0 1992 Gordon and Breach Science Publishers S.A.
Intern. .I Neuroscience, . 1992, Vol. 63, p. 5-16
Reprints available directly from the publisher Photocopying permitted by license only
Printed in the United States of America
TOPOGRAPHIC BRAIN MAPPING OF EMOTIONRELATED HEMISPHERE ASYMMETRIES RUPERT ROSCHMANN AND WERNER WITTLING* Lehrstuhl fur Biopsychologie und Klinische Psychologie, Katholische Universitat Eichstatt, 0-8078 Eichstatt, FRG
Int J Neurosci Downloaded from informahealthcare.com by Nyu Medical Center on 02/08/15 For personal use only.
(Received October 5 . 1990)
The study used topographic brain mapping of visual evoked potentials to investigate emotion-related hemisphere asymmetries. The stimulus material consisted of color photographs of human faces, grouped into two emotion-related categories: normal faces (neutral stimuli) and faces deformed by dermatological diseases (emotional stimuli). The pictures were presented tachistoscopically to 20 adult right-handed subjects. Brain activity was recorded by 30 EEG electrodes with linked ears as reference. The waveforms were averaged separately with respect to each of the two stimulus conditions. Statistical analysis by means of significance probability mapping revealed significant differences between stimulus conditions for two periods of time, indicating right hemisphere superiority in emotion-related processing. The results are discussed in terms of a 2-stage-model of emotional processing in the cerebral hemispheres. Keywords: topographic brain mapping, evoked potentials, emotional hemisphere asymmetry, emotion, lateralization.
During the last few years there has been a rapid growth of interest in emotion-related hemisphere asymmetries. The majority of clinical and experimental studies has supported the hypothesis that the right cerebral hemisphere plays a dominant role in the processing of emotional stimuli and the regulation of emotional responses (Gainotti, 1983; Silberman and Weingartner, 1986). An alternative hypothesis that the right hemisphere is specialized for negative emotions and the left hemisphere for positive emotions also found some support (Davidson, 1984). Among the various research strategies used to study the phenomenon one important approach involves the recording of brain electric activity during emotion-related processing. Most of these studies dealt with EEG measures to determine relative hemispheric activation (Tucker, Stenslie, Roth and Shearer, 1981; Ahern and Schwartz, 1985; Cole and Ray, 1985). In contrast, very few investigations used the method of evoked potentials (Ivanitsky, Kurnitskaya and Sobotka, 1986). In nearly all these studies the recordings relied upon a small number of electrodes. Topographic mapping of scalp-recorded brain electrical activity, a method primarily developed by Duffy and co-workers (Duffy , Burchfiel and Lombroso, 1979; Duffy, Bartels and Burchfiel, 1981), has gained increasing importance in clinical practice and neuropsychological research (Duffy, 1986; Maurer, 1989). Until now there has been a lack of studies, however, which use topographic brain mapping of evoked potentials to investigate hemisphere asymmetries in emotion-related stimulus processing. In the present study we therefore attempted to examine whether differences exist in the topography of brain potentials evoked by neutral versus emotional stimuli. The following questions were investigated in detail: Is it possible to differentiate between the cerebral processing of emotional and neutral stimuli by means of topographic brain mapping of evoked potentials? Are there any differences between the cerebral hemispheres during processing of emotional and neutral stimuli? 5
6
R . ROSCHMANN AND W . WITTLING
Are there intrahemispheric regions especially relevant for the processing of emotional stimuli? METHOD Subjects
Int J Neurosci Downloaded from informahealthcare.com by Nyu Medical Center on 02/08/15 For personal use only.
