Visual evoked potentials, heart rate responses and memory to emotional pictorial stimuli

https://doi.org/10.1016/S0167-8760(97)00751-4Get rights and content

Abstract

Although the effects of emotional stimuli on event-related cortical potentials, heart rate, and memory have been extensively studied, the association of these variables in a single study has been neglected. The influence of pleasant, unpleasant, and neutral photographic slides on visual evoked potentials (VEPs), heart rate responses, and free recall, was investigated in 20 normal subjects. VEPs were recorded from Cz and Pz locations, and analyses were performed on both amplitudes and latencies of identifiable endogenous peaks (P2, N2 and P3), and mean amplitude in the 100–200-ms, 400–600-ms, and 600–900-ms latency ranges. An emotional effect was present on VEPs starting from about 282 ms on, as revealed by the N2, P3, and late components. Both pleasant and unpleasant slides yielded larger cortical positivity as compared to neutral ones. Peak latencies did not show any emotional effect. Heart rate data showed a deceleratory response that was larger to unpleasant slides. Free recall of the projected slides showed a better performance for emotional slides compared to neutral ones. VEPs and memory data showed the same pattern: both pleasant and unpleasant slides induced larger positivity in the event-related potentials and were better remembered than neutral slides. Positive correlations were found between the late negative VEPs component (600–900 ms), recorded from Cz, and heart rate deceleration (r=0.62), and between P3 (at Pz location) and the number of remembered slides (r=0.53). © 1997 Elsevier Science B.V.

Introduction

The investigation of event-related potentials to emotional visual stimuli has generally shown higher cortical positivity in response to emotional material as compared to neutral material. Such pattern has been viewed as indicative of deeper processing of the emotional information. However a number of problems characterized previous studies on this topic, as the fact that the visual stimuli covered a restricted semantic category and/or were not standardized.

Lifshitz (1966)and Begleiter et al. (1967)first reported the influence of emotional visual stimuli on event-related cortical responses. Later on, Radilovà and coworkers (Radilovà, 1982; Radilovà et al., 1983, Radilovà et al., 1984) studied visual evoked potentials (VEPs) while subjects watched emotional pictures. In the first study (Radilovà, 1982), unpleasant visual stimuli, tachistoscopically presented, produced more positive P3 waves as compared to neutral stimuli. Two other experiments (Radilovà et al., 1983, Radilovà et al., 1984) showed that P3 was more positive for erotic (sexual slides) than non-erotic (landscapes, flowers, etc.) slides. These studies yielded the conclusion that the emotional impact of visual stimuli increased the amplitude of the P3 independent of the pleasant or aversive characteristics of the induced emotion (Radilovà, 1989). Nevertheless, the comparison between negative and positive emotions was not included in a single study by these authors.

Data from Johnston and coworkers (Johnston et al., 1986; Johnston and Wang, 1991) were consistent with the above-cited results. In their first study (Johnston et al., 1986), subjects watched sexual (opposite and same sex models), pleasant (babies), neutral (ordinary people) and unpleasant (dermatological pictures) slides. Female model pictures induced larger P3 and P4 (max. at 540 ms) than both pleasant and unpleasant slides, which, in turn, yielded larger cortical positivity compared to neutral slides. Johnston and Wang (1991), using the same visual stimuli, additionally supported the finding that P3 was larger for emotional than for neutral material.

Those studies that investigated lateralized visual evoked potentials to emotional pictures also yielded similar results (Vanderploeg et al., 1987; Laurian et al., 1991): larger P3 to emotional material compared to neutral one; data on laterality were not so clearcut.

In order to obtain a replicable data set, in our laboratory (Mini et al., 1996), visual evoked potentials were investigated to standardized emotional slides. Slides were selected from the International Affective Picture System (IAPS: Center for the Study of Emotion and Attention, 1995), a cross-cultural validated set that includes a wide range of semantic categories varying along the valence dimension (pleasant, unpleasant, and neutral). Results showed more cortical positivity to emotional than neutral slides, in both the analyzed latency ranges: 300–400 ms and 400–500 ms.

