Stimulus-Preceding Negativity and heart rate changes in anticipation of affective pictures

Transcription

1 International Journal of Psychophysiology 65 (2007) Stimulus-Preceding Negativity and heart rate changes in anticipation of affective pictures Silvia Poli, Michela Sarlo, Marta Bortoletto, Giulia Buodo, Daniela Palomba Department of General Psychology, University of Padova, Via Venezia 8, Padova, Italy Received 24 July 2006; received in revised form 10 February 2007; accepted 18 February 2007 Available online 24 February 2007 Abstract The aim of the present study was to investigate the affective modulation of the Stimulus-Preceding Negativity (SPN) and heart rate changes in anticipation of visual stimuli differing in emotional content. A S1 S2 task was employed with a word (S1) indicating the content of a subsequent emotional picture (S2). Both valence and arousal dimensions were manipulated by presenting positive and negative pictures, high and low in arousal. Irrespective of valence, the amplitude of the SPN resulted to be significantly larger preceding high rather than low arousal pictures, indicating that the SPN does reflect the intensity of the motivational engagement ascribed to affective stimuli. Heart rate responses showed a deceleration preceding S2, which was larger preceding high arousal stimuli in comparison with neutral stimuli. Results suggest a coherent response pattern in both cortical and peripheral measures during affective anticipation Elsevier B.V. All rights reserved. Keywords: Stimulus-Preceding Negativity; Heart rate; Anticipation; Emotion 1. Introduction Emotional anticipation is an essential regulating factor of human behavior, enabling anticipation of positive outcomes and avoidance of danger. It helps us in the choice of alternative actions and possibly helps us to survive. In psychophysiology, emotional anticipation can be investigated by means of slow cortical potentials, using a two-stimulus task, where a first stimulus (S1 or warning stimulus) indicates the occurrence of a second stimulus (S2 or imperative stimulus), which usually requires a motor response. A negative slow wave, known as Contingent Negative Variation (CNV; Walter et al., 1964), has been observed to develop in the interval between the two stimuli. When this interval is sufficiently long (2 or more seconds), the CNV consists of at least two waves: an early and a late wave (Connor and Lang, 1969; Loveless and Sanford, 1974a,b). The early wave is larger at fronto-central sites and has been associated with the properties of the first stimulus (Loveless and Sanford, 1974a,b), whereas the late wave is larger at centro-parietal sites and has been Corresponding author. Tel.: ; fax: address: silvia.poli@unipd.it (S. Poli). associated with expectancy (Loveless and Sanford, 1974a,b) and motor preparation (Rohrbaugh and Gaillard, 1983). The CNV late wave has also been observed when no motor response to the second stimulus is required. In these circumstances, however, the second stimulus has to be relevant enough to evoke a reliable anticipatory response. This is the case for emotional stimuli. When no motor response to the second stimulus is required, the CNV late wave is thought to reflect only the anticipatory processes related to the second stimulus, and it has been named Stimulus-Preceding Negativity (SPN; Brunia, 1988). Several paradigms have been employed to investigate the SPN during emotional anticipation. The SPN has been identified by presenting an emotional stimulus (usually a picture) as S2, which was signaled by a neutral stimulus, such as a tone (Amrhein et al., 2005; Klorman and Ryan, 1980; Lumsden et al., 1986; Simons et al., 1979), by employing a threat-of-shock paradigm in which a threat cue was associated with the possibility of an electric shock delivery (Baas et al., 2002; Böcker et al., 2001), and by manipulating the emotional valence of a feedback about a previously performed task (usually a time estimation task) by means of rewards or punishments (Kotani et al., 2001, 2003; Ohgami et al., 2004) /$ - see front matter 2007 Elsevier B.V. All rights reserved. doi: /j.ijpsycho

2 S. Poli et al. / International Journal of Psychophysiology 65 (2007) When visual emotional stimuli were employed as S2, only single comparisons have been assessed, i.e. positive pictures vs. neutral pictures or negative pictures vs. neutral pictures. Simons et al. (1979) showed male subjects pictures of nude females in comparison with neutral pictures and found a larger SPN prior to the emotional contents. Along with neutral pictures, Klorman and Ryan (1980) presented pictures depicting mutilated bodies, while both Amrhein et al. (2005) and Lumsden et al. (1986) presented stimuli with fear-related contents (e.g., snakes, spiders). Results generally indicated larger SPN amplitudes preceding emotional pictures, with only one study (Lumsden et al., 1986) reporting larger SPN amplitudes preceding neutral pictures. The other paradigms employed in the literature are based on the delivery of a monetary reward (Kotani et al., 2001, 2003; Ohgami et al., 2004) or a punishment consisting of noise or painful stimuli, associated with the feedback about a previously performed task (Kotani et al., 2001) or following the presentation of a symbolic threat cue (Baas et al., 2002; Böcker et al., 2001). The results of these studies consistently indicate that the SPN is larger in reward than in no-reward conditions and in anticipation of a shock/noise presentation as compared with a neutral condition. The only study using both rewards and punishments (Kotani et al., 2001) did not directly compare these