictures. Overall, results were suggestive of robust preparatory responding across response modalities that was enhanced when anticipating the upcoming presentation of emotio nal, highly arousing pictures. Both heart rate and cortical ERP showed an initial orienting response to cues that was heightened for cues predicting pictures of romance and violence than for pictures of mundane scenes; in heart rate, this orienting response was an i nitial deceleration, and in the ERP a negative voltage wave occurred over frontal sensors from 5001000ms after cue onset. All response modalities subsequently evidenced i ncreasing response modulation during the latter half of the anticipatory interval. Heart rate decelerated prio r to picture onset, frontal ERP showed a slow, negative voltage wave that increased as anticipation continued, and startle responses were potentiated most for startle probes presented m ost proximal to picture onset. Each of these responses was enhanced (in the case of heart rate) or effected (in the case of ERP and startle blink) when anticipating highly arousing, emotional pictures, supporting the hypothesis that anticipation of emotional pictures elicits ro bust physiological responding. Across modalities, physiology showed a gr aded enhancement as the onset of picture presentation became more imminent consistent with the animal model (Miller, 1951) T his graded response was heightened in advance of unpleasant compared to pleasant picture presentation, consistent with the idea suggested by animal research that avoidance gradi ents are stronger than approach ones (Miller, 1951; Kahneman & Tversky, 200)
31 Modulation of startle reflexes during anticipation were consistent with previous data (Sabatinelli et al., 2001; Nitschke et al., 2002), with heightened reflexes when anticipating the presentation of highly arousing pictures. Moreover, current data extend these findings by characterizing gradients in reflex modulation acr oss the anticipatory interval. Because the startle reflex is sensitive to attentional, action, and motivational demands, it is modulated by emotion differently depending on the task conte xt (Bradley, 2000). In perceptual contexts, modulation is sensitive to the emotional valence (pleasant, neutral, or unpleasant) of the picture (Bradley, Codispoti, & Lang, 2006), reflecting activation of appetitive or avoidant motivational systems (Bradley et al., 2006). In situations where emotional events are mentally imagined, however, the startle response is potentiated whether imagined scenes are pleasant or unpleasant, compared to neutral, and is thus taken to primarily index processes relating to act ion engagement that are more robustly evoked by highly arousing imagined events (Miller et al., 2002). Because anticipation of emotional pictures in the current study was associated with increasing startle potentiation, it appears that anticipation evokes similar mobilization for action as seen during mental imagery, regardless of whether the specific action is approachor avoidancemotivated. Unlike in imagery, however, this mobilization increases over time, in preparation for the presentation of increasi ngly imminent anticipated emotional event. With t his account of startle potentiation during anticipation in mind it is interesting to note the difference between the curren t findings and those reported during anticipation of making a response to receive a reward or avoid punishment ( Lw et al. 2008) I n anticipation of an active response, startle reflex has been found to be
32 inhibited, rather than p otentiated, as the onset of a motive stimulus approaches ( Lw et al. 2008) It is possible that the different patterns in these reflex data reflect different contextual demands, specifically in whether the stimulus anticipated connotes an active r esponse or passive perception. Report of startle inhibition prior to a motivated reaction time task is consistent wit h previous findings of startle inhibition prior to simple reaction time tasks (e.g., Anthony, 1985), and potentially indicates a general dampening of motor activity to facilitate detect ion of and reaction to a particular response stimulus. However, w hen anticipating picture perception as in the current study, motor quies cence is not functional, and potentiation rather than inhibition is seen as a function of increasing arousal assoc iated with imminence of an emotional stimulus. These differences in startle modulation are all the more interesting given the parallels between heart rate reac tivity in these two paradigms. Innervations of the heart by both the parasympathetic and sympathetic divisions of the autonomic nervous system interact to slow or speed up heart rate depending upon current environmental deman ds (Sokolov & Cacioppo, 1997). When a stimulus appears in the environment, heart rate decelerates under primarily parasympathetic influence to facilitate orienting and stimulus intake, such as that seen i n an animal threatened by a distant sti mulus (Campbell et al., 1997). Conversely, heart rate accelerates under sympathetic influence in situations where resources must be rapidly organized (such as when a threat approaches rapidly; Masterson & Crawford, 1982), in situations of heavy cognitive load (e.g., Althaus, Mulder, Mulder, Van Roon, & Minderaa, 1998), and in situations in which action is mentally generated (such as in imager y; e.g., Miller et al., 2002). The literature on the modulation of heart rate response in emotional contexts is consistent with these