20 normal adult subjects (10 males, 10 females) aged 20 to 32 years participated in the experiment. All subjects were students at the Catholic University of Eichstatt. All subjects were right-handed. The “Edinburgh Handedness Inventory” (Oldfield, 1971) in the modified version of Williams (1986) was used to determine hand preference. Stimulus Material and Stimulus Presentation Subjects were shown 130 color slides of human faces grouped into two emotionrelated categories. Photographs of normal human faces served as neutral stimuli, while faces deformed by dermatological diseases were presented as emotional stimuli. In a pilot study the neutral stimuli were rated as slightly positive, whereas the emotional stimuli represented a clearly negative, aversive emotional quality. The series of emotional and neutral stimuli each included 65 different slides and were paralleled with respect to formal criteria. The slides were presented tachistoscopically with a stimulus duration of 150 ms in central vision. They were projected onto a constantly illuminated screen with a small fixation cross in the center. The fixation cross also matched the center of each slide. The pictures of both categories were presented in a random order sequence. The subjects were instructed to pay close attention to the pictures projected. Directly before a slide was projected the subject was asked to fixate the cross and keep it in focus during the presentation. After each slide the subject was asked to tell whether the picture represented a normal or a deformed face. This was to assure a sufficient level of attention on the part of the subject while avoiding higher cognitive demands. Datu Acquisition and Recording Procedures The brain mapping system we used was constructed in accordance with the BEAM technique of Duffy et al. (1979, 1981). The electrical activity of the brain was recorded from 30 electrodes attached to the scalp with collodion. The electrodes were placed according to the system depicted in Fig. 1, which allows a high electrodedensity with electrodes positioned systematically symmetric to the midline. Monopolar recordings with a time constant of 0.3 s and a upper cutoff of 32 Hz were made with linked earlobes as a reference. In order to detect artifacts horizontal and vertical eye movements were registered. Data acquisition and stimulus presentation were under the control of an Eltec 68K Computer, which digitized the EEG and EOG at 1.79 ms intervals providing a sampling rate of more than 550 Hz. The recordings covered a 200 ms pre- and a 1600 ms poststimulus period for each slide presentation. Recording was carried out in a soundattenuated, electrically shielded chamber. The potentials obtained for each subject were averaged separately for emotional and neutral stimulus conditions. Grand means for the whole subject group were cal-
Int J Neurosci Downloaded from informahealthcare.com by Nyu Medical Center on 02/08/15 For personal use only.
EMOTION-RELATED HEMISPHERE ASYMMETRIES
EAR
LEFT
INION
.:6
FIGURE 1 Placement of electrodes
7
NASION
RiGXT EAR
. ?
*
8
R. ROSCHMANN AND W. WITTLING
culated for both conditions. For statistical analysis topographic t-test comparisons for repeated measures were performed and depicted by means of significance probability mapping. A more detailed description of the methods can be found in Roschmann ( 1 990).
Int J Neurosci Downloaded from informahealthcare.com by Nyu Medical Center on 02/08/15 For personal use only.
RESULTS
To illustrate the waveforms obtained in this experiment, a typical example of a grand mean is depicted in Fig. 2. If we neglect the early peaks with a latency less than 300 ms, the potential is characterized primarily by a positive component at about 350 ms, representing a P300, followed by a long lasting negativity with its maximum at about 900 ms. Between these two extrema two peaks of relative negativity and positivity can be observed, which are, depending on electrode location, more pronounced in the anterior region. Topogruphic Mupping of Evoked Potentials In the following, the sequences of the topographic mappings of the grand means evoked by neutral and emotional stimuli are briefly described. Examples of the mappings obtained at some specific time points are illustrated in Fig. 3. Directly after stimulus onset the maps are very similar, representing a comparable, nearly neutral base level in both conditions. With a latency of about 250 ms a marked positivity developed in both series in the occipital region and spread in the anterior direction. In the interval between 350 and 370 ms this positivity reached its greatest
FIGURE 2 Example of a grand mean potential. The vertical linc marks stimulus onset that is after an elapsed time of 200 ms. Downward deflection indicates positivity.
EMOTION-RELATED HEMISPHERE ASYMMETRIES
9
Int J Neurosci Downloaded from informahealthcare.com by Nyu Medical Center on 02/08/15 For personal use only.