conditions. The emotional value of the expected stimulus has been manipulated differently in the above-mentioned studies. Indeed, while in the feedback paradigms participants anticipate an outcome that depends on their task performance (i.e., they do not know exactly which stimulus will be presented trial by trial), in the threat-of-shock and in the picture-viewing paradigms participants are informed by the warning stimulus about the emotional content of the imperative stimulus. The employment of a picture-viewing paradigm appears to be more suitable to investigate the anticipation of stimuli differing in emotional content. Indeed, it allows manipulating a specific emotional state in anticipation. Moreover, a number of studies indicated that the presentation of pictures varying in emotional content can reliably elicit pronounced affective reactions (e.g., Bradley et al., 1993, 2001). During picture processing, several psychophysiological measures show variations that are differently modulated by the pleasantness or the intensity of the emotional content (e.g. Bradley et al., 1993). These parameters are described in terms of valence and arousal dimensions, according to the dimensional model of emotion (e.g., Lang et al., 1993) and are supposed to be the most relevant parameters describing emotional experience. More specifically, valence is the affective dimension describing the pleasantness or unpleasantness of the emotional state and is supposed to be associated with the activation of the appetitive or the defensive motivational system, whereas arousal is the affective dimension related to the intensity of motivational activation and engagement (Bradley et al., 2001). These dimensions have never been manipulated in the study of anticipation by means of slow cortical potentials. Moreover, the cortical distribution of SPN in anticipation of emotional stimuli has not been fully investigated in the literature. In most picture-viewing paradigms, indeed, few midline electrodes have been employed (Lumsden et al., 1986; Klorman and Ryan, 1980; Simons et al., 1979). When the SPN distribution was investigated using reward punishment paradigms (Kotani et al., 2001; Ohgami et al., 2004), it showed right lateralization; however the use of feedback in these studies suggests that this effect may not be specific of emotional anticipation but may rather be associated with the feedback itself. In fact, an SPN right lateralization in anticipation of a feedback has often been reported in the literature (Brunia et al., 2000; Damen and Brunia, 1985, 1987) but the same effect is not observed in anticipation of other kinds of stimuli (e.g., an instruction about a subsequent task or a probe in an arithmetic task). Heart rate (HR) changes are another measure of anticipation which have been investigated along with cortical measures in few studies. These studies usually employed a two-stimulus paradigm with fairly long interstimulus intervals (e.g., Klorman and Ryan, 1980; Simons et al., 1979). Under these conditions, a classical triphasic HR waveform, including a first deceleration (D1), a midinterval acceleration (A1) and a subsequent pronounced deceleration (D2), has frequently been observed. In reviewing some of these studies, Simons (1988) noted that the D2 component systematically occurs in a two-stimulus paradigm, during both cognitive and affective anticipations, and can develop independently of cortical negativity (Lacey and Lacey, 1974). On the other hand, in the absence of motor responses, cortical negativity clearly develops only in anticipation of affective stimuli. Simons then suggested that cardiac deceleration in the absence of cortical negativity could be an index of a merely cognitive anticipation while concurrent cardiac and cortical responses could more specifically indicate a state of emotional anticipation. In the present study, both the SPN and HR changes have been recorded. The general purpose of the study was to investigate emotional anticipation by analyzing the affective modulation of the SPN and HR changes to visual stimuli whose emotional content was signaled by congruent warning cues. An adapted form of the picture-viewing paradigm was employed by using a word (S1) naming the content of a subsequent picture (S2). Importantly, the use of such a paradigm allowed inducing a specific affective state during anticipation because participants knew which particular emotional stimulus they were going to face trial by trial. The first aim was to manipulate the valence dimension by comparing the anticipation of positive and negative emotional stimuli within the same study. Although most of the above-mentioned studies demonstrated that negative and positive emotions elicited a larger anticipatory response in comparison with neutral stimuli, it is still unknown whether there are any differences in the SPN amplitude between positive and negative contents. Moreover, this study was designed to verify whether the arousal level affected the modulation of the SPN. Since this dimension has never been systematically considered in previous studies, its contribution to the SPN modulation is still unknown. Another purpose of the study was to investigate the cortical distribution of the SPN in anticipation of positive and negative emotional stimuli in order to clarify whether the right hemispheric preponderance normally recorded for the cortical distribution of the SPN in the reward punishment paradigms might also be identified in anticipation of visual emotional stimuli. This study was also aimed at exploring the cardiac response pattern in anticipation and its relationship with cortical responses.

