- Email: [email protected]
Decreased P3 amplitudes elicited by negative facial emotion in manic patients: Selective deficits in emotional processing
Neuroscience Letters 481 (2010) 92–96
Contents lists available at ScienceDirect
Neuroscience Letters journal homepage: www.elsevier.com/locate/neulet
Decreased P3 amplitudes elicited by negative facial emotion in manic patients: Selective deficits in emotional processing Vin Ryu a,b , Suk Kyoon An a,b , Hye Hyeon Jo b , Hyun Sang Cho a,b,∗ a b
Department of Psychiatry, College of Medicine, Yonsei University, Gyeonggi, South Korea Institute of Behavioral Science in Medicine, Yonsei University College of Medicine, South Korea
a r t i c l e
i n f o
Article history: Received 25 April 2010 Received in revised form 28 May 2010 Accepted 18 June 2010 Keywords: Bipolar disorder Negative emotion P3
a b s t r a c t Manic patients have impairments in recognizing negative emotional stimuli. However, there have been few studies on manic patients’ neurophysiological responses to facial emotions. We measured the P3 event-related potentials using facial emotional stimuli to investigate whether the impairment in recognition of negative emotions is greater in maniac patients. We recruited twenty manic patients and twenty controls. A visual oddball paradigm was used with facial pictures: happy, neutral, sad, fear, and disgust emotions. While P3 amplitudes of emotional stimuli were significantly larger than those of neutral stimuli in controls, the amplitudes were not significantly different from those for neutral pictures in manic patients. Repeated-measures analysis of variance on P3 amplitudes revealed significant interaction effects of paired emotions as sad-neutral, disgust-neutral, fear-neutral, but not in the happy-neutral emotion pairs. These differential P3 responses suggest that manic patients may have abnormal neurophysiological activity when evaluating negative facial emotions. Thus, these findings may give the evidence for reduced negative emotion recognition of manic patients. © 2010 Elsevier Ireland Ltd. All rights reserved.
Manic patients usually have difficulty in social relations due to impaired social cognition [19] and negativistic social behaviors to virtual humans [20] as well as mood regulations. In the Stroop task, both manic and depressed patients demonstrated slower response to negative emotional words, but not to positive words [24]. Manic patients also showed impaired processing of negative facial emotions, but not of positive emotions, which was inversely correlated with manic severity [22,23]. Poor processing of negative emotion has been regarded as affective bias, characterizing the manic state [25]. The P3, an event-related potential, is a positive wave occurring at about 300 ms after stimulus onset. P3 is thought to be involved in the allocation of brain resources, context updating, and attentional engagement related to memory processing [28]. P3 elicited by emotional facial expression may be related to the process of evaluation or categorization after detection and simple perception of emotional stimuli [18]. Recently, serial studies suggested that auditory P3 responses can be valid endophenotypes in psychotic bipolar patients [14,15,29]. We previously reported that negative facial emotions generated smaller P3 amplitudes than positive emotions in schizophrenic patients, but the opposite was true for healthy
∗ Corresponding author at: Department of Psychiatry, College of Medicine, Yonsei University, Gyeonggi, South Korea. Tel.: +82 31 760 9403; fax: +82 31 764 8662. E-mail address: [email protected] (H.S. Cho). 0304-3940/$ – see front matter © 2010 Elsevier Ireland Ltd. All rights reserved. doi:10.1016/j.neulet.2010.06.059
subjects [2]. This finding was also reported in a recent study on electrophysiological responses to facial emotion pictures [33]. In healthy people, emotional pictures elicit increased P3 amplitudes than neutral ones irrespective of inducing emotions [11,17]. Phasic arousal changes indicate the subjects’ energetic reaction to stimuli to affect the availability of attentional resources, and P3 amplitude might reflect this process [21]. Although bipolar disorder is a major mood disorder with impaired emotional processing, to our knowledge, no studies have examined P3 responses to affective facial expressions in manic patients. In this study, we measured P3 responses to determine whether negative facial emotion processing is impaired in manic patients. We thought it was quite likely that, when compared with neutral emotion, manic patients show smaller P3 amplitudes to negative facial emotions than healthy controls, but not to positive facial emotions. Therefore, we tried to find neurophysiological evidence of impairment in processing negative emotional in manic patients. Twenty patients (10 males, 10 females, age = 32.6 ± 9.5 years, Young Mania Rating Scale (YMRS) = 17.8 ± 5.7, Montgomery and Åsberg Depression Rating Scale (MADRS) = 4.0 ± 3.7) were recruited from the inpatients of Severance Mental Health Hospital. All patients met the criteria of non-psychotic bipolar I disorder, manic episode according to the DSM-IV [1]. Diagnosis was confirmed by the MINI [31]. We excluded patients with mixed type due to possible confounding effects of depressed mood. Diagnostic workup was done by two psychiatrists (V.R. and H.S.C.). Patients with a
V. Ryu et al. / Neuroscience Letters 481 (2010) 92–96
93
Fig. 1. (a) Illustration of the procedure. (b) Grand averages of P3 and EEG electrode arrangements.
history of closed head injury or neurological diseases and any other current axis I disorders were excluded. At EEG recording, all patients were clinically stable and scored 13 or higher on YMRS [35]. For control subjects (11 males, 9 females, age = 30.2 ± 11.5), we posted a recruitment notice on a website and selected 20 healthy subjects from respondents. Healthy volunteers had no history of bipolar disorder or other psychiatric disease and did not show any mood or thought abnormalities during the interview. All subjects were right-handed as indicated by the Annett’s handedness questionnaire [3]. Intelligence quotient (IQ) was evaluated by a short form of Korean Wechsler Adult Intelligence Test comprising three subtests: “information”, “digit span”, and “picture completion”. Participants’ subjective depressive symptoms were measured by the Beck Depression Inventory [5]. Their mood status was also objectively measured by YMRS [35] and MADRS [4]. This study was approved by the Institutional Review Board of Severance Mental Health Hospital, and conducted according to the Declaration of Helsinki. Written informed consents were obtained from all participants with adequate understanding. The subjects were shown black and white pictures of five emotional faces of two men and two women. The facial pictures were selected from the Japanese and Caucasian Facial Expression and Emotions and Neutral Faces sets [26]. We selected Japanese pictures because of higher recognition rates of Japanese faces than Caucasians in our previous study. Our recent data from 143 healthy Korean volunteers demonstrated that the average recognition rate was fair: 99.6% to four happy faces, 83.0% to four sad faces, 71.7% to four fear faces, and 72.1% to four disgusted faces (unpublished). As indicated, we selected four pictures for each emotion (happiness, sadness, fear, disgust, and neutrality). These 20 pictures were grouped into five blocks. Checkerboard pictures were non-target, and emotional pictures were target. Subjects were instructed to respond to target stimuli by pressing a button on the keypad and not to respond to non-target stimuli. They were also required to feel the emotion of presented faces, which instruction was according to our previous report of measuring the P3 amplitude evoked by the facial emotional stimuli [2]. Stimuli were presented 400 times in a fixed pseudo-randomized order with 2:8 ratio (target:non-target).