extent and was most pronounced in the central and parietal region. An example of this spatial patterning of amplitudes at 350 ms is depicted in Fig. 3. At about 400 ms the positivity is replaced in the prefrontal region by slight negative and neutral amplitudes. This relative negativity is for the most part restricted to the right frontal area and seems to be more pronounced in the emotional stimulus condition as the example for the 531 ms latency in Fig. 3 illustrates. Starting with a latency of about 600 ms in both map series another negativity developed in the occipital region and spread slowly in the anterior direction. With a latency of 700 ms it becomes obvious that emotional stimuli yield higher negative amplitudes and a more extensive negativity, encompassing the whole posterior region. The maps for 768 ms and 950 ms in Fig. 3 can serve as examples for this process. The negativity lasted in both conditions until about 1500 ms. t-Test Probability Mapping Topographic t-test comparisons between emotional and neutral stimulus conditions reveal no differences significant at the 5% level in the first phase of the potentials between stimulus onset and 350 ms latency. In the period between 350 and about 400 ms some regional differences become apparent, which are restricted to the frontal region and reflect a higher positivity evoked by neutral stimuli. Consistent significant differences can be detected in the interval between 5 10 ms and 560 ms in the anterior region. Fig. 4 illustrates the development of the differences in the prefrontal region (517 ms) and shows that the differences obtained were almost exclusively restricted to the right frontal area (531 ms, 542 ms, 551 ms). In the whole period, regions of significant differences indicate that emotional stimuli evoked a higher negativity than neutral stimuli. Following a longer period revealing no significant results at about 670 ms, new regional differences occurred in the right temporooccipital area and developed gradually. Fig. 5 illustrates this process for some time points (671 ms, 689 ms, 701 ms, 710 ms). For the significant areas, emotional stimuli again produced a higher negativity than neutral stimuli. At about 760 ms latency it becomes obvious, as also depicted in Fig. 5, that the pattern of significant amplitude differences has expanded over the whole posterior region but has its greatest intensity over the right hemisphere. This is indicated by some regional differences significant at the 1% level. Significant differences existed until more than 1200 ms latency but remained restricted to the posterior scalp region.
DISCUSSION Topographic brain mapping reveals marked differences between brain potentials evoked by emotional and neutral stimuli. The results suggest that the differences occurring in specific time intervals are related to hemispheric asymmetries and are topographically restricted to specific intrahemispheric regions. According to the results of t-test probability mapping, the most consistent differences between the emotional and neutral stimulus condition can be observed in two periods of the event-related potentials. Both of these show a higher negativity in response to emotional stimuli. A first phase, which seems to characterize the processing of emotional stimuli, occurs between 510 ms and 560 ms and is topographically restricted to the anterior region of the right hemisphere. Starting at about 670
Int J Neurosci Downloaded from informahealthcare.com by Nyu Medical Center on 02/08/15 For personal use only.
R . ROSCHMANN AND W. WITTLING
z
I
c c
0
v,
rr)
d
M ul
II
II
U
Y
111s
FIGURE 3 Examples of topographic maps of the grand means obtained for the neutral and the emotional stimulus condition. Amplitude values are color coded as shown on the voltage scale. Light yellow indicates maximum positive amplitudes followed by dark yellow, light red and dark red. Grey represents the lack of potential differences. In the negative direction the scale varies from dark blue to a very light blue indicating maximum negative voltages. The latencies are indicated on the left side. (See Color Plates I and 11).
t = 950
t = 768 ins
Int J Neurosci Downloaded from informahealthcare.com by Nyu Medical Center on 02/08/15 For personal use only.
+
-
Int J Neurosci Downloaded from informahealthcare.com by Nyu Medical Center on 02/08/15 For personal use only.
12
R . ROSCHMANN AND W. WITTLING
m C
z? c
Y
II
Int J Neurosci Downloaded from informahealthcare.com by Nyu Medical Center on 02/08/15 For personal use only.
EMOTION-RELATED HEMISPHERE ASYMMETRIES
c1
II
2
c
iI
13
Int J Neurosci Downloaded from informahealthcare.com by Nyu Medical Center on 02/08/15 For personal use only.
14
R. ROSCHMANN AND W. WIlTLING
Int J Neurosci Downloaded from informahealthcare.com by Nyu Medical Center on 02/08/15 For personal use only.