3 34 S. Poli et al. / International Journal of Psychophysiology 65 (2007) According to Simons (1988), HR deceleration has been hypothesized to develop in anticipation of both emotional and neutral contents and to be associated with cortical negativity only in anticipation of emotional pictures. Moreover, the paradigm used in the present study allows investigating the cardiac responses during anticipation as a function of valence and arousal. Previous studies have indeed demonstrated that HR changes differentiate emotional stimuli during picture processing, mainly reflecting the valence dimension, with unpleasant stimuli eliciting the greater deceleration (e.g. Bradley et al., 2001; Palomba et al., 1997). This effect has never been investigated during anticipation. In order to address these issues, both valence and arousal of the emotional stimuli were manipulated in the present study. Amplitude and scalp distribution of the SPN and HR changes during anticipation were examined. 2. Methods 2.1. Participants Twenty right-handed undergraduate students (10 females) aged years (M=24.85, S.D.=2.21) participated in the study after providing their written informed consent. All participants had normal or corrected-to-normal vision Stimuli and procedure Participants were seated in a comfortable armchair in an electrically shielded room at a distance of 1.20 m from the computer screen. During the entire experiment, they were monitored by a video camera. A two-stimulus task was employed. The first stimulus (S1) was a word lasting 1500 ms that signaled the content of a picture (S2) presented 4500 ms later and lasting 2000 ms. Six words (Italian translation of erotica, nature, object, people, filth, blood), balanced for length, were selected to match the content of the following six picture categories: Erotic Couples (High Arousal Positive Valence), Nature (Low Arousal Positive Valence), Injuries (High Arousal Negative Valence), Pollution (Low Arousal Negative Valence), Household Objects and Neutral People (neutral categories). The two neutral categories were employed in order to balance the two valence and arousal dimensions. A total of 72 stimuli (12 pictures per category) were selected from the International Affective Picture System (IAPS; Lang et al., 2001) and presented in two randomized sequences across subjects. 1 1 IAPS catalogue numbers (between brackets mean±standard deviation of valence and arousal ratings obtained by using the Self-Assessment Manikin 9- point scale) for the pictures used in this study are: Erotic Couples (6.8±0.3, 6.6±0.3): ; Nature (6.9±0.5, 3.8±0.9): ; Household Objects (5.0±0.1, 2.5±0.5): ; Neutral People (4.9±0.5, 3.2±0.4): ; Pollution (3.1±0.5, 4.2±0.6): ; Injuries (1.8±0.3, 6.9±0.3): The IAPS normative ratings for the pictures selected showed a significant difference between high and low arousal categories and between positive and negative categories (F(5, 66)= , p b and F(5, 66)=311.31, p b 0.001, respectively). The post-hoc comparisons between the two high arousal categories, as well as between the two low arousal categories showed no significant differences in arousal ratings. The post-hoc comparisons between the two positive categories showed no significant differences in valence ratings; conversely, a difference between the normative valence ratings of the two negative categories was found ( p b 0.001), due to an objective difficulty in finding negative low arousal material. Participants were asked to attend to the word and to the subsequent picture. No motor response to the second stimulus was required. Participants were instructed to wait for about 2 s after picture offset and then to rate their emotional experience on the dimensions of valence (unpleasant pleasant) and arousal (calm excited) using the paper-and-pencil version of the Self-Assessment Manikin (SAM; Lang et al., 2001), which consists of a 9- point rating scale for each of the above-mentioned dimensions. For this reason, intertrial period ranged randomly from 20 to 26 s. Although this evaluation required a hand movement, this was delayed enough to prevent it from affecting the SPN Physiological recordings and data analysis The electroencephalogram (EEG), electrooculogram (EOG) and electrocardiogram (ECG) were recorded using SynAmps amplifiers (NeuroScan, Inc., El Paso, TX). An Electro-Cap tin electrode helmet was used to record the EEG from 58 sites, according to the widening of the International System (Jasper, 1958), referred to linked mastoids. Eye movements and blinks were detected by 4 tin electrodes placed supra- and suborbitally to the left eye and 1 cm external to the outer canthus of each eye. Electrode impedance was kept below 10 kω. Physiological signals were sampled at 500 Hz with filter settings from DC to 100 Hz. The EEG signal was epoched into 8300 ms intervals (from 300 ms before S1 until 2000 ms after S2 onset), corrected for blinks (Semlitsch et al., 1986), re-filtered off-line with a lowpass filter at 30 Hz (12 db/oct, zero phase filter) and linear detrended in order to correct each epoch for slow DC shifts. For the SPN analysis, a smaller epoch from this interval was cut, starting 3500 ms before S2, until 800 ms after S2. This time window was chosen in order to set the baseline in the ms interval preceding picture presentation, which corresponds to the mid-interval trough. According to Simons (1988), the SPN measurements taken from the mid-interval trough are indeed more sensitive to within-subject variables and more closely related to concomitant autonomic measures (p. 252). Moreover, the use of this baseline would prevent the SPN from being affected by the processing of S1. Only trials with amplitudes not exceeding ± 60 μv were accepted for averaging. The amplitude of SPN was defined as the mean negativity between 200 ms and 0 (picture onset). The electrocardiogram (ECG) was recorded using two Ag/ AgCl electrodes placed below the right clavicle and on the