Duration of stimuli presentation was 200 ms/picture and interstimulus interval were 800 ms (Fig. 1a). The session consisted of one practice block and five actual test blocks, which sequences were counterbalanced with subjects. Each block consisted of checkerboard pictures and five emotional stimuli in a random order. Subjects were seated in a comfortable reclining chair at an eyedistance of 50 cm from the computer monitor (visual angle of 9◦ × 12◦ ) in a dimly lighted, quiet, electrically shielded electroencephalography room. Subjects were instructed to concentrate on the center of the monitor to avoid eye-blinking as much as possible. It took about 400 s to present each block which was followed by an intersession break. The subjects’ performances were monitored by the closed-circuit camera, and no subjects were sleepy during the experiment. After EEG recording, the subjects performed a facial emotion recognition task presented on the monitor. These task stimuli consisted of the pictures used in the EEG recording. All participants were asked to select their perceived emotion of the facial pictures of a list of five emotions (neutral, sad, happy, fear, and disgust). EEG activity was recorded from 64-channel AgCl lead cap according to the international 10/10 system with a 0.05–100 Hz analogue band-pass and a sampling rate of 1000 Hz/channel (SynAmpsII). The recordings were referenced to linked electrodes placed on the left and right mastoid processes. Eye blinks and movements were monitored by electrodes placed near the outer canthus and beneath the left eye. Electrode impedance was maintained at 10 k or below. Analysis of EEG was carried out on the off-line basis. Salient noises of EEG were eliminated by inspection. EEG was amplified by 0.15–40 Hz band-pass filter. For controlling ocular movement artifacts, trials were adjusted by regression from electrooculograms [30]. Artifacts were rejected if F1, F2, FP1, and FP7 potential amplitudes exceeded ±50 V. Low-pass filter was 8.5 Hz for removing muscular movement, noise, and alpha-wave activity. Event-related potentials (ERP) were averaged between prestimulus 100 ms and post-stimulus 900 ms. The number of epochs included in the analysis was 65.5 ± 10.3 with a minimum of 46 of a total of 80 trials. P3 was defined as the largest positive peak between 250 ms and 550 ms after stimulus onset at Pz electrode.
94
V. Ryu et al. / Neuroscience Letters 481 (2010) 92–96
Table 1 Results of Emotion (5) × Region (3) × Hemisphere (3) × Group (2) ANOVA for P3 amplitudes. Effect Group (G) Emotion (E) Region (R) Hemisphere (H) E×G R×G H×G E×R E×H R×H E×H×G E×R×G R×H×G E×R×H E×R×H×G a
dfa 138 4152 276 276 4152 276 276 8304 8304 4152 8304 8304 4152 16,608 16,608
F
p
12.6 0.6 46.6 23.7 8.0 6.6 7.4 0.5 0.9 12.8 1.2 0.6 1.1 0.9 1.7
0.001 0.63 <0.001 <0.001 <0.001 0.007 0.002 0.77 0.50 <0.001 0.29 0.71 0.34 0.49 0.13
Degrees of freedom.
P3 of other electrodes were defined as the largest positive peak within ±50 ms of P3 at Pz. Peak amplitude (V) was calculated as the voltage difference between a component peak and a prestimulus baseline. Peak latency was defined as the time between stimulus onset and the peak maximum. P3 amplitudes and peak latencies were analyzed using repeatedmeasures analysis of variance to assess the effect by group (patients or controls), emotional condition (happy, neutral, sad, fear, or disgust), region (frontal, central, or parietal) and scalp electrode location (left, middle, or right hemisphere). Greenhouse–Geisser correction for non-sphericity was applied. Post hoc pairwise comparison was performed for within-group and between-group factors with Bonferroni corrections for multiple comparisons. P3 amplitude and peak latencies were collapsed across 9 electrode examining effects of emotional condition, group and interaction. Target detection and recognition rates of the facial emotions were analyzed with multivariate analysis of variance (MANOVA) between patients and controls. To evaluate the clinical variables and medication effect of patients, correlation analyses and Mann–Whitney U tests were performed. Unless indicated otherwise, data are presented as the mean ± SD. SPSS version 12.0 was used for these statistical analyses. There were no significant differences between the groups with regard to age, sex, IQ (109.1 ± 11.2, patients, 113.7 ± 9.9, controls) and duration of education (14.8 ± 1.7, patients, 15.7 ± 1.1, controls). Fifteen patients were on divalproex (serum concentration, 80.07 ± 17.00 g/mL), and the remaining five patients were on lithium (serum concentration, 0.55 ± 0.09 mmol/L). Nineteen patients had antipsychotics (531.8 ± 327.2 mg chlorpromazineequivalent dose). Grand averages of the ERP responses to different emotional conditions are demonstrated in Fig. 1b. For comparing P3 amplitudes between controls and patients, a repeated-measures group by emotional condition by region by hemisphere ANOVA for all subjects was done. It revealed group had a main effect (F(1,38) = 12.6, p < 0.001) as amplitudes in controls were larger than manic patients. It also showed significant interactions of group by emotional condition (F(4,152) = 8.0, p < 0.001). There was a main effect of region (F(2,76) = 46.6, p < 0.001) and a main effect of hemisphere (F(2,76) = 23.7, p < 0.001), which is compatible to P3 responses with highest at mid-parietal electrodes. There were significant interactions of region and group (F(2,76) = 6.6, p = 0.007), and of hemisphere and group (F(2,76) = 7.4, p = 0.002), in which amplitudes in the central and middle sites increased less in patients than controls. However, there were no other interaction effects (Table 1).
