RIGHT BRAIN DAMAGE IMPAIRS RECOGNITION OF

NEGATIVE EMOTIONS
M.K. Mandai!, S.C. Tandon 2 and U.S. Asthana!
CDepartment of Psychology, 2Section of Neurosurgery, Department of Surgery, Banaras
Hindu University, Varanasi, India)

Studies with normal subjects have shown that the right hemisphere is so-
mewhat more efficient than the left hemisphere in the processing of emotions
(see reviews by Campbell, 1982; Leventhal and Tomarken, 1986; Tucker, 1981).
Studies with clinical subjects also confirm the association between the right
hemisphere and the processing of emotional stimuli. There is strong evidence
that right hemisphere-damaged (RHD) patients show deficit in remembering
emotional material (Wechsler, 1973), comprehending emotional tone of voice
(Heilman et aI., 1984), appreciating humorous stimuli (Bihrle et aI., 1986), and
perceiving emotional faces (Bowers et aI., 1985; Borod et aI., 1986). Dekoskyet
ai. (1980) tested unilateral brain-damaged patients on a number of emotion
judgment tasks, and found that RHD patients were impaired in most of them.
Left hemisphere-damaged (LHD) patients were superior to RHD patients but
were somewhat inferior to normal controls. Cicone, Wapner and Gardner
(1980) tested both unilateral and bilateral brain-damaged patients on an emo-
tion judgment task, and found that bilateral frontal-Iesioned patients and RHD
patients were equally poor in their judgment of facial emotion.
Researches also suggest that the hemispheres are specialized for the valence
of emotional stimuli (Ehrlichman, 1987; Tucker, 1981). For example, a right
hemisphere superiority for processing negative emotions (Ley and Bryden, 1979;
Suberi and McKeever, 1977) and a left hemisphere superiority for processing
positive emotions (Reuter-Lorenz and Davidson, 1981; Reuter-Lorenz, Givis
and Moscovitch, 1983) have been reported. Clinical studies have suggested that
left hemisphere damage results in pessimistic and depressive mood state, and
right hemisphere damage in superficial joviality or indifference (see reviews by
Leventhal and Tomarken, 1986; Tucker, 1981). In view of this, it would be in-
teresting to see whether the RHD patients' deficit in understanding emotional
stimuli is specific to negative emotions or generalized. Borod et ai. (1986) tested
this hypothesis and found that RHD patients were significantly impaired in the
perception of negative emotions but not in the perception of positive emotions
compared to LHD patients and normal controls. The present study reexamines
the issue with two more purposes in mind.
First, emotions (Ekman, 1976) may be dimensionalized both in terms of po-
sitive-negative and aroused-nonaroused emotions (Abelson and Sermat, 1962;

Cortex, (1991) 27, 247-253

248 M.K. MandaI, S.C. Tandon and H.S. Asthana

Osgood, 1966). MandaI (1986) observed that categories of emotion, especially

the negative ones, differ widely in the level of arousal. Negative emotions such
as fear, anger involve high emotional arousal and negative emotion such as sad-
ness (also disgust) involve low emotional arousal (Pilowsky and Bassett, 1980).
Emotion-specific activity in the autonomic nervous system distinguishes not
only between positive and negative emotions, but also amongst negative emo-
tions (Ekman, Levenson and Friesen, 1983). In the light of these observations,
the fundamental categories of emotion may be regrouped in terms of positive
(happiness, surprise), negative-nonaroused (sadness, disgust), and negative-ar-
oused (fear, anger) emotions, and it would be interesting to see whether the
RHD patients' deficit is specific to either of these dimensional constituents of
emotions.
A second purpose of the study is to examine brain-damaged patients' ability
to understand emotions from an inverted facial orientation. Yin (1970) ob-
served that RHD patients' (with posterior lesion) deficit in facial recognition is
specific to normal facial orientation and not to inverted facial orientation. Stu-
dies done with Yin's effect have yielded both supportive and nonsupportive re-
sults (see review by Bruyer, 1986). The present research does not aim to test
Yin's effect of orientation-specific facial processing, but to examine the differ-
ential effect of facial orientation on emotional judgment by brain-damaged pa-
tients. The emotion effect is expected to be neutralized to a greater extent in an
inverted facial orientation (Campbell, 1982).