The brief review of studies presented here highlights a quite general agreement about a larger positivity to emotional pictorial stimuli as compared to neutral ones. This effect was mainly showed in the P3 and later components.

In general, cardiovascular changes reflect metabolic adjustment to environmental demands. Therefore, heart rate modifications should be expected during almost every emotional state. Although heart rate responses have been largely investigated during emotional induction, the data available up to now are controversial (reviews on this argument are in Wagner, 1989; Levenson, 1992; Zajonc and McIntosh, 1992; Cacioppo et al., 1993).

Either physiological or psychophysiological factors account for some divergent findings. First, the heart is doubly innervated by sympathetic and parasympathetic nervous systems, which can vary reciprocally, independently or coactively in front of different stimuli (Berntson et al., 1991, Berntson et al., 1993). Second, the different experimental methods used to induce emotions, particularly imagery vs perception, may be responsible for different effects (Lang et al., 1993; Palomba, 1993; Philippot, 1993).

Cardiac changes during emotional perception have been explained in the light of Laceys' model of information processing (Lacey and Lacey, 1970). The model indicates that stimulus intake (outward directed attention) produces heart rate deceleration associated with cortical activation. On the other hand, stimulus rejection (inwards directed processing) is associated with heart rate acceleration and cortical inhibition. According to this model, heart rate increases occurring during intense emotional states should be associated to cortical deactivation and inhibited information processing, as predicted by the defense reaction (DR). Milder emotional stimulations are likely to produce an orienting reaction (OR); under these conditions a heart rate reduction should be observed (Sokolov, 1963; Graham and Clifton, 1966; Turpin, 1986).

Indeed, when subjects watch visual stimuli varying along the valence dimension, cardiac deceleration prevails over acceleratory changes. Moreover, larger heart rate decelerations occur following unpleasant stimuli as compared to pleasant ones (Winton et al., 1984; Vrana et al., 1988; Patrick et al., 1993; Palomba et al., 1994). When highly-fearful subjects are presented with unpleasant, phobic material, cardiac acceleration (DR) replaces the usual orienting cardiac deceleration (Klorman et al., 1977; Lumley and Melamed, 1992). A completely different pattern is observed during imagery or social induction tasks: cardiac acceleration is often present, and the amount of heart rate increase seems to be influenced by the intensity of emotion, and to a lesser extent, by its valence (Vrana et al., 1988; Gollnisch and Averill, 1993; vanOyen Witvliet and Vrana, 1995).

In summary, cardiac responses to emotional stimuli reflect both cognitive (mostly attentional) processes and the modality of affective processing. Given the complex interaction between central and cardiovascular processes during emotional states, it might be of particular interest to study their association within a single study.

Cortical processing of emotional material can be studied through cortical indexes (as the above considered VEPs), but also indirectly by investigating emotional influences on cognitive functions such as memory.

During the first part of this century it was a common belief that pleasant events were remembered more than unpleasant ones (Matlin and Stang, 1978). However, this effect has mostly relied on speculative concepts and questionable evidence (see Singer and Salovey, 1988). In particular, the mechanisms underlying the effect were almost unclear.

Later on, several models have been proposed to explain relations between emotions and memory. Widely accepted in cognitive psychology is the assumption that retrieval is largely dependent upon mood states and context cues available at encoding and recall (Bower, 1981; see also: Blaney, 1986; Singer and Salovey, 1988, for reviews on this topic). Those studies that assessed the impact of emotional material on memory performance independent of mood and context influences, indicate that emotional events are remembered better than neutral, ordinary ones. Bradley and Baddeley (1990), and Paul and Whissell (1992), investigating memory for emotional words, showed that they were remembered more than neutral ones while pleasant and unpleasant words were not differentially recalled. These general findings, however, are still questionable due to methodological differences which make it difficult to compare results obtained in different studies (see Christianson, 1992for a review). Few studies, indeed, used standardized emotional material to assess affective memory. Bradley et al. (1992)investigated the free-recall of a large number of emotional slides, varying along the dimensions of valence and arousal, selected from the International Affective Picture System (Center for the Study of Emotion and Attention, 1995). In immediate free recall test, both pleasant and unpleasant slides resulted in better memory performance as compared to neutral ones. Long-term recall, as assessed one year later, yielded the same effect. A slight advantage for pleasant slides, found in immediate recall, disappeared in long-term recall.