While the patient group did not show the main effect of emotional conditions (F(4,76) = 1.7, p = 0.17), the control group revealed the main effect of emotional conditions (F(4,76) = 8.89, p < 0.001). Post hoc comparisons between neutral and emotional conditions within the control group revealed a significant difference in P3 amplitudes of neutral-happy (F(1,19) = 5.6), neutral-sad (F(1,19) = 16.5, p = 0.001), neutral-fearful (F(1,19) = 13.8, p = 0.001), and neutral-disgust (F(1,19) = 9.5, p = 0.006) conditions, in which amplitudes of neutral condition were lower than those of other emotional conditions. To investigate the interaction effect for group by emotional condition, we used paired emotional conditions (i.e., a neutral condition and each emotional condition) as within-group factors. While happy-neutral condition did not have interactions of group by emotional condition (F(1,38) = 2.1, p = 0.15), sad-neutral condition (F(1,38) = 20.1, p < 0.001), fear-neutral (F(1,38) = 8.6, p = 0.006), and disgust-neutral condition (F(1,38) = 15.7, p < 0.001) had interactions of group by emotional conditions (Fig. 2). Post hoc t test revealed that P3 amplitudes in manic patients were significantly lower than control group for sad (t = −4.45, p < 0.001), fear (t = −3.57, p = 0.001), disgust (t = −3.72, p = 0.001), and neutral (t = −2.03, p = 0.05). Group (F(1,38) = 15.2, p < 0.001), hemisphere (F(2,76) = 7.1, p = 0.002) and region (F(2,72) = 12.1, p < 0.001) had main effects on P3 peak latencies. P3 latencies of patients were slower than those of controls. There was an interaction of group by region (F(2,76) = 4.9, p = 0.01) as well. However, we did not observe any significant main or interaction effect of emotion, group by emotion, or group by emotion by region. MANOVA for target detection rates revealed no differences between controls and patients (F(5,34) = 1.56, p = 0.20). MANOVA for recognition rates after EEG recordings revealed no group differences (F(5,34) = 1.81, p = 0.14). Recognition rates were 90.0 ± 23.5% (patients) and 92.5 ± 14.3% (controls) for happy faces, 61.3 ± 35.8% (patients) and 52.5 ± 32.3% (controls) for neutral faces, 66.5 ± 37.8% (patients) and 88.8 ± 22.2% (controls) for sad faces, 90.0 ± 23.5% (patients) and 92.5 ± 14.3% (controls) for fear faces, and 76.3 ± 31.9% (patients) and 67.5 ± 38.1% (controls) for disgust faces. Independent t test revealed a difference in the recognition rates for sad stimuli (t = 2.23, p = 0.03) between patients and controls, but no significant difference in the other emotions. Patients’ misrecognition rate of neutral stimuli as sad (5.0 ± 10.3%) was lower than that of controls (15.0 ± 17.0%) (t = 2.25, p = 0.03). Misrecognition rates of neutral stimuli as fear was lower in patients (3.8 ± 9.2%) than in controls (11.3 ± 12.8%) (t = −2.14, p = 0.04). Misrecognition rates of sad stimuli as neutral was higher in patients (15.0 ± 17.0%) than in controls (3.8 ± 9.2%) (t = 2.60, p = 0.01). There were no differences in other misrecognition rates between patients and controls. YMRS, BPRS, BDI and MADRS scores were not significantly correlated with P3 amplitudes at Pz electrode (p values ranged from 0.20 to 0.99). P3 amplitudes and peak latencies were not significantly correlated with chlorpromazine-equivalent doses at all electrode sites for all emotional conditions (p values from 0.11 to 0.66). There were also no differences between lithium use and divalproex use by Mann–Whitney U test in P3 amplitude and latency of Pz (p values from 0.31 to 0.97). This study investigated event-related P3 responses to facial emotional stimuli in a sample of manic patients and healthy controls. The analysis of group by neutral-emotional condition on P3 amplitudes revealed significant interaction effects in condition pairs of sad-neutral, disgust-neutral, fear-neutral conditions, but not happy-neutral conditions. Manic patients demonstrated significantly lower P3 amplitudes than controls for each emotional stimulus. While healthy subjects showed larger amplitudes to all kinds of emotional, especially negative faces compared to neu-
V. Ryu et al. / Neuroscience Letters 481 (2010) 92–96
95
Fig. 2. Interactions of groups by emotional conditions (neutral vs. emotional).
tral faces, manic patients showed no difference in P3 amplitudes between emotional faces and neutral faces. P3 amplitude indexes brain activity of comparing incoming stimuli with previous ones. The attentional process that detects updating of stimuli is concomitant with P3 responses. Therefore, our findings might be the results of impaired attention reallocation strategy, especially for negative emotion in manic patients. Larger P3 amplitudes elicited by negative emotions in our healthy subjects can be supported by previous studies. It has been reported that normal subjects show larger late positive potentials (LPPs) to emotional pictures than neutral ones and increaseed LPP amplitude with emotional arousal/intensity [10]. Furthermore, healthy people exhibit stronger sensitivity and larger LPPs to negative stimuli than neutral stimuli [17]. Our study demonstrated that manic patients showed no difference in P3 amplitudes between emotional stimuli and neutral stimuli, whereas healthy controls showed increased P3 amplitude to emotional stimuli compared to neutral stimuli. These findings suggest that emotional effects, larger amplitude responses to negative stimuli, may not work in manic state. Therefore, manic patients tend to underestimate or misperceive negative emotional stimuli. As described earlier, manic patients demonstrated poor processing for negative emotional stimuli [22,23]. Thus, the differential P3 responses to emotional valences observed in this study may be a neurophysiological evidence of impaired negative facial emotion processing in mania. Most studies have shown prolonged P3 latency in psychotic or non-psychotic manic patients [27,32]. All these studies used the auditory oddball paradigms to evaluate stimulus relevance. In the present study, a visual oddball paradigm with facial emotional stimuli was used to categorize the emotion as in the behavioral studies [22,23]. P3 latency, indexing classification speed, may be related with detection and evaluation of stimuli [28]. While our manic subjects showed prolonged P3 latencies compared to controls, there were no effects of emotional condition and group by emotional condition. This finding suggests that evaluation speed on facial expression may be delayed, irrespective of emotional valences, in manic status. Reduced P3 amplitudes for negative emotions might be explained by some functional magnetic imaging (fMRI) studies. Manic patients showed weak intensity rating of sad faces, which is associated with significant attenuation in the anterior cingulate