4 S. Poli et al. / International Journal of Psychophysiology 65 (2007) lower left ribcage, according to Eintoven's DII derivation. Interbeat intervals were computed, transformed off-line to heart rate (HR) values, averaged every half-second (Graham, 1978), and then analyzed as change scores with respect to a 1-s baseline preceding S1 onset. Separate analyses were performed on SPN mean amplitudes for midline (FPz, Fz, FCz, Cz, CPz, Pz, POz, Oz) and for fortytwo of the remaining electrode sites, which were averaged to obtain the following topographical regions of interest (ROIs): frontal [F1/2, F3/4, F5/6], fronto-central [FC1/2, FC3/4, FC5/6], central [C1/2, C3/4, C5/6], temporal [F7/8, T7/8, TP7/8], centroparietal [CP1/2, CP3/4, CP5/6], parietal [P1/2, P3/4, P5/6] and parieto-occipital [PO3/4, PO5/6, PO7/8]. A repeated-measures analysis of variance (ANOVA) was then conducted on midline sites with Category and Electrode as within-subjects factors, and on the remaining electrode sites with Category, Region and Hemisphere as within-subjects factors. A statistical analysis on the event-related potentials (ERPs) to the S1 onset was also performed on midline electrode sites, in order to explore whether the word processing could have affected the anticipatory processes notwithstanding the use of a baseline subsequent S1 offset. To this end, the amplitudes of the P300 and the late positive potential (LPP) components were measured, with respect to 100-ms pre-s1 baseline. The P300 amplitude was computed as the mean amplitude in the time window between 250 and 400 ms after S1 onset. Two time windows were considered for the LPP: LPP1 (mean amplitude between 400 and 700 ms after S1 onset) and LPP2 (mean amplitude between 700 and 1000 ms after S1 onset). For each component, separate repeated-measures ANOVAs were performed with Category and Electrode as within-subjects factors. In order to identify the classical cardiac components recorded in a S1 S2 task (initial deceleration, mid-interval acceleration and subsequent deceleration), the following time windows were considered for heart rate analysis: D1 (minimum HR change score between 6000 and 4000 ms before S2), A1 (maximum HR change score between 4000 and 2000 ms before S2), D2 (minimum HR change score between 2000 ms and the onset of S2). The peak deceleration during picture viewing was also considered (PDP: Picture Deceleration Peak ) and computed as the minimum value recorded during the 2 s of picture viewing. A repeated-measures ANOVA with Category and Component as within-subjects factors was then performed. Lastly, separate repeated-measures ANOVAs were also performed on mean ratings of valence and arousal for the 6 picture categories. Where appropriate, the Greenhouse Geisser correction was used to account for violation of the assumption of sphericity in the ANOVA analyses. All post-hoc comparisons reported in the results are significant according to the Tukey test ( pb0.05). In order to test the relationship between HR deceleration and SPN modulation, a Pearson correlation was performed for each category. Mean HR D2 values and mean SPN amplitudes measured over Cz site were considered. 3. Results 3.1. Self-reports Significant Category effects were found for valence (F(5, 95) = , pb0.001) and arousal (F(5, 95)=42.86, pb0.001). Posthoc comparisons indicated that positive pictures obtained the highest (most pleasant) valence ratings, followed by neutral and negative pictures. Injuries were rated as significantly more unpleasant than Pollution. Greater arousal was reported in response to Injuries, followed by Erotic Couples, and the other categories, with the lowest arousal for Household Objects. Mean values and standard deviations are reported in Table Stimulus-Preceding Negativity The analyses along the midline sites showed that the SPN amplitude was larger over frontal, central and parietal sites, being smaller over Oz (Electrode main effect: (F(7, 133)=7.85, p b 0.001, e = 0.34). The SPN was larger for high arousal categories, independent of valence (Category main effect: (F(5, 95)=19.76, p b0.001) (Fig. 1), over all electrode sites except for Fpz, where the SPN amplitude for Erotic Couples was not significantly different from that of the other categories, and for Oz, where the SPN amplitude for Erotic Couples and Injuries was not different from that recorded for Neutral People (Category Electrode interaction: F(35, 665)=4.00, p b0.001, e=0.22). The analysis on the remaining electrode sites also showed a larger SPN amplitude for high arousal categories, independent of valence (Category main effect: F(5, 95)=20.41, pb 0.001). This was true for each cortical region whereas the SPN was smaller for Nature as compared with Neutral People over fronto-central, central, centro-parietal, parietal and parieto-occipital regions (Category Region interaction: F(30, 570)=2.71, p b 0.05, e=0.20). The Region main effect showed a tendency to significance (F(6, 114)=2.89, p=0.06, e=0.38), indicating a Table 1 Self-Assessment Manikin ratings Ratings Erotic Couples Nature Household Objects Neutral People Pollution Injuries Valence 6.97±1.30 a 7.10±0.78 a 5.18±0.66 b 5.32±0.79 b 3.29±1.02 c 1.51±0.82 d (6.8±0.3) (6.9±0.5) (5.0±0.1) (4.9±0.5) (3.1±0.5) (1.8±0.3) Arousal 4.74±2.11 a 3.22±1.63 b 1.96±0.98 c 2.64±1.20 b,c,d 3.53±1.64 b,d 6.66±1.51 e (6.6±0.3) (3.8±0.9) (2.5±0.5) (3.2±0.4) (4.2±0.6) (6.9±0.3) MS±S.D. for each SAM rating. Ratings with the same letter do not differ significantly from each other, according to the Tukey test. Between brackets mean±standard deviation of valence and arousal ratings obtained from the normative ratings and already displayed in Footnote #1. It is interesting to observe that the subjective ratings provided by the participants of this study are systematically reduced as compared to the normative ratings. This is particularly marked for Erotic Couples category.