EMOTION-RELATED HEMISPHERE ASYMMETRIES
15
ms a second period representing an emotional response has a posterior loFation and a markedly longer duration of up to more than 1000 ms latency. The higher negativity evoked by emotional stimuli also proceeds in this phase from the right hemisphere and remains more pronounced on this side. The conclusions that can be drawn from our data are limited to stimuli with a negative emotional quality. Keeping this limitation in mind, the results suggest an asymmetry in the cortical processing of emotional stimuli which consists in higher negative amplitudes over the right hemisphere. The finding of a higher negativity evoked by emotional stimuli is partially in contrast to the results of Johnston, Miller and Burleson (1986). These authors, who did not investigate hemisphere asymmetries, found two positive components in the evoked potential associated with emotional stimuli. The experiment, however, combined emotional stimulus processing with several cognitive tasks and is not directly comparable with our study. Overall, the results of our experiment could be interpreted in terms of a 2-stagemodel of emotional stimulus processing. We suggest that the first phase of a relatively higher negativity for emotional stimuli in the right anterior region represents a basic emotional evaluation. Considering the functional characteristics of the frontal cortex this may be relevant for the initiation of motor and physiological responses. The elevated negativity in the second period on posterior location could be interpreted as a correlate of conscious emotional experiencing associated with cognitive appraisal. Leaving aside these hypothetical thoughts it should also be noted that the results of the evoked potential study, reflecting a microprocess in the brain, are in line with findings of our group obtained with a totally different research strategy. In these studies a technique for lateralizing visual input was used that allows prolonged exposure of films to one hemisphere while permitting free ocular scanning. Employing this method, described in detail by Wittling (1990), marked emotion-related hemisphere asymmetries were found for several psychological, physiological and endocrine indicators. In different studies, right-hemispheric viewing of an emotional film led to a higher degree of subjective emotional reactions, a higher increase of systolic and diastolic blood pressure and a higher increase of cortisol secretion than the lefthemispheric viewing of the same film (Wittling, 1990; Wittlhg and Huger, in press). The results of the evoked potential study reported add a new perspective to these findings and also strongly support the leading role of the right hemisphere in emotional processing.
REFERENCES Ahern, G. L. & Schwartz, G . E. (1985). Differential lateralization for positive and negative emotions in the human brain: EEG spectral analysis. Neuropsychologia, 23, 745-755. Cole, H. W. & Ray, W. J . (1985). EEG correlates of emotional tasks related to attentional demands. International Journal of Psychophysiology, 3, 33-44. Davidson, R. J . (1984). Affect, cognition and hemispheric specialization. In C . E. Izard, J . Kagan & R. B . Zajonc (Eds.), Emotions, cognition and behavior (pp. 320-365). New York: Cambridge University Press. Duffy, F. H . , Bartels, P. H. & Burchfiel, J . L. (1981). Significance probability mapping: An aid in the topographic analysis of brain electrical activity. Electroencephalography and Clinical Neurophysiology, 51, 455-462. Duffy, F. H., Burchfiel, J . L. & Lombroso, C. T. (1979). Brain electrical activity mapping (BEAM): A method for extending the clinical utility of EEG and evoked potential data. Annals of Neurology, 5 , 309-321. Duffy, F. H. (Ed.). (1986). Topographic mapping of brain electrical activiry. Boston: Butterwonhs.
Int J Neurosci Downloaded from informahealthcare.com by Nyu Medical Center on 02/08/15 For personal use only.
16
R. ROSCHMANN AND W. WITTLING
Gainotti, G. (1983). Laterality of affect: The emotional behavior of right- and left-braindamaged patients. In M. S. Myslobodsky (Ed.), Hemisyndromes (pp. 175-192). New York: Academic Press. Ivanitsky, A. M., Kurnitskaya, I. V. & Sobotka, S . (1986). Cortical topography of event-related potentials to winning and loosing in a video tennis game. International Journal of Psychophysiology, 4, 149-155. Johnston, V. S . , Miller, D. R. & Burleson, M. H. (1986). Multiple P300s to emotional stimuli and their theoretical significance. Psychophysiology, 23, 684-694. Maurer, K. (Ed.). (1989). Topogruphic brain mapping of EEG and evoked potentials. Berlin: Springer. Oldfield, R. C. (1971). The assessment and analysis of handedness: The Edinburgh inventory. Neuropsychologia, 9, 97-1 13. Roschmann, R. (1 990). Emotionule Reizverarbeitung in den zerehralen Hemisphiiren: Unrersuchung zur Topographic ereigniskorrelierter Potentiale hei ernotionsbezogener visueller Stimulation. Regensburg: Pustet. Silberman, E. K . & Weingartner, H. (1986). Hemispheric lateralization of functions related to emotion. Brain and Cognition, 9, 322-353. Tucker, D. M., Stenslie, C., Roth, R., & Shearer, S. (1981). Right frontal lobe activation and right hemispheric perfoniiance decrement during a depressed mood. Archives of General Psychiatry. 38, 169- 174. Williams, S . M. (1986). Factor analysis of the Edinburgh handedness inventory. Cortex, 22, 325-326. Wittling, W. ( 1 990). Psychophysiological correlates of human brain asymmetry: Blood pressure changes during lateralized presentation of an emotionally laden film. Neuropsychologia, 28, 457-470. Wittling, W. & Pliiger, M. (in press). Neuroendocrine hemisphere asymmetries: Salivary cortisol secretion during lateralized viewing of an emotionally aversive film. Brain and Cognition.