gyrus and amygdala [23]. Moreover, their corticolimbic regions with frontal and temporal cortices and limbic and paralimbic regions were underactivated during the explicit processing of sad faces [9]. Manic patients also showed increased blood oxygen leveldependent (BOLD) responses in the amygdala to positive affective stimuli, not to negative stimuli [6]. Generally, the model of neural activation of underlying P3 generation reveals that external stimuli are maintained and monitored by the frontal areas and anterior cingulate regions, and then they are transmitted to the temporal and parietal areas that produce P3 responses [7,28]. Thus, it seems that decreased activation of some cortical or corticolimbic areas during emotional processing is involved in reduced P3 amplitude to negative stimuli observed in this study [9,23]. In this study, we could not identify whether the differential P3 response to each emotion shown in manic state will be observed in euthymic state as enduring impairments. Behavioral studies on facial emotion perception in euthymic bipolar patients have shown inconsistent findings. Some have reported impairment in recognizing faces [8,16], but others have reported no difference in facial recognition between patients and controls [22,34]. We need to further investigate P3 responses to emotional face stimuli in euthymic patients. Our study has limitations. First, correct recognition rate of neutral faces in our study was low (61.3% in control, 52.5% in patients). In a study with 143 healthy volunteers, it was very low score in 7-point Likert scale that the recognition intensity of neutral face as other emotional faces (1.55 ± 0.78), while intensity scores of other emotions were high (5.66 ± 0.91 for happy, 3.16 ± 0.96 for sad, 4.90 ± 0.78 for fear, and 4.88 ± 1.08 for disgust) [13]. The relatively low emotional intensity might support the applicability of the comparison between emotional and neutral faces. Second, we found no significant interaction effects of group by happy-neutral condition. However, due to the small sample size, this study might not have enough power to distinguish differences between the two groups. Third, we found no statistical association of P3 amplitudes with neuroleptic dose or the type of mood stabilizers. It was reported that P3 waves are not affected by antipsychotic drugs [12]. However, we cannot completely exclude possible effects of drugs used in patients due to the small number of subjects. In summary, patients showed no P3 amplitude differences between negative and neutral faces and more reduced P3 ampli-
96
V. Ryu et al. / Neuroscience Letters 481 (2010) 92–96
tudes to negative facial emotions compared to normal subjects. These findings may give the neurophysiological evidence of impaired negative emotional processing, which is consistent with mood-congruent emotional bias in manic patients. Acknowledgements This study was supported by a faculty research grant (No. 6-2009-0118) of Yonsei University College of Medicine and the intramural grant of the Institute of Behavioral Science in Medicine, Yonsei University College of Medicine. References [1] American Psychiatric Association, Diagnostic and Statistical Manual of Mental Disorders, American Psychiatric Association, Washington, DC, 2000. [2] S.K. An, S.J. Lee, C.H. Lee, H.S. Cho, P.G. Lee, C.I. Lee, E. Lee, K.S. Roh, K. Namkoong, Reduced P3 amplitudes by negative facial emotional photographs in schizophrenia, Schizophr. Res. 64 (2003) 125–135. [3] M. Annett, A classification of hand preference by association analysis, Br. J. Psychol. 61 (1970) 303–321. [4] M. Asberg, S.A. Montgomery, C. Perris, D. Schalling, G. Sedvall, A comprehensive psychopathological rating scale, Acta Psychiatr. Scand. Suppl. (1978) 5–27. [5] A.T. Beck, C.H. Ward, M. Mendelson, J. Mock, J. Erbaugh, An inventory for measuring depression, Arch. Gen. Psychiatry 4 (1961) 561–571. [6] F. Bermpohl, U. Dalanay, T. Kahnt, B. Sajonz, H. Heimann, R. Ricken, M. Stoy, C. Hagele, F. Schlagenhauf, M. Adli, J. Wrase, A. Strohle, A. Heinz, M. Bauer, A preliminary study of increased amygdala activation to positive affective stimuli in mania, Bipolar Disord. 11 (2009) 70–75. [7] C. Bledowski, D. Prvulovic, K. Hoechstetter, M. Scherg, M. Wibral, R. Goebel, D.E. Linden, Localizing P300 generators in visual target and distractor processing: a combined event-related potential and functional magnetic resonance imaging study, J. Neurosci. 24 (2004) 9353–9360. [8] V.P. Bozikas, T. Tonia, K. Fokas, A. Karavatos, M.H. Kosmidis, Impaired emotion processing in remitted patients with bipolar disorder, J. Affect. Disord. 91 (2006) 53–56. [9] C.H. Chen, B. Lennox, R. Jacob, A. Calder, V. Lupson, R. Bisbrown-Chippendale, J. Suckling, E. Bullmore, Explicit and implicit facial affect recognition in manic and depressed States of bipolar disorder: a functional magnetic resonance imaging study, Biol. Psychiatry 59 (2006) 31–39. [10] S. Delplanque, L. Silvert, P. Hot, S. Rigoulot, H. Sequeira, Arousal and valence effects on event-related P3a and P3b during emotional categorization, Int. J. Psychophysiol. 60 (2006) 315–322. [11] M. Eimer, A. Holmes, Event-related brain potential correlates of emotional face processing, Neuropsychologia 45 (2007) 15–31. [12] J.M. Ford, P.M. White, J.G. Csernansky, W.O. Faustman, W.T. Roth, A. Pfefferbaum, ERPs in schizophrenia: effects of antipsychotic medication, Biol. Psychiatry 36 (1994) 153–170. [13] R.Y. Ha, J.I. Kang, S.K. An, H.S. Cho, Some psychological correlates affecting recognition of neutral facial emotion in young adults, J. Korean Neuropsychiatr. Assoc. 48 (2009) 481–487. [14] M.H. Hall, F. Rijsdijk, S. Kalidindi, K. Schulze, E. Kravariti, F. Kane, P. Sham, E. Bramon, R.M. Murray, Genetic overlap between bipolar illness and event-related potentials, Psychol. Med. 37 (2007) 667–678.
[15] M.H. Hall, K. Schulze, F. Rijsdijk, S. Kalidindi, C. McDonald, E. Bramon, R.M. Murray, P. Sham, Are auditory P300 and duration MMN heritable and putative endophenotypes of psychotic bipolar disorder? A maudsley bipolar twin and family study, Psychol. Med. 39 (2009) 1277–1287. [16] C.J. Harmer, L. Grayson, G.M. Goodwin, Enhanced recognition of disgust in bipolar illness, Biol. Psychiatry 51 (2002) 298–304. [17] T.A. Ito, J.T. Larsen, N.K. Smith, J.T. Cacioppo, Negative information weighs more heavily on the brain: the negativity bias in evaluative categorizations, J. Pers. Soc. Psychol. 75 (1998) 887–900. [18] R. Johnson Jr., A triarchic model of P300 amplitude, Psychophysiology 23 (1986) 367–384. [19] N. Kerr, R.I. Dunbar, R.P. Bentall, Theory of mind deficits in bipolar affective disorder, J. Affect. Disord. 73 (2003) 253–259. [20] E. Kim, J. Ku, J.J. Kim, H. Lee, K. Han, S.I. Kim, H.S. Cho, Nonverbal social behaviors of patients with bipolar mania during interactions with virtual humans, J. Nerv. Ment. Dis. 197 (2009) 412–418. [21] A. Kok, On the utility of P3 amplitude as a measure of processing capacity, Psychophysiology 38 (2001) 557–577. [22] A. Lembke, T.A. Ketter, Impaired recognition of facial emotion in mania, Am. J. Psychiatry 159 (2002) 302–304. [23] B.R. Lennox, R. Jacob, A.J. Calder, V. Lupson, E.T. Bullmore, Behavioural and neurocognitive responses to sad facial affect are attenuated in patients with mania, Psychol. Med. 34 (2004) 795–802. [24] H.M. Lyon, M. Startup, R.P. Bentall, Social cognition and the manic defense: attributions, selective attention, and self-schema in bipolar affective disorder, J. Abnorm. Psychol. 