5 36 S. Poli et al. / International Journal of Psychophysiology 65 (2007) larger SPN amplitude over the centro-parietal region as compared with the temporal region ( p = 0.058). No differences between right and left hemispheres were identified ERPs to S1 onset The analyses on P300 amplitude showed no significant difference between categories. The Category main effect was significant for the LPP1 component (F(5, 95)=4.00, pb0.01), indicating a greater positivity for Erotic Couples as compared with Nature, Pollution and Neutral People, and for the LPP2 Fig. 2. Heart rate changes during the 6 s preceding and the 2 s following picture onset, for the six categories. component (F(5, 95) = 3.49, p b 0.01), indicating a greater positivity for Erotic Couples and Injuries as compared with Neutral People. For each component, the Electrode main effect was also significant (F(7, 133) =4.32, pb 0.05, e =0.22; F(7, 133) =21.16, pb0.001, e=0.32 and F(7, 133)=17.59, pb0.001, e=0.33, for P300, LPP1 and LPP2, respectively) showing that the components were larger over the central and parietal sites as compared with the frontal sites Heart rate A typical triphasic response consisting of a brief initial deceleration followed by a slight HR increase and a secondary deceleration was identified, particularly for high arousal pictures (Fig. 2). As expected, each component was significantly different from the others (Component main effect: F(3, 57)=74.04, pb 0.001, e=0.68). A Category main effect was found (F(5, 95)= 3.38, p b 0.01), indicating a difference between high arousal categories and Household Objects category ( p b 0.05) and the Category Component interaction (F(15, 285) = 3.57, p b 0.001, e=0.43) clarified that the D2 component was significantly larger during the anticipation of Erotic Couples as compared with the neutral categories ( pb0.01). As for the PDP component, both Erotic Couples and Injuries showed larger deceleration than emotional low arousal and neutral categories ( pb0.01). No other difference was found to be significant Correlations between HR D2 component and SPN amplitude The correlation between mean HR D2 values and mean SPN amplitudes over Cz site was not significant for any category (Erotic Couples: r=0.15, p=0.52; Nature: r=0.08, p=0.75; Household Objects: r= 0.34, p=0.14; Neutral People: r= 0.09, p=0.71; Pollution: r=0.16, p=0.49; Injuries: r=0.08, p=0.74). 4. Discussion Fig. 1. For purposes of exemplification, the figure shows the SPN recorded only over the Fz, Cz and Pz sites. For the sake of clarity, waveforms were re-filtered at 10 Hz. The present study was designed to investigate the affective modulation of the SPN and HR changes in anticipation of emotional stimuli. A set of six different picture categories drawn

6 S. Poli et al. / International Journal of Psychophysiology 65 (2007) from the International Affective Picture System was presented. Both valence and arousal dimensions were manipulated. Consistent with previous results, the SPN amplitude was found to be larger in anticipation of emotional rather than neutral pictures. However, this effect was only significant for high arousal pictures. In fact, the SPN amplitude during the anticipation of low arousal contents was very small and did not differ significantly from that recorded during the anticipation of neutral categories. This result was highly significant at all cortical regions and was not affected by the processing of S1. Indeed, a significant difference between categories was only found in the late phases of the processing of S1, indicating a greater allocation of attentional resources, reflected in a greater cortical positivity, for high arousal categories as compared with neutral or low arousal categories. If this effect had influenced the anticipatory processes, a greater positivity, instead of a greater negativity, would also have been expected during the interstimulus interval. This was not the case; therefore it may be argued that the word processing did not affect the anticipatory processes. Similarly, the results found for the SPN were not affected by the significant difference in self-reports between the two high arousal categories. In fact, High Arousal Negative pictures were rated by participants as more arousing than High Arousal Positive pictures, although their normative a priori ratings were not significantly different. This result suggests that emotional anticipation and expectancy affected subjective experience, thus leading to some discrepancies from the normative ratings. In particular, unlike negative stimuli, positive stimuli modulate subjective arousal ratings in that subjective arousal reported after picture viewing is reduced compared to normative ratings. Although the mechanism involved in mediating this effect requires further investigation, it did not influence cortical anticipation