Intern. .I Neuroscience, . 1992, Vol. 63, p. 5-16
Reprints available directly from the publisher Photocopying permitted by license only
Printed in the United States of America
TOPOGRAPHIC BRAIN MAPPING OF EMOTIONRELATED HEMISPHERE ASYMMETRIES RUPERT ROSCHMANN AND WERNER WITTLING* Lehrstuhl fur Biopsychologie und Klinische Psychologie, Katholische Universitat Eichstatt, 0-8078 Eichstatt, FRG
Int J Neurosci Downloaded from informahealthcare.com by Nyu Medical Center on 02/08/15 For personal use only.
(Received October 5 . 1990)
The study used topographic brain mapping of visual evoked potentials to investigate emotion-related hemisphere asymmetries. The stimulus material consisted of color photographs of human faces, grouped into two emotion-related categories: normal faces (neutral stimuli) and faces deformed by dermatological diseases (emotional stimuli). The pictures were presented tachistoscopically to 20 adult right-handed subjects. Brain activity was recorded by 30 EEG electrodes with linked ears as reference. The waveforms were averaged separately with respect to each of the two stimulus conditions. Statistical analysis by means of significance probability mapping revealed significant differences between stimulus conditions for two periods of time, indicating right hemisphere superiority in emotion-related processing. The results are discussed in terms of a 2-stage-model of emotional processing in the cerebral hemispheres. Keywords: topographic brain mapping, evoked potentials, emotional hemisphere asymmetry, emotion, lateralization.
During the last few years there has been a rapid growth of interest in emotion-related hemisphere asymmetries. The majority of clinical and experimental studies has supported the hypothesis that the right cerebral hemisphere plays a dominant role in the processing of emotional stimuli and the regulation of emotional responses (Gainotti, 1983; Silberman and Weingartner, 1986). An alternative hypothesis that the right hemisphere is specialized for negative emotions and the left hemisphere for positive emotions also found some support (Davidson, 1984). Among the various research strategies used to study the phenomenon one important approach involves the recording of brain electric activity during emotion-related processing. Most of these studies dealt with EEG measures to determine relative hemispheric activation (Tucker, Stenslie, Roth and Shearer, 1981; Ahern and Schwartz, 1985; Cole and Ray, 1985). In contrast, very few investigations used the method of evoked potentials (Ivanitsky, Kurnitskaya and Sobotka, 1986). In nearly all these studies the recordings relied upon a small number of electrodes. Topographic mapping of scalp-recorded brain electrical activity, a method primarily developed by Duffy and co-workers (Duffy , Burchfiel and Lombroso, 1979; Duffy, Bartels and Burchfiel, 1981), has gained increasing importance in clinical practice and neuropsychological research (Duffy, 1986; Maurer, 1989). Until now there has been a lack of studies, however, which use topographic brain mapping of evoked potentials to investigate hemisphere asymmetries in emotion-related stimulus processing. In the present study we therefore attempted to examine whether differences exist in the topography of brain potentials evoked by neutral versus emotional stimuli. The following questions were investigated in detail: Is it possible to differentiate between the cerebral processing of emotional and neutral stimuli by means of topographic brain mapping of evoked potentials? Are there any differences between the cerebral hemispheres during processing of emotional and neutral stimuli? 5
6
R . ROSCHMANN AND W . WITTLING
Are there intrahemispheric regions especially relevant for the processing of emotional stimuli? METHOD Subjects
Int J Neurosci Downloaded from informahealthcare.com by Nyu Medical Center on 02/08/15 For personal use only.