MATERIALS AND METHOD

Subjects
Two groups of right-handed hemisphere-damaged patients (RHD: N = 10, mean
age=33 .5 yr, mean education =9.8 yr; LHD: N=1O, mean age=27.6 yr, mean educa-
tion = 8.2 yr) were selected for experimental purpose from the Section of Neurosurgery, De-
partment of Surgery, Banaras Hindu University, Varanasi. Two groups of normal controls
were also selected on the basis of handedness (Left-handers: N = 10, mean age = 20.6 yr, mean
education = 11.8 yr; Right-handers: N = 10, mean age = 26.5 yr, mean education = 12.0 yr) .
Although the brain-damaged patients were all right-handed, left-handed control subjects
were included along with right-handed controls to assess the extent of variation in perform-
ance within the non-neurologic population. The hand preference for each subject was as-
sessed using the Edinburgh Handedness Inventory (Oldfield, 1971). Right-handed subjects
with a score of below + 25 and left-handed subjects with a score of below - 25 on this in-
ventory were excluded from the study. Familial handedness was also ascertained for further
confirmation. Subjects mostly belonged to middle-class socio-economic status.
Lesion sites of all the patients were confirmed by CT scan and further verified at oper-
ation (Table I). Although four patients with left-hemisphere damage were aphasic at the be-
ginning, they had sufficient language skill at the time of testing. Patients with (a) obvious bi-
lateral damage, (b) unilateral mUltiple lesion, (c) lesion not restricted to cortex, (d) previous
psychiatric disorder, and (e) inability to keep the set required for carrying out the task, were
not included in the sample.
Photographs
Four sets of photographs, each set comprising of facial expressions of six emotions
(4 x 6 = 24), happiness, sadness, fear, anger, surprise, disgust, were selected from a series of
such photographs developed by Mandai (1984). Both male and female faces (2 sets of pho-
 Right brain damage and recognition oj emotions 249

TABLE!

Characteristics oj Brain-damaged Patients

Patient Etiology Lesion site Lesion size**
(in cm)
RHO
I. Glioma Fronto-temporal 4x3x3
2. Extradural haematoma Parietal 8x3x4
3. Subdural haematoma Fronto-temporal *'
4. Extradural haematoma Frontal 5x3x3
5. Glioma Parietal 4x3x4
6. Extradural haematoma Parietal 8x4x4
7. Chronic subdural haematoma Fronto-parietal
8. Intracerebral haematoma Temporal 4x3x3
9. Compound depressed fractive Fronto-parietal 4x4x2
10. Glioma Temporal 4x5x3
LHO
1 Glioma Frontal 4x5x3
2. Gliasis Parietal 2x2xl
3. Proencephalic cyst Parieto-occipital 3x2x4
4. Glioma Parietal 6x4x4
5. Glioma Fronto-parietal 4x5x3
6. Extradural haematoma Fronto-temporal 6x4x4
7. Ependymoma Temporal 3x4x3
8. Abscess Frontal 3x4x4
9. Extradural haematoma Parieto-occipital 4x4x5
10. Intracerebral haematoma Frontal 3x3x4
*' 60 ml of blood evacuated .
• 2 100 ml of blood evacuated.

** Length, breadth and depth.

tographs for each) were used. The detailed accounts and standardization of photographs are
not given since they have already been published (Mandai, 1987). The photographs were
black and white glossy prints measuring 19 x 17 cm, and each was a full face photograph of
an adult Indian subject.

Procedure

The first phase of the study began with photographs of facial emotions presented in an
upright orientation. Subject's ability to judge facial emotion was tested in two tasks, match-
ing and verbal labelling. Both tasks required the subjects to judge a set of six facial emo-
tions, happiness, sadness, fear, anger, surprise, disgust. In each task, two sets of photo-
graphs, one set comprised of photographs of male expressors and one set comprised of pho-
tographs of female expressors, were used.
In the matching task, six facial emotion photographs of male expressors (test) were ran-
domly placed before the subject along with a facial emotion photograph (target) of a female
expressor. Subjects were required to match the emotion expressed in the target photograph
with one of the test photographs. All the six target photographs were considered one by one,
and each time the test photographs were arranged in a random order. The whole procedure
was repeated in a second session in which photographs of female expressors were considered
'test' and photographs of male expressors 'target' . A judgment was considered correct when
an emotion was accurately matched in both sessions. In the labelling task, subjects were
shown these 12 photographs (6 emotions x male-female expressors) one by one in a random
order, and were instructed to label facial photographs by emotion word, the list of which was
given in advance. A judgment was considered correct when both photographs (male-female)
were correctly labelled.
In the second phase of the study, the photographs were presented in an inverted orien-
tation, and an identical procedure to the first phase of the study was followed, except that
250 M.K. MandaI, S.C. Tandon and H.S. Asthana

(a) two new sets (male-female expressors) of photographs were used, and (b) in the matching
task, the test photographs were presented in an upright position while the target photograph
was presented in an inverted position.