108 (1999) 273–282. [25] W. Mansell, R. Pedley, The ascent into mania: a review of psychological processes associated with the development of manic symptoms, Clin. Psychol. Rev. 28 (2008) 494–520. [26] D. Matsumoto, P. Ekman, Japanese and Caucasian Facial Expressions of Emotion (JACFEE) Slides, San Francisco, CA, 1988. [27] B.F. O’Donnell, J.L. Vohs, W.P. Hetrick, C.A. Carroll, A. Shekhar, Auditory eventrelated potential abnormalities in bipolar disorder and schizophrenia, Int. J. Psychophysiol. 53 (2004) 45–55. [28] J. Polich, Updating P300: an integrative theory of P3a and P3b, Clin. Neurophysiol. 118 (2007) 2128–2148. [29] K.K. Schulze, M.H. Hall, C. McDonald, N. Marshall, M. Walshe, R.M. Murray, E. Bramon, Auditory P300 in patients with bipolar disorder and their unaffected relatives, Bipolar Disord. 10 (2008) 377–386. [30] H.V. Semlitsch, P. Anderer, P. Schuster, O. Presslich, A solution for reliable and valid reduction of ocular artifacts, applied to the P300 ERP, Psychophysiology 23 (1986) 695–703. [31] D.V Sheehan, Y. Lecrubier, K.H. Sheehan, P. Amorim, J. Janavs, E. Weiller, T. Hergueta, R. Baker, G.C. Dunbar, The Mini-International Neuropsychiatric Interview (M.I.N.I.): the development and validation of a structured diagnostic psychiatric interview for DSM-IV and ICD-10, J. Clin. Psychiatry 59 (Suppl. 20) (1998) 22–33, quiz 34–57. [32] W.K. Strik, M. Ruchsow, S. Abele, A.J. Fallgatter, T.J. Mueller, Distinct neurophysiological mechanisms for manic and cycloid psychoses: evidence from a P300 study on manic patients, Acta Psychiatr. Scand. 98 (1998) 459–466. [33] B.I. Turetsky, C.G. Kohler, T. Indersmitten, M.T. Bhati, D. Charbonnier, R.C. Gur, Facial emotion recognition in schizophrenia: when and why does it go awry? Schizophr. Res. 94 (2007) 253–263. [34] H.R. Venn, J.M. Gray, B. Montagne, L.K. Murray, D. Michael Burt, E. Frigerio, D.I. Perrett, A.H. Young, Perception of facial expressions of emotion in bipolar disorder, Bipolar Disord. 6 (2004) 286–293. [35] R.C. Young, J.T. Biggs, V.E. Ziegler, D.A. Meyer, A rating scale for mania: reliability, validity and sensitivity, Br. J. Psychiatry 133 (1978) 429–435.
Contents lists available at ScienceDirect
Neuroscience Letters journal homepage: www.elsevier.com/locate/neulet
Decreased P3 amplitudes elicited by negative facial emotion in manic patients: Selective deficits in emotional processing Vin Ryu a,b , Suk Kyoon An a,b , Hye Hyeon Jo b , Hyun Sang Cho a,b,∗ a b
Department of Psychiatry, College of Medicine, Yonsei University, Gyeonggi, South Korea Institute of Behavioral Science in Medicine, Yonsei University College of Medicine, South Korea
a r t i c l e
i n f o
Article history: Received 25 April 2010 Received in revised form 28 May 2010 Accepted 18 June 2010 Keywords: Bipolar disorder Negative emotion P3
a b s t r a c t Manic patients have impairments in recognizing negative emotional stimuli. However, there have been few studies on manic patients’ neurophysiological responses to facial emotions. We measured the P3 event-related potentials using facial emotional stimuli to investigate whether the impairment in recognition of negative emotions is greater in maniac patients. We recruited twenty manic patients and twenty controls. A visual oddball paradigm was used with facial pictures: happy, neutral, sad, fear, and disgust emotions. While P3 amplitudes of emotional stimuli were significantly larger than those of neutral stimuli in controls, the amplitudes were not significantly different from those for neutral pictures in manic patients. Repeated-measures analysis of variance on P3 amplitudes revealed significant interaction effects of paired emotions as sad-neutral, disgust-neutral, fear-neutral, but not in the happy-neutral emotion pairs. These differential P3 responses suggest that manic patients may have abnormal neurophysiological activity when evaluating negative facial emotions. Thus, these findings may give the evidence for reduced negative emotion recognition of manic patients. © 2010 Elsevier Ireland Ltd. All rights reserved.
Manic patients usually have difficulty in social relations due to impaired social cognition [19] and negativistic social behaviors to virtual humans [20] as well as mood regulations. In the Stroop task, both manic and depressed patients demonstrated slower response to negative emotional words, but not to positive words [24]. Manic patients also showed impaired processing of negative facial emotions, but not of positive emotions, which was inversely correlated with manic severity [22,23]. Poor processing of negative emotion has been regarded as affective bias, characterizing the manic state [25]. The P3, an event-related potential, is a positive wave occurring at about 300 ms after stimulus onset. P3 is thought to be involved in the allocation of brain resources, context updating, and attentional engagement related to memory processing [28]. P3 elicited by emotional facial expression may be related to the process of evaluation or categorization after detection and simple perception of emotional stimuli [18]. Recently, serial studies suggested that auditory P3 responses can be valid endophenotypes in psychotic bipolar patients [14,15,29]. We previously reported that negative facial emotions generated smaller P3 amplitudes than positive emotions in schizophrenic patients, but the opposite was true for healthy
∗ Corresponding author at: Department of Psychiatry, College of Medicine, Yonsei University, Gyeonggi, South Korea. Tel.: +82 31 760 9403; fax: +82 31 764 8662. E-mail address: [email protected] (H.S. Cho). 0304-3940/$ – see front matter © 2010 Elsevier Ireland Ltd. All rights reserved. doi:10.1016/j.neulet.2010.06.059
subjects [2]. This finding was also reported in a recent study on electrophysiological responses to facial emotion pictures [33]. In healthy people, emotional pictures elicit increased P3 amplitudes than neutral ones irrespective of inducing emotions [11,17]. Phasic arousal changes indicate the subjects’ energetic reaction to stimuli to affect the availability of attentional resources, and P3 amplitude might reflect this process [21]. Although bipolar disorder is a major mood disorder with impaired emotional processing, to our knowledge, no studies have examined P3 responses to affective facial expressions in manic patients. In this study, we measured P3 responses to determine whether negative facial emotion processing is impaired in manic patients. We thought it was quite likely that, when compared with neutral emotion, manic patients show smaller P3 amplitudes to negative facial emotions than healthy controls, but not to positive facial emotions. Therefore, we tried to find neurophysiological evidence of impairment in processing negative emotional in manic patients. Twenty patients (10 males, 10 females, age = 32.6 ± 9.5 years, Young Mania Rating Scale (YMRS) = 17.8 ± 5.7, Montgomery and Åsberg Depression Rating Scale (MADRS) = 4.0 ± 3.7) were recruited from the inpatients of Severance Mental Health Hospital. All patients met the criteria of non-psychotic bipolar I disorder, manic episode according to the DSM-IV [1]. Diagnosis was confirmed by the MINI [31]. We excluded patients with mixed type due to possible confounding effects of depressed mood. Diagnostic workup was done by two psychiatrists (V.R. and H.S.C.). Patients with a
V. Ryu et al. / Neuroscience Letters 481 (2010) 92–96
93
Fig. 1. (a) Illustration of the procedure. (b) Grand averages of P3 and EEG electrode arrangements.