20 normal adult subjects (10 males, 10 females) aged 20 to 32 years participated in the experiment. All subjects were students at the Catholic University of Eichstatt. All subjects were right-handed. The “Edinburgh Handedness Inventory” (Oldfield, 1971) in the modified version of Williams (1986) was used to determine hand preference. Stimulus Material and Stimulus Presentation Subjects were shown 130 color slides of human faces grouped into two emotionrelated categories. Photographs of normal human faces served as neutral stimuli, while faces deformed by dermatological diseases were presented as emotional stimuli. In a pilot study the neutral stimuli were rated as slightly positive, whereas the emotional stimuli represented a clearly negative, aversive emotional quality. The series of emotional and neutral stimuli each included 65 different slides and were paralleled with respect to formal criteria. The slides were presented tachistoscopically with a stimulus duration of 150 ms in central vision. They were projected onto a constantly illuminated screen with a small fixation cross in the center. The fixation cross also matched the center of each slide. The pictures of both categories were presented in a random order sequence. The subjects were instructed to pay close attention to the pictures projected. Directly before a slide was projected the subject was asked to fixate the cross and keep it in focus during the presentation. After each slide the subject was asked to tell whether the picture represented a normal or a deformed face. This was to assure a sufficient level of attention on the part of the subject while avoiding higher cognitive demands. Datu Acquisition and Recording Procedures The brain mapping system we used was constructed in accordance with the BEAM technique of Duffy et al. (1979, 1981). The electrical activity of the brain was recorded from 30 electrodes attached to the scalp with collodion. The electrodes were placed according to the system depicted in Fig. 1, which allows a high electrodedensity with electrodes positioned systematically symmetric to the midline. Monopolar recordings with a time constant of 0.3 s and a upper cutoff of 32 Hz were made with linked earlobes as a reference. In order to detect artifacts horizontal and vertical eye movements were registered. Data acquisition and stimulus presentation were under the control of an Eltec 68K Computer, which digitized the EEG and EOG at 1.79 ms intervals providing a sampling rate of more than 550 Hz. The recordings covered a 200 ms pre- and a 1600 ms poststimulus period for each slide presentation. Recording was carried out in a soundattenuated, electrically shielded chamber. The potentials obtained for each subject were averaged separately for emotional and neutral stimulus conditions. Grand means for the whole subject group were cal-
Int J Neurosci Downloaded from informahealthcare.com by Nyu Medical Center on 02/08/15 For personal use only.
EMOTION-RELATED HEMISPHERE ASYMMETRIES
EAR
LEFT
INION
.:6
FIGURE 1 Placement of electrodes
7
NASION
RiGXT EAR
. ?
*
8
R. ROSCHMANN AND W. WITTLING
culated for both conditions. For statistical analysis topographic t-test comparisons for repeated measures were performed and depicted by means of significance probability mapping. A more detailed description of the methods can be found in Roschmann ( 1 990).
Int J Neurosci Downloaded from informahealthcare.com by Nyu Medical Center on 02/08/15 For personal use only.
RESULTS
To illustrate the waveforms obtained in this experiment, a typical example of a grand mean is depicted in Fig. 2. If we neglect the early peaks with a latency less than 300 ms, the potential is characterized primarily by a positive component at about 350 ms, representing a P300, followed by a long lasting negativity with its maximum at about 900 ms. Between these two extrema two peaks of relative negativity and positivity can be observed, which are, depending on electrode location, more pronounced in the anterior region. Topogruphic Mupping of Evoked Potentials In the following, the sequences of the topographic mappings of the grand means evoked by neutral and emotional stimuli are briefly described. Examples of the mappings obtained at some specific time points are illustrated in Fig. 3. Directly after stimulus onset the maps are very similar, representing a comparable, nearly neutral base level in both conditions. With a latency of about 250 ms a marked positivity developed in both series in the occipital region and spread in the anterior direction. In the interval between 350 and 370 ms this positivity reached its greatest
FIGURE 2 Example of a grand mean potential. The vertical linc marks stimulus onset that is after an elapsed time of 200 ms. Downward deflection indicates positivity.
EMOTION-RELATED HEMISPHERE ASYMMETRIES
9
Int J Neurosci Downloaded from informahealthcare.com by Nyu Medical Center on 02/08/15 For personal use only.