RESULTS

The percentages of correct judgment by the groups are shown in Table II.
Data were analysed with a 4 x 2 x 2 x 3 split-plot repeated measures analysis of
variance. Task condition (matching-labelling), facial orientation (upright-in-
verted), and emotion (positive, negative-aroused, negative-nonaroused) were
the within-subject factors. Main effects except facial orientation were signifi-
cant (see Table III). A significant main effect of group demonstrated superior-
ity of performance by left and right-handed normal subjects followed by LHD
and RHD patients. Post-hoc analysis by Duncan New Multiple Range Test re-
vealed that normal controls were significantly more accurate than LHD pa-
tients (p < .001), who however were significantly more accurate than RHD pa-
tients (p < .01). A significant main effect of task indicated that performance was
more accurate in the matching compared to the labelling task. Main effect of
emotion was also significant. Negative-aroused (fear, anger) emotions were less
accurately identified compared to negative-nonaroused (sadness, disgust) or
positive emotions (happiness, surprise) (Duncan test p < .001). Except the in-
teraction of group X emotion, all other interaction effects were nonsignificant
(Table III).
A significant group x emotion interaction revealed that (a) the difference in
judged accuracy between positive and negative-nonaroused emotions was non-
significant in any group (p > .01), except RHD patients who were significantly
more accurate in judging positive than negative-nonaroused emotions (p < .01),
(b) although negative-aroused emotions were less poorly recognized than other
emotions by each group, RHD patients were significantly more deficient in re-

TABLE II

Percent of Correct Judgment for Photographs of Facial Emotions Presented in Upright and Inverted"
Orientation
Emotion Positive Negative-nonaroused Negative-aroused
Group Task Happiness Surprise Sadness Disgust Fear Anger
Right Match 100 (100) 80 (60) 100 (100) 60 (90) 40 (50) 50 (30)
handers Label 100 (100) 70 (70) 70 (90) 50 (60) 20 (10) 30 (40)

Left Match 100 (100) 60 (70) 90 (100) 100 (100) 30 (40) 30 (10)
handers Label 100 (100) 50 (50) 90 (90) 80 (70) 20 (00) 20 (30)

RH Match 80 (50) 60 (50) 20 (30) 40 (20) 20 (10) 00 (00)

damage Label 80 (60) 10 (10) 60 (50) 00 (00) 00 (30) 10 (10)

LH Match 50 (40) 70 (30) 80 (70) 50 (30) 70 (60) 00 (20)

damage Label 70 (80) 00 (20) 50 (50) 00 (10) 10 (20) 10 (10)
* Shown within parenthesis.
RH =right hemisphere; LH =left hemisphere.
 Right brain damage and recognition oj emotions 251

TABLE III

Summary oj 4 x 2 x 2 xSplit-plot
3 Factorial Analysis oj Variance
Source of variation Sum of square d.L Mean square F ratio
A (Groups) 42.910 3 14.300 41.44***
Sub. within group 12.450 36 0.345
B (Orientations) 0.130 1 0.130 0.656
AxB 0.553 3 0.184 0.929
B xSub. within group 7.150 36 0.198
C (Tasks) 5.630 1 5 .630 17.48***
A xC 1.247 3 0.416 1.29
C x Sub . within group 11.620 36 0.322
D (Emotions) 66.350 2 33.175 59.24*"
AxD 15.296 6 2 .550 4.55**
D xSub. within group 40.350 72 0.560
BC 0.313 1 0.313 1.12
ABC 0.816 3 0.272 0.974
BC x Sub. within group 10.050 36 0.279
BD 0.667 2 0.333 1.266
ABD 2.362 6 0.393 1.49
BD x Sub. within group 18.950 72 0.263
CD 1.090 2 0.545 1.758
ACD 2.543 6 0.423 1.364
CD x Sub . within group 22.380 72 0.310
BCD 0.308 2 0.154 0.652
ABCD 0.478 6 0.796 0.334
BCD x Sub. within group 17.050 72 0.236
Total 280.693 479
*** < .001; ** < .01.

cogmzmg negative-aroused emotions when compared with other groups

(p < .01), (c) LHD patients were superior to RHD patients in judging either ne-
gative-aroused or negative-nonaroused emotions (p < .01), but not positive
emotions (p > .01).