processes. Positive and negative high arousal stimuli, particularly those employed in this study, are especially relevant for the survival of individuals and a number of previous studies have shown that they are able to naturally activate the appetitive and defensive motivational systems respectively, eliciting the most pronounced reactivity (Bradley et al., 2001). In fact, the largest skin conductance changes (Bradley et al., 2001; Codispoti et al., 2001; Cuthbert et al., 1996; Lang et al., 1993, 1998)andthelargest and most sustained late positive potential (LPP) (Cuthbert et al., 2000; Schupp et al., 2000, 2003, 2004) were obtained during the viewing of high arousal pictures. Thus, the SPN seems not to reflect a generic affective anticipation but rather the motivational engagement ascribed to affective stimulus anticipation. According to Bradley et al. (2001), not all psychophysiological measures may discriminate defensive and appetitive disposition, and not all measures vary monotonically with motive intensity. The present data provide information about affective picture pre-processing and suggest that both cortical slow waves the SPN and the LPP are primarily modulated by motivational intensity, whereas the valence of emotional stimuli is confirmed to be better discriminated by changes in other measures (e.g. facial electromyographic responses; Bradley et al., 1993, 2001). The same mechanism involved in the activation of the brain motivational circuits during the processing of emotionally arousing stimuli may then be assumed to operate also in anticipation of the same material. As mentioned above, the modulation of the SPN seems to be very similar to that of the LPP. This positive slow wave, which is recorded a few hundred milliseconds after stimulus onset and which may be sustained for several seconds, clearly discriminates affective from neutral material. It is especially larger in response to high arousal emotional stimuli, reflecting a sustained attention set to these contents (Schupp et al., 2000, 2004). It has been assumed that cortical negativity, as measured by slow event-related potentials, is an indicator of an enhanced excitability of cortical networks, reflecting a preparatory state for cerebral processing, whereas cortical positivity, as measured by slow event-related potentials, would indicate a decrease in this excitability (or a cortical inhibition), reflecting a greater allocation of perceptual processing resources (Birbaumer et al., 1990; Rockstroh et al., 1989; Schupp et al., 1994). Thus, it might be suggested that an increase in LPP positivity is an index of cortical processing of motivationally significant material, whereas an increase in the pre-stimulus SPN negativity, related to the same material, would be an index of cortical expectancy and preparation, which is enhanced for the same evolutionary reasons. Furthermore, the hypothesis that comparable brain regions are involved in both stimulus anticipation and perception was proposed by William James (1892). This hypothesis has been tested in recent years by several functional magnetic resonance imaging (fmri) studies, but with controversial results. In particular, the anticipation of a visual emotional stimulus was investigated by Simmons et al. (2004) and Ueda et al. (2003) with a cue that predicted whether a positive or a negative picture would follow, as was done in the present study. The results of these studies suggest the presence of a remarkable overlap between networks involved in emotional anticipation and perception, including insula, parahippocampal gyrus, prefrontal cortex, amygdala, anterior cingulate cortex and the occipital cortex. However, a recent study employing a cue that did not provide information about stimulus valence did not support these results and rather suggested the existence of partially dissociable networks, including anterior cingulate cortex, cingulate motor area, and parieto-occipital regions during anticipation processes and amygdala, insula, prefrontal cortex, cerebellum and occipitotemporal areas during perception (Bermpohl et al., 2006). The question is still open and undoubtedly represents a promising area for future research. The present study was aimed at investigating differences in SPN scalp distribution during anticipation of positive and negative emotional stimuli. The results showed that the SPN preceding emotional stimuli are bilaterally distributed. The right hemispheric preponderance usually shown for the SPN cortical distribution in the reward punishment paradigms was not observed using the present paradigm. This result challenges the view that the SPN right hemispheric preponderance recorded in time estimation tasks would be determined by an involvement of the right hemisphere in emotional processing (Kotani et al., 2001, 2003) and indicates that it would rather be associated to