history of closed head injury or neurological diseases and any other current axis I disorders were excluded. At EEG recording, all patients were clinically stable and scored 13 or higher on YMRS [35]. For control subjects (11 males, 9 females, age = 30.2 ± 11.5), we posted a recruitment notice on a website and selected 20 healthy subjects from respondents. Healthy volunteers had no history of bipolar disorder or other psychiatric disease and did not show any mood or thought abnormalities during the interview. All subjects were right-handed as indicated by the Annett’s handedness questionnaire [3]. Intelligence quotient (IQ) was evaluated by a short form of Korean Wechsler Adult Intelligence Test comprising three subtests: “information”, “digit span”, and “picture completion”. Participants’ subjective depressive symptoms were measured by the Beck Depression Inventory [5]. Their mood status was also objectively measured by YMRS [35] and MADRS [4]. This study was approved by the Institutional Review Board of Severance Mental Health Hospital, and conducted according to the Declaration of Helsinki. Written informed consents were obtained from all participants with adequate understanding. The subjects were shown black and white pictures of five emotional faces of two men and two women. The facial pictures were selected from the Japanese and Caucasian Facial Expression and Emotions and Neutral Faces sets [26]. We selected Japanese pictures because of higher recognition rates of Japanese faces than Caucasians in our previous study. Our recent data from 143 healthy Korean volunteers demonstrated that the average recognition rate was fair: 99.6% to four happy faces, 83.0% to four sad faces, 71.7% to four fear faces, and 72.1% to four disgusted faces (unpublished). As indicated, we selected four pictures for each emotion (happiness, sadness, fear, disgust, and neutrality). These 20 pictures were grouped into five blocks. Checkerboard pictures were non-target, and emotional pictures were target. Subjects were instructed to respond to target stimuli by pressing a button on the keypad and not to respond to non-target stimuli. They were also required to feel the emotion of presented faces, which instruction was according to our previous report of measuring the P3 amplitude evoked by the facial emotional stimuli [2]. Stimuli were presented 400 times in a fixed pseudo-randomized order with 2:8 ratio (target:non-target).
Duration of stimuli presentation was 200 ms/picture and interstimulus interval were 800 ms (Fig. 1a). The session consisted of one practice block and five actual test blocks, which sequences were counterbalanced with subjects. Each block consisted of checkerboard pictures and five emotional stimuli in a random order. Subjects were seated in a comfortable reclining chair at an eyedistance of 50 cm from the computer monitor (visual angle of 9◦ × 12◦ ) in a dimly lighted, quiet, electrically shielded electroencephalography room. Subjects were instructed to concentrate on the center of the monitor to avoid eye-blinking as much as possible. It took about 400 s to present each block which was followed by an intersession break. The subjects’ performances were monitored by the closed-circuit camera, and no subjects were sleepy during the experiment. After EEG recording, the subjects performed a facial emotion recognition task presented on the monitor. These task stimuli consisted of the pictures used in the EEG recording. All participants were asked to select their perceived emotion of the facial pictures of a list of five emotions (neutral, sad, happy, fear, and disgust). EEG activity was recorded from 64-channel AgCl lead cap according to the international 10/10 system with a 0.05–100 Hz analogue band-pass and a sampling rate of 1000 Hz/channel (SynAmpsII). The recordings were referenced to linked electrodes placed on the left and right mastoid processes. Eye blinks and movements were monitored by electrodes placed near the outer canthus and beneath the left eye. Electrode impedance was maintained at 10 k or below. Analysis of EEG was carried out on the off-line basis. Salient noises of EEG were eliminated by inspection. EEG was amplified by 0.15–40 Hz band-pass filter. For controlling ocular movement artifacts, trials were adjusted by regression from electrooculograms [30]. Artifacts were rejected if F1, F2, FP1, and FP7 potential amplitudes exceeded ±50 V. Low-pass filter was 8.5 Hz for removing muscular movement, noise, and alpha-wave activity. Event-related potentials (ERP) were averaged between prestimulus 100 ms and post-stimulus 900 ms. The number of epochs included in the analysis was 65.5 ± 10.3 with a minimum of 46 of a total of 80 trials. P3 was defined as the largest positive peak between 250 ms and 550 ms after stimulus onset at Pz electrode.
94
V. Ryu et al. / Neuroscience Letters 481 (2010) 92–96
Table 1 Results of Emotion (5) × Region (3) × Hemisphere (3) × Group (2) ANOVA for P3 amplitudes. Effect Group (G) Emotion (E) Region (R) Hemisphere (H) E×G R×G H×G E×R E×H R×H E×H×G E×R×G R×H×G E×R×H E×R×H×G a
dfa 138 4152 276 276 4152 276 276 8304 8304 4152 8304 8304 4152 16,608 16,608
F
p
12.6 0.6 46.6 23.7 8.0 6.6 7.4 0.5 0.9 12.8 1.2 0.6 1.1 0.9 1.7
0.001 0.63 <0.001 <0.001 <0.001 0.007 0.002 0.77 0.50 <0.001 0.29 0.71 0.34 0.49 0.13
Degrees of freedom.