DISCUSSION

The finding that normal controls were significantly superior in recognizing

facial emotions than LHD patients, who in turn were significantly superior than
RHD patients, has already been reported by Dekosky et al. (1980). In fact, a
number of studies advocated general impairment in emotional sensistivity fol-
lowing right hemisphere-damage (Cicone et aI., 1980; Bowers et aI., 1985; Bor-
od et aI., 1986). Studies also observed that LHD patients are significantly more
accurate in matching compared to labelling emotion recognition task (Cicone et
al., 1980; Dekosky et aI., 1980; Kolb and Taylor, 1981). The current data did
not confirm this proposition, although LHD patients, in comparison to other
groups, had the greatest advantage in matching task (see Table II). The inter-
action of group x task did not reach a significance level.
The findings that deserve special mention are that (a) RHD patients' global
252 M.K. Mandai, s.c. Tandon and H.S. Asthana

impairment in emotional understanding was largely attributable to impairment

in negative emotions, aroused (fear, anger) or nonaroused (sadness, disgust), (b)
RHD patients did not differ from LHD patients in understanding positive emo-
tions, and (c) the difference in performance between the presentation of two fa-
cial orientations (upright-inverted) was nonsignificant in any group.
Borod et al. (1986) observed that RHD patients are significantly more im-
paired than LHD patients and normal controls in perceiving negative emotions.
The deficiency of RHD patients to understand negative emotions partly sub-
stantiates the notion of 'lateralization of the emotional processes' (see Ehrlich-
man, 1987) which assumes left hemisphere superiority in processing positive
emotions (Reuter-Lorenz et aI., 1983) and right hemisphere superiority in pro-
cessing negative emotions (Ley and Bryden, 1979; Suberi and McKeever, 1977).
The left hemisphere effect is not supported by the current data, although LHD
patients, in comparison to other groups, were less accurate in recognizing po-
sitive emotions. It was assumed that since emotion effect is reduced in an in-
verted facial orientation, RHD patients would show significant impairment in
emotional judgment for inverted faces compared to upright faces. The present
findings did not confirm the assumption. Possibly, the reduced emotion effect
in inverted face is somewhat different from less intense emotion expressed in
upright face. This is because variation of emotional intensity in upright face is
accompanied by variation in structural composition of facial components, whi-
le structural similarities persist in inverted facial orientation with its effect re-
duced. It may be possible that the right hemisphere is more sensitive to struc-
tural differences between extreme facial emotions (Campbell, 1982). Investi-
gators found that the right hemisphere processes extreme facial emotions more
efficiently than less extreme facial emotions (Ley and Bryden, 1979). Although
the current data are supportive of this notion, extensive researches are neces-
sary before arriving at a conclusion.

ABSTRACT

Patients with right or left hemisphere-damage and normal control groups were asked to
judge facial emotions from photographs presented in two orientations - upright, inverted.
Responses were elicited with a matching and a verbal labelling task. Normal controls were
significantly superior in the judgment of facial emotions than left hemisphere-damaged pa-
tients, who in turn were significantly superior than right hemisphere-damaged patients. Ne-
gative-aroused (fear, anger) and negative-nonaroused (sadness, disgust) facial expressions
were recognized with significantly greater accuracy by left hemisphere-damaged patients
compared to right hemisphere-damaged patients; the group difference in performance was
nonsignificant for positive (happiness , surprise) emotions.

Acknowledgement. This research was supported in part by the University Grants Com-
mission of India through Career Award (No . F. 5-10/87, HR-l) to M.K. MandaI.

REFERENCES

ABELSON, R.P., and SERMAT, V. Multidimensional scaling of facial expressions. Journal of Experimental
Psychology, 63: 546-554, 1962.
BIHRLE, A.M., BROWNELL, H.H., POWELSON, 1., and GARDNER, H . Comprehension of humorous and
 Right brain damage and recognition of emotions 253