7 38 S. Poli et al. / International Journal of Psychophysiology 65 (2007) some more specific processes probably related to feedback processing (see also Brunia, 1988). This result also supports the view that the SPN is not a unitary phenomenon but a class of anticipatory responses (Van Boxtel and Böcker, 2004) and contributes to define the features of the emotional SPN. Heart rate responses in anticipation of emotional stimuli were also examined in this study. The results indicate that, throughout the whole interstimulus interval, larger heart rate deceleration is produced in response to high arousal as compared with neutral categories, whereas during anticipation (D2 component) the greatest deceleration is found for High Arousal Positive stimuli. Therefore, heart rate changes during anticipation are sensitive to the arousal manipulation. Furthermore, the results largely support the view that emotional anticipation of high arousal pictures is associated with both cortical negativity and heart rate deceleration (Simons, 1988). This pattern is also coherent for neutral stimuli, which are preceded by cardiac deceleration even in the absence of cortical negativity. Low arousal stimuli revealed a similar pattern to that found for neutral stimuli. Although this pattern is not supported by the results of Pearson correlations between HR deceleration and SPN amplitudes, a lack of significant results is often reported in literature when inter-individual correlations across subject means are obtained (Lazarus et al., 1963). Intraindividual correlations across trials might have highlighted a significant relationship between HR and SPN measures. However, it was not possible to use this approach, since event-related potentials are, by definition, obtained by average procedures. A limitation of the study might be the small number of trials per category. However, this is not uncommon when emotional processing is the target of investigation, because, in this case, it is particularly important to avoid feelings of boredom and fatigue, which could be present with long experimental sessions. In conclusion, the present study has shown that the SPN is a clear indicator of emotional anticipation reflecting the motivational engagement involved in the task. In contrast with results provided by feedback paradigms, an absence of hemispheric differentiation in anticipation of emotional stimuli has been highlighted. Finally, heart rate deceleration during anticipation has been identified, being larger for high arousal categories than for low arousal and neutral stimuli, showing that heart rate responses are a sensitive index of emotional anticipation. References Amrhein, C., Pauli, P., Dengler, W., Wiedemann, G., Covariation bias and its physiological correlates in panic disorder patients. J. Anxiety Disord. 19, Baas, J.M.P., Kenemans, J.L., Böcker, K.B.E., Mijnhardt, F., Verbaten, M.N., Threat-induced cortical processing and startle potentiation. NeuroReport 13, Bermpohl, F., Pascual-Leone, A., Amedi, A., Merabet, L.B., Fregni, F., Gaab, N., Alsop, D., Schlaug, G., Northoff, G., Dissociable networks for the expectancy and perception of emotional stimuli in the human brain. NeuroImage 30, Birbaumer, N., Elbert, T., Canavan, A., Rockstroh, B., Slow potentials of the cerebral cortex and behavior. Physiol. Rev. 70, Böcker, K.B.E., Baas, J.M.P., Kenemans, J.L., Verbaten, M.N., Stimuluspreceding negativity induced by fear: a manifestation of affective anticipation. Int. J. Psychophysiol. 43, Bradley, M.M., Greenwald, M.K., Hamm, A.O., Affective picture processing. In: Birbaumer, N., Öhman, A. (Eds.), The Structure of Emotion. Hogrefe & Huber, Toronto, pp Bradley, M.M., Codispoti, M., Cuthbert, B.N., Lang, P.J., Emotion and motivation I: defensive and appetitive reactions in picture processing. Emotion 1, Brunia, C.H.M., Movement and stimulus preceding negativity. Biol. Psychol. 26, Brunia, C.H.M., De Jong, B.M., van den Berg-Lenssen, M.M.C., Paans, A.M.J., Visual feedback about time estimation is related to a right hemisphere activation measured by PET. Exp. Brain Res. 130, Codispoti, M., Bradley, M.M., Lang, P.J., Affective reactions to briefly presented pictures. Psychophysiology 38, Connor, W.H., Lang, P.J., Cortical slow wave and cardiac rate responses in stimulus orientation and reaction time conditions. J. Exp. Psychol. 82, Cuthbert, B.N., Bradley, M.M., Lang, P.J., Probing picture perception: attention and emotion. Psychophysiology 33, Cuthbert, B.N., Schupp, H.T., Bradley, M.M., Birbaumer, N., Lang, P.J., Brain potentials in affective picture processing: covariation with autonomic arousal and affective report. Biol. Psychol. 52, Damen, E.J.P., Brunia, C.H.M., Slow brain potentials related to movement and visual feedback in a response timing task. Biol. Psychol. 