33 distinctions, as heart rate decelerates more markedly following exposure to emotional than neutral pictures (Bradley et al., 2001), while in emotional imagery paradigms it is heart rate acceleration t hat is the more pronounced when imagining emotional, compared to neutral, scenes (Witvliet & Vrana, 1995; Miller et al., 2002). In anticipation, the triphasic cardiac waveform found i n the current study is typical. Heart rate is primarily deceleratory as an anticipated stimulus approaches regardless of whether the upcoming stimulus involves making a response (Lacey & Lacey, 1964, 1966; Chase, Graham, & Graham, 1968; Simons et al., 1979) or not (Graham, Putnam, & Lev itt, 1975; Poli et al., 2007). As such, he art rate deceleration seems to represent a general orientation to either a just presented or a tobe presented stimulus, regardless of the task demands associated with that stimulus. Given their apparent differential sensitivity to task demands, then, it i s understandable that startle findings were discrepant between the current study and a previous report of active anticipation (e.g., Lw et al., 2008), while patterns of heart rate m odulation found in these studies were similar. Data suggest that, in the active context, anticipatory orientation involves simultaneously increasing attentional vigilance and readiness to make a functional response. In the passive context, current data suggest that decreasing discrimination of responding and a more generally aroused state accompany increasing vigilance. Finally, the ERP data in this study are co nsistent with heart rate in show ing increasing modulation to anticipated stimuli as their presentation approaches in time. D ifferently from heart rate, however, ERPs during anticipation appear to be unique to anti cipation of emotional stimuli. This is consistent with previous research on the CNV, which indicates that the late component of the CNV arises only under conditions in
34 which an anticipated stimulus is in some way s alient (Poli et al., 2007). The CNV has been found to be separately sensitive to different forms of upcoming stimulus salience, including indication of required motor response (Rohrbaugh & Gaillard, 1983), nonaffective attentional capture (e.g., Poon, Tho mson, Williams, & Marsh, 1974), and approach or avoidance relevance (Simons et al., 1979; Poli et al., 2007). The CNV therefore, appears to index expectancy that occurs when an anticipated stimulus is both motiva ting and when it requires special attention (Loveless & Sanford, 1974). The appearance of the CNV in situations of emotional relevance is consistent with a view of emotion as an action disposition (Frijda, 1987). That is, emotional anticipation evokes a similar ERP response as does motor anticipati on, suggesting that emotional stimulus exposure reflects a type of lowlevel response context. Thus, the anticipating organism must respond to a stimulus, when it appears, in an appropriate fashion. Consistent with this idea are data that show an apparent additivity effect of motor response requirement and emotional anticipation (e.g., Simons et al., 1979). In these situations, emotional anticipation is enhanced by the presence of a motor response set, which facil itates response to stimuli that otherwise do not entail an overt response. Thus, the ERP data are consistent with heart rate in suggesting that anticipation of emotional pictures is a situation that involves preparation to quickly detect a motivating stimulus even though this context does not inv olve an overt, organized motor response. In this study, enhancement of early orientation to cues predicting emotional stimuli was also an aspect of the anticipatory response, and was apparent i n both heart rate and the ERP. E arly modulation can be explained in terms of a number of different potential processes. First, it could indicate a type of conditioning process, in which the
35 cue comes to possess arousal value through its association w ith the anticipated picture. Perhaps a more likely explanation is t hat enhancement of early orienting to emotional cues reflects compliance with task instructions and immediate anticipatory engagement. This intriguing possibility is worthy of further study, to determine the exact nature of early orienting differences and their function in the anticipatory response in humans. Overall, the current data along with other research suggest that an ticipation of emotion al stimuli serves to modulate physiological reactivity, and that the exact nature of this modulation is d etermine d by response demands. In the current study, responding found during anticipation of emotional events is consistent generally increasing arousal, characterized by both enhanced attentional vigilance and indiscriminate motor reflex facilitation. Previous work (Low et al., 2008) suggests that the addition of a contingent response component to this context results in subsequent specification of motor responding rather than generalization of response, while increasing attentional vigilance remains. Taken togeth er, these findings provide preliminary evidence that emotional tone generally increases organismic arousal during anticipation, while response demands increase vigilance but also ability to make a directed response. It could perhaps be the case that, in di sordered or exaggerated anticipation, the former overrides the later. In the social phobic anticipating giving a speech, for instance, arousal is increased but the ability to direct that arousal into the service of functional goals is lost, resulting in a nonfunctional and therefore catastrophic seeming anxiety. Future research that systematically examines this question could significantly impact our understanding of anticipatory anxiety and, potentially, anxiety disorders.