P3 of other electrodes were defined as the largest positive peak within ±50 ms of P3 at Pz. Peak amplitude (V) was calculated as the voltage difference between a component peak and a prestimulus baseline. Peak latency was defined as the time between stimulus onset and the peak maximum. P3 amplitudes and peak latencies were analyzed using repeatedmeasures analysis of variance to assess the effect by group (patients or controls), emotional condition (happy, neutral, sad, fear, or disgust), region (frontal, central, or parietal) and scalp electrode location (left, middle, or right hemisphere). Greenhouse–Geisser correction for non-sphericity was applied. Post hoc pairwise comparison was performed for within-group and between-group factors with Bonferroni corrections for multiple comparisons. P3 amplitude and peak latencies were collapsed across 9 electrode examining effects of emotional condition, group and interaction. Target detection and recognition rates of the facial emotions were analyzed with multivariate analysis of variance (MANOVA) between patients and controls. To evaluate the clinical variables and medication effect of patients, correlation analyses and Mann–Whitney U tests were performed. Unless indicated otherwise, data are presented as the mean ± SD. SPSS version 12.0 was used for these statistical analyses. There were no significant differences between the groups with regard to age, sex, IQ (109.1 ± 11.2, patients, 113.7 ± 9.9, controls) and duration of education (14.8 ± 1.7, patients, 15.7 ± 1.1, controls). Fifteen patients were on divalproex (serum concentration, 80.07 ± 17.00 g/mL), and the remaining five patients were on lithium (serum concentration, 0.55 ± 0.09 mmol/L). Nineteen patients had antipsychotics (531.8 ± 327.2 mg chlorpromazineequivalent dose). Grand averages of the ERP responses to different emotional conditions are demonstrated in Fig. 1b. For comparing P3 amplitudes between controls and patients, a repeated-measures group by emotional condition by region by hemisphere ANOVA for all subjects was done. It revealed group had a main effect (F(1,38) = 12.6, p < 0.001) as amplitudes in controls were larger than manic patients. It also showed significant interactions of group by emotional condition (F(4,152) = 8.0, p < 0.001). There was a main effect of region (F(2,76) = 46.6, p < 0.001) and a main effect of hemisphere (F(2,76) = 23.7, p < 0.001), which is compatible to P3 responses with highest at mid-parietal electrodes. There were significant interactions of region and group (F(2,76) = 6.6, p = 0.007), and of hemisphere and group (F(2,76) = 7.4, p = 0.002), in which amplitudes in the central and middle sites increased less in patients than controls. However, there were no other interaction effects (Table 1).
While the patient group did not show the main effect of emotional conditions (F(4,76) = 1.7, p = 0.17), the control group revealed the main effect of emotional conditions (F(4,76) = 8.89, p < 0.001). Post hoc comparisons between neutral and emotional conditions within the control group revealed a significant difference in P3 amplitudes of neutral-happy (F(1,19) = 5.6), neutral-sad (F(1,19) = 16.5, p = 0.001), neutral-fearful (F(1,19) = 13.8, p = 0.001), and neutral-disgust (F(1,19) = 9.5, p = 0.006) conditions, in which amplitudes of neutral condition were lower than those of other emotional conditions. To investigate the interaction effect for group by emotional condition, we used paired emotional conditions (i.e., a neutral condition and each emotional condition) as within-group factors. While happy-neutral condition did not have interactions of group by emotional condition (F(1,38) = 2.1, p = 0.15), sad-neutral condition (F(1,38) = 20.1, p < 0.001), fear-neutral (F(1,38) = 8.6, p = 0.006), and disgust-neutral condition (F(1,38) = 15.7, p < 0.001) had interactions of group by emotional conditions (Fig. 2). Post hoc t test revealed that P3 amplitudes in manic patients were significantly lower than control group for sad (t = −4.45, p < 0.001), fear (t = −3.57, p = 0.001), disgust (t = −3.72, p = 0.001), and neutral (t = −2.03, p = 0.05). Group (F(1,38) = 15.2, p < 0.001), hemisphere (F(2,76) = 7.1, p = 0.002) and region (F(2,72) = 12.1, p < 0.001) had main effects on P3 peak latencies. P3 latencies of patients were slower than those of controls. There was an interaction of group by region (F(2,76) = 4.9, p = 0.01) as well. However, we did not observe any significant main or interaction effect of emotion, group by emotion, or group by emotion by region. MANOVA for target detection rates revealed no differences between controls and patients (F(5,34) = 1.56, p = 0.20). MANOVA for recognition rates after EEG recordings revealed no group differences (F(5,34) = 1.81, p = 0.14). Recognition rates were 90.0 ± 23.5% (patients) and 92.5 ± 14.3% (controls) for happy faces, 61.3 ± 35.8% (patients) and 52.5 ± 32.3% (controls) for neutral faces, 66.5 ± 37.8% (patients) and 88.8 ± 22.2% (controls) for sad faces, 90.0 ± 23.5% (patients) and 92.5 ± 14.3% (controls) for fear faces, and 76.3 ± 31.9% (patients) and 67.5 ± 38.1% (controls) for disgust faces. Independent t test revealed a difference in the recognition rates for sad stimuli (t = 2.23, p = 0.03) between patients and controls, but no significant difference in the other emotions. Patients’ misrecognition rate of neutral stimuli as sad (5.0 ± 10.3%) was lower than that of controls (15.0 ± 17.0%) (t = 2.25, p = 0.03). Misrecognition rates of neutral stimuli as fear was lower in patients (3.8 ± 9.2%) than in controls (11.3 ± 12.8%) (t = −2.14, p = 0.04). Misrecognition rates of sad stimuli as neutral was higher in patients (15.0 ± 17.0%) than in controls (3.8 ± 9.2%) (t = 2.60, p = 0.01). There were no differences in other misrecognition rates between patients and controls. YMRS, BPRS, BDI and MADRS scores were not significantly correlated with P3 amplitudes at Pz electrode (p values ranged from 0.20 to 0.99). P3 amplitudes and peak latencies were not significantly correlated with chlorpromazine-equivalent doses at all electrode sites for all emotional conditions (p values from 0.11 to 0.66). There were also no differences between lithium use and divalproex use by Mann–Whitney U test in P3 amplitude and latency of Pz (p values from 0.31 to 0.97). This study investigated event-related P3 responses to facial emotional stimuli in a sample of manic patients and healthy controls. The analysis of group by neutral-emotional condition on P3 amplitudes revealed significant interaction effects in condition pairs of sad-neutral, disgust-neutral, fear-neutral conditions, but not happy-neutral conditions. Manic patients demonstrated significantly lower P3 amplitudes than controls for each emotional stimulus. While healthy subjects showed larger amplitudes to all kinds of emotional, especially negative faces compared to neu-
V. Ryu et al. / Neuroscience Letters 481 (2010) 92–96
95
Fig. 2. Interactions of groups by emotional conditions (neutral vs. emotional).