extent and was most pronounced in the central and parietal region. An example of this spatial patterning of amplitudes at 350 ms is depicted in Fig. 3. At about 400 ms the positivity is replaced in the prefrontal region by slight negative and neutral amplitudes. This relative negativity is for the most part restricted to the right frontal area and seems to be more pronounced in the emotional stimulus condition as the example for the 531 ms latency in Fig. 3 illustrates. Starting with a latency of about 600 ms in both map series another negativity developed in the occipital region and spread slowly in the anterior direction. With a latency of 700 ms it becomes obvious that emotional stimuli yield higher negative amplitudes and a more extensive negativity, encompassing the whole posterior region. The maps for 768 ms and 950 ms in Fig. 3 can serve as examples for this process. The negativity lasted in both conditions until about 1500 ms. t-Test Probability Mapping Topographic t-test comparisons between emotional and neutral stimulus conditions reveal no differences significant at the 5% level in the first phase of the potentials between stimulus onset and 350 ms latency. In the period between 350 and about 400 ms some regional differences become apparent, which are restricted to the frontal region and reflect a higher positivity evoked by neutral stimuli. Consistent significant differences can be detected in the interval between 5 10 ms and 560 ms in the anterior region. Fig. 4 illustrates the development of the differences in the prefrontal region (517 ms) and shows that the differences obtained were almost exclusively restricted to the right frontal area (531 ms, 542 ms, 551 ms). In the whole period, regions of significant differences indicate that emotional stimuli evoked a higher negativity than neutral stimuli. Following a longer period revealing no significant results at about 670 ms, new regional differences occurred in the right temporooccipital area and developed gradually. Fig. 5 illustrates this process for some time points (671 ms, 689 ms, 701 ms, 710 ms). For the significant areas, emotional stimuli again produced a higher negativity than neutral stimuli. At about 760 ms latency it becomes obvious, as also depicted in Fig. 5, that the pattern of significant amplitude differences has expanded over the whole posterior region but has its greatest intensity over the right hemisphere. This is indicated by some regional differences significant at the 1% level. Significant differences existed until more than 1200 ms latency but remained restricted to the posterior scalp region.
DISCUSSION Topographic brain mapping reveals marked differences between brain potentials evoked by emotional and neutral stimuli. The results suggest that the differences occurring in specific time intervals are related to hemispheric asymmetries and are topographically restricted to specific intrahemispheric regions. According to the results of t-test probability mapping, the most consistent differences between the emotional and neutral stimulus condition can be observed in two periods of the event-related potentials. Both of these show a higher negativity in response to emotional stimuli. A first phase, which seems to characterize the processing of emotional stimuli, occurs between 510 ms and 560 ms and is topographically restricted to the anterior region of the right hemisphere. Starting at about 670
Int J Neurosci Downloaded from informahealthcare.com by Nyu Medical Center on 02/08/15 For personal use only.
R . ROSCHMANN AND W. WITTLING
z
I
c c
0
v,
rr)
d
M ul
II
II
U
Y
111s
FIGURE 3 Examples of topographic maps of the grand means obtained for the neutral and the emotional stimulus condition. Amplitude values are color coded as shown on the voltage scale. Light yellow indicates maximum positive amplitudes followed by dark yellow, light red and dark red. Grey represents the lack of potential differences. In the negative direction the scale varies from dark blue to a very light blue indicating maximum negative voltages. The latencies are indicated on the left side. (See Color Plates I and 11).
t = 950
t = 768 ins
Int J Neurosci Downloaded from informahealthcare.com by Nyu Medical Center on 02/08/15 For personal use only.
+
-
Int J Neurosci Downloaded from informahealthcare.com by Nyu Medical Center on 02/08/15 For personal use only.
12
R . ROSCHMANN AND W. WITTLING
m C
z? c
Y
II
Int J Neurosci Downloaded from informahealthcare.com by Nyu Medical Center on 02/08/15 For personal use only.
EMOTION-RELATED HEMISPHERE ASYMMETRIES
c1
II
2
c
iI
13
Int J Neurosci Downloaded from informahealthcare.com by Nyu Medical Center on 02/08/15 For personal use only.
14
R. ROSCHMANN AND W. WIlTLING
Int J Neurosci Downloaded from informahealthcare.com by Nyu Medical Center on 02/08/15 For personal use only.
EMOTION-RELATED HEMISPHERE ASYMMETRIES
15
ms a second period representing an emotional response has a posterior loFation and a markedly longer duration of up to more than 1000 ms latency. The higher negativity evoked by emotional stimuli also proceeds in this phase from the right hemisphere and remains more pronounced on this side. The conclusions that can be drawn from our data are limited to stimuli with a negative emotional quality. Keeping this limitation in mind, the results suggest an asymmetry in the cortical processing of emotional stimuli which consists in higher negative amplitudes over the right hemisphere. The finding of a higher negativity evoked by emotional stimuli is partially in contrast to the results of Johnston, Miller and Burleson (1986). These authors, who did not investigate hemisphere asymmetries, found two positive components in the evoked potential associated with emotional stimuli. The experiment, however, combined emotional stimulus processing with several cognitive tasks and is not directly comparable with our study. Overall, the results of our experiment could be interpreted in terms of a 2-stagemodel of emotional stimulus processing. We suggest that the first phase of a relatively higher negativity for emotional stimuli in the right anterior region represents a basic emotional evaluation. Considering the functional characteristics of the frontal cortex this may be relevant for the initiation of motor and physiological responses. The elevated negativity in the second period on posterior location could be interpreted as a correlate of conscious emotional experiencing associated with cognitive appraisal. Leaving aside these hypothetical thoughts it should also be noted that the results of the evoked potential study, reflecting a microprocess in the brain, are in line with findings of our group obtained with a totally different research strategy. In these studies a technique for lateralizing visual input was used that allows prolonged exposure of films to one hemisphere while permitting free ocular scanning. Employing this method, described in detail by Wittling (1990), marked emotion-related hemisphere asymmetries were found for several psychological, physiological and endocrine indicators. In different studies, right-hemispheric viewing of an emotional film led to a higher degree of subjective emotional reactions, a higher increase of systolic and diastolic blood pressure and a higher increase of cortisol secretion than the lefthemispheric viewing of the same film (Wittling, 1990; Wittlhg and Huger, in press). The results of the evoked potential study reported add a new perspective to these findings and also strongly support the leading role of the right hemisphere in emotional processing.