36 Future Research T o lend further credence to the suggestions in this report future research could directly manipulate respons e demands following anticipation by including responseand noresponse conditions within the same paradigm This research could then systema tically examine effect s of response demands on emotional responding in a v ariety of measurement modalities. Related to this further examination of timing effects in anticipation, which systematically explores the anticipatory interval (e.g., length, predictability, etc), would be important in helping to determine to what extent the differences found in this study reflect temporal certainty. Finally, given the apparent parallel in startle response findings, f uture research could also examine anticipation concurrently with emotio nal imagery in a withinsubject design, to determine if parallels exist between these processes. Functional imaging will play an important role in the study of anticipation as well, as it will help to explicate the neural network of anticipatory processing Such research is expected to implicate limbic structures (e.g., the amygdala and insula) activated in other emotional paradigms in addition to structures such as the anterior cingulate cortex and hippocampus, which may serve to maintain representations o f ant icipated stimuli in the brain. It will also be important to compare neural correlates of emotion in different contexts (e.g., imagery, anticipation) within subject s to determine which brain structures are ubiquitous across emotional contexts an d which are context dependent. Research of this type will be invaluable in further clarifying the nature of neural, physiological, and psychological process during anticipation of emotional stimuli, and may provide further ground for inference as to the exact rel ationship between emotion, function, and reflex.
37 Given the characterization of anticipation in terms of response gradients based on threat and approach reactions in animals, it is likely that the work described above and future research using this paradigm would be clinically relevant. Existing research already provides preliminary indication of this; a previous study found potentiation of response in the anticipation of snake pictures in snakephobic men (Sabatinelli et al., 2001), despite the fact that these pictures are rated as low in arousal (Lang et al., 2008) and themselves inconsistently evoke potentiation of response (Bradley et al., 2001). Given the prevalence of anticipatory anxiety across DSM IV anxiety disorder diagnoses (DSMIV TR), it is likely that exaggerations in the motivational respo nse gradients characterized in this study wou ld be apparent across anxiety disorders. In focal fear disorders such as phobia, it is possible that this exaggeration may be relegated to anticipation of the feared stimulus, or it may be a more generalized anxious response to the process of anticipation itself. Research using multiple paradigms and fMRI methodologies would be useful to determine the nature of anticipati on deficits in anxiety. Such research would sig nificantly impact the understanding and treatment of anxiety disorders, and could potentially identify new biological targets for the assessment of therapeutic change. Limitations Although analysis of response to pictures themselves was not a stated aim of the current research, the lack of apparent startle response modulation during picture perception was a surprising finding. While startle inhibition was apparent for pleasant contents, no evidence of potentiation when viewing unpleasant pictures was appare nt, and the overall effect of hedonic content on reflex modulation during perception was not significant. One possible explanation for this effect is that knowing the hedonic content
38 of the upcoming picture prior to its presentation dam pens its affective e ngagement. Alternative explanations, such as experimenter error, must be ruled out through replication, however, to determine the validity of this explanation. Oth er limitations could be noted. For instance, limited number of specific contents anticipated in this study may account for any differences during pleasant and unpleasant anticipation. Replication of these results using a wider variety of cued categories would help to address this issue. Another potential limitation is the possible contamination of the ERP with eyeblink responses to the startle probes. The use of eye movement correc tion minimizes this possibility however, and reanalysis of CNV data using only trials in which startle probes were not presented during anticipation produced similar re sults as the analyses reported here.