The purpose of the present experiment was to investigate changes in visual evoked potentials and heart rate during the viewing of standardized emotional photographic pictures, and to assess the immediate free recall for the same slides.

According to the literature, both pleasant and unpleasant emotional stimuli were hypothesized to yield larger cortical positivity compared to neutral ones. This effect was expected to occur in VEP endogenous components in a latency range between 200 and 900 ms. Moreover, on the basis of our previous study (Mini et al., 1996), in which Fz and Cz locations yielded similar scalp amplitudes, in the present study visual evoked potentials were obtained only from two recording sites (Cz and Pz).

Heart rate was expected to show larger decelerations following unpleasant stimuli as compared to pleasant ones, as already reported in previous studies (Winton et al., 1984; Vrana et al., 1988; Patrick et al., 1993; Palomba et al., 1994). Moreover, according to Laceys' model (Lacey and Lacey, 1970) and further demonstrations of positive relationships between heart rate deceleratory changes and cortical negativity (Hatfield et al., 1987; Putnam, 1990; Rockstroh and Elbert, 1990; Walker and Sandman, 1982), it was expected a positive relationship between visual evoked potentials negativity and heart rate decelerations in response to emotional material.

Recall for emotional stimuli was assessed as a subjective measure of emotional processing. In line with Bradley et al. (1992), both pleasant and unpleasant slides were expected to yield a better memory performance as compared to neutral ones. Furthermore, the influence of emotional stimuli on encoding can be fruitfully tested by recording event-related potentials during stimulus presentation and by comparing the direction of changes in both event-related potentials and memory performance to pleasant, unpleasant, and neutral material. Indeed, in the literature, the P3 component has been related to the updating of representations in working memory, by the context-updating model (Donchin and Coles, 1988). Updating of memory representation is assumed to facilitate subsequent recall. Some studies have shown that event-related potentials to subsequently recalled or recognized stimuli were more positive than those to forgotten ones. Moreover, the P3 (and/or later positive components) varied as a function of subsequent memory performance (e.g. Neville et al., 1986; Paller et al., 1987; Paller and Kutas, 1992; Sanquist et al., 1980). Therefore, in the present study it was predicted that those slides that evoked more positive endogenous waves would also be better remembered.

Section snippets

Subjects

Subjects were twenty psychology students (17 female and 3 male; mean age: 25.3 years) participating as a course requirement. All subjects had normal vision and were not suffering from any neurological disorder. All subjects gave informed consent to the participation in the study.

Stimuli

Sixty emotional slides

Visual evoked potentials

The grand-averaged VEP waveforms for the three emotional conditions are shown in Fig. 1.

Mean amplitude of each selected latency range, along with peak amplitudes and latencies are reported in Table 1.

All statistical analyses, including 100–200 ms, 400–600 ms, and 600–900 ms time windows, and P2, N2 and P3 peak amplitudes, showed a highly significant location effect, with Pz always more positive than Cz (all P<0.0007). There were no latency differences between the two locations. The interaction

Event-related potentials to emotional pictorial stimuli and memory performance

Processing visual emotional material produced significant changes in both the central nervous system (CNS) and cardiac activity.

In agreement with our expectation no significant effects due to physical differences between stimuli were detected in the first analyzed latency range, the 100–200 ms window. The earliest emotional effect was found about 282 ms after stimulus onset. N2, P3, and later components (400–600 ms and 600–900 ms latency ranges) showed that emotional slides, both pleasant and

Acknowledgements

The Authors wish to thank Stefano Bravi for his help in data collection, Prof. L. Stegagno and Terry W. Njoroge for the revision of the manuscript.