20, 195. Damen, E.J.P., Brunia, C.H.M., Changes in heart rate and slow brain potentials related to motor preparation and stimulus anticipation in a time estimation task. Psychophysiology 24, Graham, F.K., Constraints on measuring heart rate and period sequentially through real and cardiac time. Psychophysiology 15, James, W., Text-Book of Psychology. Macmillan, London. Jasper, H.H., The ten-twenty electrode system of the international federation. Electroencephalogr. Clin. Neurophysiol. 10, Klorman, R., Ryan, R.M., Heart rate, contingent negative variation, and evoked potentials during anticipation of affective stimulation. Psychophysiology 17, Kotani, Y., Hiraku, S., Suda, K., Aihara, Y., Effect of positive and negative emotion on stimulus-preceding negativity prior to feedback stimuli. Psychophysiology 38, Kotani, Y., Kishida, S., Hiraku, S., Suda, K., Ishii, M., Aiharab, Y., Effects of information and reward on stimulus-preceding negativity prior to feedback stimuli. Psychophysiology 40, Lacey, B.C., Lacey, J.I., Studies of heart rate and other bodily processes in sensorimotor behavior. In: Obrist, P.A. (Ed.), Cardiovascular Psychophysiology: Current Issues in Response Mechanisms, Biofeedback, and Methodology. Aldine, Chicago, pp Lang, P.J., Bradley, M.M., Cuthbert, B.N., Emotion, motivation, and anxiety: brain mechanisms and psychophysiology. Biol. Psychiatry 44, Lang, P.J., Bradley, M.M., Cuthbert, B.N., International affective picture system (IAPS): instruction manual and affective ratings. Technical Report A-5. The Center for Research in Psychophysiology, University of Florida, Gainsville, FL. Lang, P.J., Greenwald, M.K., Bradley, M.M., Hamm, A.O., Looking at pictures: affective, facial, visceral, and behavioural reactions. Psychophysiology 30, Lazarus, R.S., Speisman, J.C., Mordkoff, A.M., The relationship between autonomic indicators of psychological stress: heart rate and skin conductance. Psychosom. Med. 25, Loveless, N.E., Sanford, A.J., 1974a. Effects of age on the contingent negative variation and preparatory set in a reaction time task. J. Gerontol. 29, Loveless, N.E., Sanford, A.J., 1974b. Slow potential correlates of preparatory set. Biol. Psychol. 1, Lumsden, J., Howard, R.C., Fenton, G.W., The contingent negative variation (CNV) to fear-related stimuli in acquisition and extinction. Int. J. Psychophysiol. 3, Ohgami, Y., Kotani, Y., Hiraku, S., Aihara, Y., Ishii, M., Effects of reward and stimulus modality on stimulus-preceding negativity. Psychophysiology 41, Palomba, D., Angrilli, A., Mini, A., Visual evoked potentials, heart rate responses and memory to emotional pictorial stimuli. Int. J. Psychophysiol. 27,

8 S. Poli et al. / International Journal of Psychophysiology 65 (2007) Rockstroh, B., Elbert, T., Canavan, A., Lutzenberger, W., Birbaumer, N., Slow Brain Potentials and Behaviour, 2nd edn. Urban and Schwarzenberg, Munich. Rohrbaugh, J.W., Gaillard, A.W.K., Sensory and motor aspects of the contingent negative variation. In: Gaillard, A.W.K., Ritter, W. (Eds.), Tutorials in ERP Research: Endogenous Components. Elsevier, Amsterdam, pp Schupp, H.T., Lutzenberger, W., Rau, H., Birbaumer, N., Positive shifts of event-related potentials: a state of cortical disfacilitation as reflected by the startle reflex probe. Electroencephalogr. Clin. Neurophysiol. 90, Schupp, H.T., Cuthbert, B.N., Bradley, M.M., Cacioppo, J.T., Ito, T., Lang, P.J., Affective picture processing: the late positive potential is modulated by motivational relevance. Psychophysiology 37, Schupp, H.T., Junghöfer, M., Weike, A.I., Hamm, A.O., Emotional facilitation of sensory processing in the visual cortex. Psychol. Sci. 14, Schupp, H.T., Cuthbert, B.N., Bradley, M.M., Hillman, C.H., Hamm, A.O., Lang, P.J., Brain processes in emotional perception: motivated attention. Cogn. Emot. 18, Semlitsch, H.V., Anderer, P., Schuster, P., Presslich, O., A solution for reliable and valid reduction of ocular artifacts, applied to the P300. Psychophysiology 23, Simons, R.F., Event-related slow brain potentials: a perspective from ANS psychophysiology. In: Ackles, P.K., Jennings, J.R., Coles, M.G.H. (Eds.), Advances in Psychophysiology, vol. 3. Elsevier, Amsterdam, pp Simons, R.F., Öhman, A., Lang, P.J., Anticipation and response set: cortical, cardiac, and electrodermal correlates. Psychophysiology 16, Simmons, A., Matthews, S.C., Stein, M.B., Paulus, M.P., Anticipation of emotionally aversive visual stimuli activates right insula. NeuroReport 15, Ueda, K., Okamoto, Y., Okada, G., Yamashita, H., Hori, T., Yamawaki, S., Brain activity during expectancy of emotional stimuli: an fmri study. NeuroReport 14, Van Boxtel, G.J.M., Böcker, K.B.E., Cortical measures of anticipation. J. Psychophysiol. 18, Walter, W.G., Cooper, R., Aldridge, V.J., McCallum, W.C., Winter, A.L., Contingent negative variation: an electric sign of sensorimotor association and expectancy in the human brain. Nature 203,