39 LIST OF REFERENCES Althaus, M., Mulder, L. J. M., Mulder, G., Van Roon, A. M., & Minderaa, R. B. (1998 ) Influence of respiratory activity on the cardiac response pattern to mental effort. Psychophysiology, 35, 420 430 Anthony, B. J., & Putnam, L. E. (1985). Cardiac and blink reflex concomitants of attentional selectivity: A comparison of adults and young children. Psychophysiology, 22, 508516. Bradley, M. M. (2000). Emotion and motivation. IN J.T. Cacioppo & L.G. Tassinary (Eds.), Handbook of psychophysiology (2nd ed.; pp. 602 642). New York: Cambridge University Press. Bradley, M. M., Codispoti, M., Cuthbert, B. N., & Lang, P. J. (2001). Emotion and motivation I: Defensive and appetitive reactions in picture pr ocessing. Emotion, 1(3), 276 298. Bradley, M. M., Codispoti, M., & Lang, P. J. (2006). A multi process account of startle modulation during affective perception. Psychophysiology, 43, 486 497. Campbell, B. A., Wood, G., & McBride, T. (1997). Origins of orienting and defensive responses: An evolutionary perspective. In P.J. Lang, R.F. Simons, & M.T. Balaban (Eds.), Attention and orienting: Sensory and motivational processes (pp. 97135). Hillsdale, NJ: Erlbaum. Chase, W. G., Graham, F. K., & Graham, D. T (1968). Components of HR response in anticipation of reaction time and exercise tasks. Journal of Experimental Psychology, 76, 642648. Costa, V. D., Lang, P. J., Sabatinelli, D., Versace, F., & Bradley, M. M. (2010). Emotional imagery: Assessing pleasure and arousal in the brains reward circuitry. Human Brain Mapping, 31, 14461457. Davis, M. (1997). Neurobiology of fear responses: the role of the amygdala. Journal of Neuropsychiatry & Clinical Neuroscience, 9, 382 402. Davis, M. (1998). Anatomic and physiologic substrates of emotion in an animal model. Journal of Clinical Neurophysiology, 15, 378387. Kapp, B. S., Whalen, P. J., Supple, W. F., & Pascoe J. P. (1992). Amygdaloid contributions to conditioned arousal and sensory information process ing. In J.P. Aggleton (Ed.), The amygdala: Neurobiological aspects of emotion, memory, and mental dysfunction (pp. 229 254). New York: Wiley Liss.
40 Faneslow, M. S. (1994). Neural organization of the defensive behavior system responsible for fear. Psycho nomic Bulletin & Review, 1, 429438. Frijda (1987). Emotion, cognitive structure, and action tendency. Cognition and Emotion, 1, 115 143. Gatchel, R. J., & Lang, P. J. (1973). Accuracy of psychophysical judgments and physiological response amplitude. J ournal of Experimental Psychology, 98, 175 183. Graham, F. K., Putnam, L. E., & Leavitt, L. A. (1975). Lead stimulation effects on human cardiac orienting and blink reflexes. Journal of Experimental Psychology, 104, 161 169. Grillon, C., Ameli, R., Woo ds, S. W., Merikangas, K., & Davis, M. (1991). Fear potentiated startle in humans: Effects of anticipatory anxiety on the acoustc startle reflex. Psychophysiology, 28, 588 595. Headrick, M. W. & Graham, F. K. (1969). Multiplecomponenet heart rate responses conditioned under paced respiration. Journal of Experimental Psychology, 79, 486494. Kahneman, D., & Tversky, A. (2000). Choices, values, and frames. New York: Cambridge University Press. Lacey, J. I. & Lacey, B. C. (1974). On heart rate responses and behavior: a reply to Elliot. Journal of Personality & Social Psychology, 1, 1 18. Lacey, B. C., & Lacey, J. I. (1966). Twoway communication between the heart and the brain. Significance of time within the cardiac cycle. American Psychologist, 33 99113. Lacey, J. I., (1967). Somatic response patterning and stress: Some revisions of activation theory. In Appley, M.H., & Trumbull, R. (Eds.), Psychological Stress: Issues in Research. New York: AppletonCentury Crofts. Lang, P. J. (1994). The varieties of emotional experience : A meditation on James Lange theory. Psychological Review, 101, 211 221. Lang, P.J., & Hnatiow, M. (1962). Stimulus repetition and the heart rate response. Journal of Comprehensive Physiological Psychology, 55, 781 785.
41 Lang, P. J., Bradley, M. M., & Cuthbert, B. N. (1997). Motivated attention: Affect, activation and action. In P.J. Lang, R.F. Simons, & M.T. Balaban (Eds.), Attention and orienting: Sensory and motivational processes (pp. 97 135). Hillsdale, NJ: Erlbaum. Lang, P.J., Bradley, M.M., & Cuthbert, B.N. (2008). International affective picture system (IAPS): Affective ratings of pictures and instruction manual. Technical Report A 8. University of Florida, Gainesville, FL. Lang, P. J., & Davis, M. (2006). Emotio n, motivation, and the brain: reflex foundations in animal and human research. Progressive Brain Research, 156, 3 29. Loveless & Sanford (1974). Slow potential correlates of preparatory set. Biological Psychology, 1(4), 303 314. Lw, A., Lang, P.J., Smi th, J.C., & Bradley, M.M. (2008). Both predator and prey: Emotional arousal in threat and reward. Psychological Science, 19, 865873. Masterson, F.A., & Crawford, M. (1982). The defense motivation system: A theory of avoidance behavior. The Behavioral and Brain Sciences, 5, 661 696. 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, 212 225. Miller, G. A., Levin, D. N., Kozak, M. J., Cook, E. W. III, McLean, A., & Lang P. J. (1987). Individual differences in imagery and the psychophysiology of emotion. Cognition and Emotion, 1, 367390. Miller, M. W., Patrick, C. J., & Levenston, G. K. (2002). Affective imagery an d the startle response: Probing mechanisms of modulation during pleasant scenes, personal experiences, and discrete negative emotions. Psychophysiology, 39, 519529. Miller, N. E. (1951). Comments on theoretical models illustrated by the development of a theory of conflict behavior. Journal of Personality, 20, 82100. Nitschke, J.B., Larson, C. L., Smoller, M. J., Navin, S. D., Pederson, A. J. C., Ruffalo, D., et al. (2002). Startle potentiation in aversive anticipation: Evidence for state but not trait effects. Psychophysiology, 39, 254 258.