REFERENCES Ahern, G. L. & Schwartz, G . E. (1985). Differential lateralization for positive and negative emotions in the human brain: EEG spectral analysis. Neuropsychologia, 23, 745-755. Cole, H. W. & Ray, W. J . (1985). EEG correlates of emotional tasks related to attentional demands. International Journal of Psychophysiology, 3, 33-44. Davidson, R. J . (1984). Affect, cognition and hemispheric specialization. In C . E. Izard, J . Kagan & R. B . Zajonc (Eds.), Emotions, cognition and behavior (pp. 320-365). New York: Cambridge University Press. Duffy, F. H . , Bartels, P. H. & Burchfiel, J . L. (1981). Significance probability mapping: An aid in the topographic analysis of brain electrical activity. Electroencephalography and Clinical Neurophysiology, 51, 455-462. Duffy, F. H., Burchfiel, J . L. & Lombroso, C. T. (1979). Brain electrical activity mapping (BEAM): A method for extending the clinical utility of EEG and evoked potential data. Annals of Neurology, 5 , 309-321. Duffy, F. H. (Ed.). (1986). Topographic mapping of brain electrical activiry. Boston: Butterwonhs.
Int J Neurosci Downloaded from informahealthcare.com by Nyu Medical Center on 02/08/15 For personal use only.
16
R. ROSCHMANN AND W. WITTLING
Gainotti, G. (1983). Laterality of affect: The emotional behavior of right- and left-braindamaged patients. In M. S. Myslobodsky (Ed.), Hemisyndromes (pp. 175-192). New York: Academic Press. Ivanitsky, A. M., Kurnitskaya, I. V. & Sobotka, S . (1986). Cortical topography of event-related potentials to winning and loosing in a video tennis game. International Journal of Psychophysiology, 4, 149-155. Johnston, V. S . , Miller, D. R. & Burleson, M. H. (1986). Multiple P300s to emotional stimuli and their theoretical significance. Psychophysiology, 23, 684-694. Maurer, K. (Ed.). (1989). Topogruphic brain mapping of EEG and evoked potentials. Berlin: Springer. Oldfield, R. C. (1971). The assessment and analysis of handedness: The Edinburgh inventory. Neuropsychologia, 9, 97-1 13. Roschmann, R. (1 990). Emotionule Reizverarbeitung in den zerehralen Hemisphiiren: Unrersuchung zur Topographic ereigniskorrelierter Potentiale hei ernotionsbezogener visueller Stimulation. Regensburg: Pustet. Silberman, E. K . & Weingartner, H. (1986). Hemispheric lateralization of functions related to emotion. Brain and Cognition, 9, 322-353. Tucker, D. M., Stenslie, C., Roth, R., & Shearer, S. (1981). Right frontal lobe activation and right hemispheric perfoniiance decrement during a depressed mood. Archives of General Psychiatry. 38, 169- 174. Williams, S . M. (1986). Factor analysis of the Edinburgh handedness inventory. Cortex, 22, 325-326. Wittling, W. ( 1 990). Psychophysiological correlates of human brain asymmetry: Blood pressure changes during lateralized presentation of an emotionally laden film. Neuropsychologia, 28, 457-470. Wittling, W. & Pliiger, M. (in press). Neuroendocrine hemisphere asymmetries: Salivary cortisol secretion during lateralized viewing of an emotionally aversive film. Brain and Cognition.