non-humorous materials by left and right brain-damaged patients. Brain and Cognition, 5: 399-412,
1986.
BOROD, J.C., KOFF, E., LORCH, M.P., and NICHOLAS, M. The expression and perception of facial emo-
tion in brain-damaged patients. Neuropsychologia, 24: 169-180, 1986.
BOWERS, D., BAUER, R.M., COSLETT, H.B., and HEILMAN, K.M. Processing of faces by patients with un-
ilaterallesions. I. Dissociation between judgments of facial affect and facial identity. Brain and Cog-
nition, 4: 258-272, 1985.
BRUYER, R. Cerebral and behavioral asymmetries in the processing of 'unusual' faces: A review. In H.D.
Ellis, M.A. Jeeves, F. Newcombe and A.W. Young (Eds.), Aspect of Face Processing. Dordrecht:
Nijhoff, 1986.
CAMPBELL, R. The lateralization of emotion: A critical review. International Journal of Psychology, 17:
211-229, 1982.
OCONE, M. , WAPNER, W., and GARDNER, H. Sensitivity to emotional expressions and situations in or-
ganic patients. Cortex, 16: 145-158, 1980.
DAVIDSON, R.J. Hemispheric asymmetry and emotion. In K.R. Scherer and P. Ekman (Eds.), Approach-
es to Emotion. Hillsdale: Erlbaum, 1984.
DEKOSKY, S.T., HEILMAN, K.M., BOWERS, D., and VALENSTEIN, E. Recognition and discrimination of
emotional faces and pictures. Brain and Language, 9: 206-214, 1980.
EKMAN, P. Pictures of Facial Affect. California, Consulting Psychologists' Press, 1976.
EKMAN, P., LEVENSON, R.W., and FRIESEN, W.V. Autonomic nervous system activity distinguishes among
emotions. Science, 221: 1208-1210, 1983.
EHRLICHMAN, H. Hemispheric asymmetry and positive-negative affect. In D. Ottoson (Ed.), Duality and
Unity of the Brain. London: MacMillan, 1987.
HEILMAN, K.M., BOWERS, D., SPEEDIE, L., and COSLETT, H.B. Comprehension of affective and nonaf-
fective prosody. Neurology, 34: 917-921, 1984.
KOLB, B., and TAYLOR, L. Affective behavior in patients with localized cortical excisions: Role of lesion
site and side. Science, 214: 89-91, 1981.
LEVENTHAL, H., and TOMARKEN, A.J. Emotion: Today's problems. Annual Review of Psychology, 37:
565-610, 1986.
LEY, R.G., and BRYDEN, M.P. Hemispheric differences in processing emotions and faces. Brain and Lan-
guage, 7: 127-138, 1979.
MANDAL, M.K. An experimental study on facial expression of emotion. Unpublished doctoral disserta-
tion, University of Calcutta, 1984.
MANDAL, M.K. Judgment of facial affect among depressives and schizophrenics. British Journal ofClin-
ical Psychology, 25: 87-92, 1986.
MANDAL, M.K. Decoding of facial emotions, in terms of expressiveness, by schizophrenics and depres-
sives. Psychiatry, 50: 371-376, 1987.
OLDFIELD, R.C. The assessment and analysis of handedness. The Edinburgh Inventory. Neuropsycho-
logia, 9: 97-114, 1971.
OSGOOD, C.E. Dimensionality of the semantic space for communication via facial expressions. Scandi-
navian Journal of Psychology, 7: 1-30, 1966.
PILOWSKY, I., and BASSETT, D. Schizophrenia and the response to facial emotions. Comprehensive Psy-
chiatry, 21: 236-244, 1980.
REUTER-LoRENZ, P., and DAVIDSON, R.J. Differential contributions of the two cerebral hemispheres for
perception of happy and sad faces. Neuropsychologia, 19: 609-614, 1981.
REUTER-LoRENZ, P., GIVIS, R.P., and MOSCOVITCH, M. Hemispheric specialization and the perception
of emotion: Evidence from right-handers and from inverted and non-inverted left-handers. Neuro-
psychologia, 21: 687-692, 1983.
SUBERI, M., and McKEEVER, W.F. Differential right hemispheric memory storage of emotional and non-
emotional faces. Neuropsychologia, 15: 757-768, 1977.
TUCKER, D.M. Lateral brain function, emotion, and conceptualization. Psychological Bulletin, 89: 19-46,
1981.
WECHSLER, A.F. The effect of organic brain disease on recall of emotionally charged versus neutral nar-
rative text. Neurology, 23: 130-135, 1973.
YIN, R. Face recognition by brain-injured patients: A dissociable ability? Neuropsychologia, 8: 395-402,
1970.

M.K. Mandai, Department of Psychology, Banaras Hindu University, Varanasi, 221005, India.