References (66)

  • Begleiter, H., Gross, M.M. and Kissin, B. (1967) Evoked cortical responses to affective visual stimuli....
  • Berntson, G.G., Cacioppo, J.T. and Quigley, K.S. (1991) Autonomic determinism: the modes of autonomic control, the...
  • Berntson, G.G., Cacioppo, J.T. and Quigley, K.S. (1993) Cardiac psychophysiology and autonomic space in humans:...
  • Birbaumer, N. and Elbert, T. (1988) P3: byproduct of a byproduct. Behav. Brain Sci., 11:...
  • Birbaumer, N., Elbert, T., Canavan, A. and Rockstroh, B. (1990) Slow potentials of the cerebral cortex and behavior....
  • Birbaumer, N., Lutzenberger, W., Elbert, T. and Trevorrow, T. (1994) Threshold variations in cortical cell assemblies...
  • Blaney, P.H. (1986) Affect and memory. Psychol. Bull., 99:...
  • Bower, G.H. (1981) Mood and memory. Am. Psychol., 36:...
  • Bradley, B.P. and Baddeley, A.D. (1990) Emotional factors in forgetting. Psychol. Med., 20:...
  • Bradley, M.M., Greenwald, M.K., Petry, M. and Lang, P.J. (1992) Remembering pictures: Pleasure and arousal in memory....
  • Cacioppo, J.T., Klein, D.J., Berntson, G.G. and Hatfield, E. (1993) The psychophysiology of emotion. In: M. Lewis and...
  • Center for the Study of Emotion and Attention — CSEA-NIMH (1995) The International Affective Picture System (IAPS;...
  • Christianson, S.-Å. (1992) Remembering emotional events: potential mechanisms. In: S.-Å. Christianson (Ed.), The...
  • Donchin, E. and Coles, M.G.H. (1988) Is the P300 component a manifestation of context updating? Behav. Brain Sci., 11:...
  • Gollnisch, G. and Averill, J.R. (1993) Emotional imagery: Strategies and correlates. Cognit. Emot., 7:...
  • Graham, F.K. (1980) Representing cardiac activity in relation to time. In: I. Martin and P.H. Venables (Eds.),...
  • Graham, F.K. and Clifton, R.K. (1966) Heart rate change as component of the orienting response. Psychol. Bull., 65:...
  • Greenhouse, S. and Geisser, S. (1958) An extension of Box's results on the use of the F distribution in multivariate...
  • Greenwald, M.K., Cook, E.W. III and Lang, P.J. (1989) Affective judgment and psychophysiological response: dimensional...
  • Hatfield, B.D., Landers, D.M. and Ray, W.J. (1987) Cardiovascular-CNS interactions during a self-paced, intentional...
  • Hitchcock, J.M. and Davis, M. (1991) Efferent pathway of the amygdala involved in conditioned fear as measured with the...
  • Iwata, J., Chida, K. and LeDoux, J. (1987) Cardiovascular responses elicited by stimulation of neurons in the central...
  • Jasper, H.H. (1958) The ten-twenty electrode system of the International Federation. EEG Clin. Neurophysiol., 10:...
  • Johnston, V.S., Miller, D.R. and Burleson, M.H. (1986) Multiple P3s to emotional stimuli and their theoretical...
  • Johnston, V.S. and Wang, X.T. (1991) The relationship between menstrual phase and the P3 component of ERPs....
  • Klorman, R., Weissberg, R.P. and Wiesenfeld, A.R. (1977) Individual differences in fear and autonomic reactions to...
  • Lacey, J.I. and Lacey, B.C. (1970) Some autonomic-central nervous system interrelationships. In: P. Black (Ed.),...
  • Lang, P.J., Bradley, M.M. and Cuthbert, B.N. (1990) Emotion, attention, and the startle reflex. Psychol. Rev., 97,...
  • Lang, P.J., Greenwald, M.K., Bradley, M.M. and Hamm, A.O. (1993). Looking at pictures: affective, facial, visceral, and...
  • Laurian, S., Bader, M., Lanares, J. and Oros, L. (1991) Topography of event-related potentials elicited by visual...
  • LeDoux, J.E. (1995) Emotion: clues from the brain. Annu. Rev. Psychol., 46:...
  • Levenson, R.W. (1992) Autonomic nervous system patterning in emotion. Psychol. Sci., 3:...
  • Lifshitz, K. (1966) The averaged evoked cortical response to complex visual stimuli. Psychophysiology, 3:...
  • Cited by (302)

    View all citing articles on Scopus
    View full text