42 Phelps, E. A., OConnor, K. J., Gatenby, J. C., Gore, J. C., Grillon, C., & Davis, M. (2001). Activation of the left amygdala to a cognitive representation of fear. Nature Neuroscience, 4(4), 437 441. Poli, S., S arlo, M., Bortoletto, M., Buodo, G., Palomba, D. (2007). Stimulus Preceding Negativity and heart rate changes in anticipation of affective pictures. International Journal of Psychophysiology, 65, 32 39. Rohrbaugh, J. W., & Gaillard, A. W. K. (1983). Sensory and motor aspects of the contingent negative variation. In Gaillard, A.W.K, Ritter, W. (Eds.), Tutorials in ERP Research (pp. 269310): Endogenous Components. Elsevier, Amsterdam. Sabatinelli, D., Bradley, M. M., & Lang, P. J. (2001). Affective star tle modulation in anticipation and perception. Psychophysiology, 38, 719 722. Sabatinelli, D., Lang, P. J., Bradley, M. M., & Flaisch, T. (2006). The neural basis of narrative imagery: emotion and action. Progressive Brain Research, 156, 93 103. Sabatinelli, D., Lang, P. J., Keil, A., & Bradley, M. M. (2006). Emotional perception: Correlation of functional MRI and event related potentials. Cerebral Cortex, 17, 10851091. Simons, R. F., Ohman, A., & Lang, P. J. (1979). Anticipation and response set: Cor tical, cardiac, and electrodermal correlates. Psychophysiology, 16, 222 233. Skolnick, A,J. & Davidson, R.J. (2002). Affective modulation of eyeblink startle with reward and threat. Psychophysiology, 39, 835 850. Sokolov, E. N., & Cacioppo, J. T. (1997). Orienting and defense reflexes: Vector coding the cardiac response. In P. J. Lang, R. F. Simons, & M. T. Balaban (Eds.), Attention and orienting: Sensory and motivational processes (pp. 122) Mahwah, NJ: Erlbaum Witvliet, C. V., & Vrana, S. R. (1995). Psychophysiological responses as indices of affective dimensions. Psychophysiology, 32, 436443.
43 BIOGRAPHICAL SKETCH Christopher Sege was born in 1985 in Mountain View, California. One of two children, he grew up in Boston, Massachusetts, graduating f rom LincolnSudbury High School in 2004. He earned his bachelors degrees in Psychology and English from Boston College (BC) in 2008. While at BC, Christopher worked for 3 years in a basic emotion laboratory, studying affective conditioning processes in undergraduates. Following graduation from BC, Christopher began work at the Boston VA Medical Center in a posttraumatic stress disorder (PTSD) laboratory. While at the Boston VA, Christopher ran a study of secondorder conditioning in Vietnam veterans with PTSD, and assisted with administrative duties, data analysis, and writeup for several other projects. Christopher earned valuable clinical experience while at the VA, spending a week on base at Twenty Nine Palms Marine Base in California, and routinely co nducting neuropsychology assessments as part of the MAVERIC Study on Aging. Christopher was a second author on two published manuscripts from his time at the VA, along with numerous poster presentations and a symposium presentation. Christopher is current ly a graduate student in the Clinical & Health Psychology program at University of Florida, pursuing his doctorate in clinical psychology. He works in an emotion laboratory, the Center for the Study of Emotion and Attention, studying emotional perception a nd process in undergraduates and anxious patients. Upon receipt of his doctorate degree, Christopher hopes to complete his internship in a VA or academic clinic setting before moving back to Boston for his postdoctoral work.