tral faces, manic patients showed no difference in P3 amplitudes between emotional faces and neutral faces. P3 amplitude indexes brain activity of comparing incoming stimuli with previous ones. The attentional process that detects updating of stimuli is concomitant with P3 responses. Therefore, our findings might be the results of impaired attention reallocation strategy, especially for negative emotion in manic patients. Larger P3 amplitudes elicited by negative emotions in our healthy subjects can be supported by previous studies. It has been reported that normal subjects show larger late positive potentials (LPPs) to emotional pictures than neutral ones and increaseed LPP amplitude with emotional arousal/intensity [10]. Furthermore, healthy people exhibit stronger sensitivity and larger LPPs to negative stimuli than neutral stimuli [17]. Our study demonstrated that manic patients showed no difference in P3 amplitudes between emotional stimuli and neutral stimuli, whereas healthy controls showed increased P3 amplitude to emotional stimuli compared to neutral stimuli. These findings suggest that emotional effects, larger amplitude responses to negative stimuli, may not work in manic state. Therefore, manic patients tend to underestimate or misperceive negative emotional stimuli. As described earlier, manic patients demonstrated poor processing for negative emotional stimuli [22,23]. Thus, the differential P3 responses to emotional valences observed in this study may be a neurophysiological evidence of impaired negative facial emotion processing in mania. Most studies have shown prolonged P3 latency in psychotic or non-psychotic manic patients [27,32]. All these studies used the auditory oddball paradigms to evaluate stimulus relevance. In the present study, a visual oddball paradigm with facial emotional stimuli was used to categorize the emotion as in the behavioral studies [22,23]. P3 latency, indexing classification speed, may be related with detection and evaluation of stimuli [28]. While our manic subjects showed prolonged P3 latencies compared to controls, there were no effects of emotional condition and group by emotional condition. This finding suggests that evaluation speed on facial expression may be delayed, irrespective of emotional valences, in manic status. Reduced P3 amplitudes for negative emotions might be explained by some functional magnetic imaging (fMRI) studies. Manic patients showed weak intensity rating of sad faces, which is associated with significant attenuation in the anterior cingulate
gyrus and amygdala [23]. Moreover, their corticolimbic regions with frontal and temporal cortices and limbic and paralimbic regions were underactivated during the explicit processing of sad faces [9]. Manic patients also showed increased blood oxygen leveldependent (BOLD) responses in the amygdala to positive affective stimuli, not to negative stimuli [6]. Generally, the model of neural activation of underlying P3 generation reveals that external stimuli are maintained and monitored by the frontal areas and anterior cingulate regions, and then they are transmitted to the temporal and parietal areas that produce P3 responses [7,28]. Thus, it seems that decreased activation of some cortical or corticolimbic areas during emotional processing is involved in reduced P3 amplitude to negative stimuli observed in this study [9,23]. In this study, we could not identify whether the differential P3 response to each emotion shown in manic state will be observed in euthymic state as enduring impairments. Behavioral studies on facial emotion perception in euthymic bipolar patients have shown inconsistent findings. Some have reported impairment in recognizing faces [8,16], but others have reported no difference in facial recognition between patients and controls [22,34]. We need to further investigate P3 responses to emotional face stimuli in euthymic patients. Our study has limitations. First, correct recognition rate of neutral faces in our study was low (61.3% in control, 52.5% in patients). In a study with 143 healthy volunteers, it was very low score in 7-point Likert scale that the recognition intensity of neutral face as other emotional faces (1.55 ± 0.78), while intensity scores of other emotions were high (5.66 ± 0.91 for happy, 3.16 ± 0.96 for sad, 4.90 ± 0.78 for fear, and 4.88 ± 1.08 for disgust) [13]. The relatively low emotional intensity might support the applicability of the comparison between emotional and neutral faces. Second, we found no significant interaction effects of group by happy-neutral condition. However, due to the small sample size, this study might not have enough power to distinguish differences between the two groups. Third, we found no statistical association of P3 amplitudes with neuroleptic dose or the type of mood stabilizers. It was reported that P3 waves are not affected by antipsychotic drugs [12]. However, we cannot completely exclude possible effects of drugs used in patients due to the small number of subjects. In summary, patients showed no P3 amplitude differences between negative and neutral faces and more reduced P3 ampli-
96
V. Ryu et al. / Neuroscience Letters 481 (2010) 92–96
tudes to negative facial emotions compared to normal subjects. These findings may give the neurophysiological evidence of impaired negative emotional processing, which is consistent with mood-congruent emotional bias in manic patients. Acknowledgements This study was supported by a faculty research grant (No. 6-2009-0118) of Yonsei University College of Medicine and the intramural grant of the Institute of Behavioral Science in Medicine, Yonsei University College of Medicine. References [1] American Psychiatric Association, Diagnostic and Statistical Manual of Mental Disorders, American Psychiatric Association, Washington, DC, 2000. [2] S.K. An, S.J. Lee, C.H. Lee, H.S. Cho, P.G. Lee, C.I. Lee, E. Lee, K.S. Roh, K. Namkoong, Reduced P3 amplitudes by negative facial emotional photographs in schizophrenia, Schizophr. Res. 64 (2003) 125–135. [3] M. Annett, A classification of hand preference by association analysis, Br. J. Psychol. 61 (1970) 303–321. [4] M. Asberg, S.A. Montgomery, C. Perris, D. Schalling, G. Sedvall, A comprehensive psychopathological rating scale, Acta Psychiatr. Scand. Suppl. (1978) 5–27. [5] A.T. Beck, C.H. Ward, M. Mendelson, J. Mock, J. Erbaugh, An inventory for measuring depression, Arch. Gen. Psychiatry 4 (1961) 561–571. [6] F. Bermpohl, U. Dalanay, T. Kahnt, B. Sajonz, H. Heimann, R. Ricken, M. Stoy, C. Hagele, F. Schlagenhauf, M. Adli, J. Wrase, A. Strohle, A. Heinz, M. Bauer, A preliminary study of increased amygdala activation to positive affective stimuli in mania, Bipolar Disord. 11 (2009) 70–75. [7] C. Bledowski, D. Prvulovic, K. Hoechstetter, M. Scherg, M. Wibral, R. Goebel, D.E. Linden, Localizing P300 generators in visual target and distractor processing: a combined event-related potential and functional magnetic resonance imaging study, J. Neurosci. 24 (2004) 9353–9360. [8] V.P. Bozikas, T. Tonia, K. Fokas, A. Karavatos, M.H. Kosmidis, Impaired emotion processing in remitted patients with bipolar disorder, J. Affect. Disord. 91 (2006) 53–56. [9] C.H. Chen, B. Lennox, R. Jacob, A. Calder, V. Lupson, R. Bisbrown-Chippendale, J. Suckling, E. Bullmore, Explicit and implicit facial affect recognition in manic and depressed States of bipolar disorder: a functional magnetic resonance imaging study, Biol. Psychiatry 59 (2006) 31–39. [10] S. Delplanque, L. Silvert, P. Hot, S. Rigoulot, H. Sequeira, Arousal and valence effects on event-related P3a and P3b during emotional categorization, Int. J. Psychophysiol. 60 (2006) 315–322. [11] M. Eimer, A. Holmes, Event-related brain potential correlates of emotional face processing, Neuropsychologia 45 (2007) 15–31. [12] J.M. Ford, P.M. White, J.G. Csernansky, W.O. Faustman, W.T. Roth, A. Pfefferbaum, ERPs in schizophrenia: effects of antipsychotic medication, Biol. Psychiatry 36 (1994) 153–170. [13] R.Y. Ha, J.I. Kang, S.K. An, H.S. Cho, Some psychological correlates affecting recognition of neutral facial emotion in young adults, J. Korean Neuropsychiatr. Assoc. 48 (2009) 481–487. [14] M.H. Hall, F. Rijsdijk, S. Kalidindi, K. Schulze, E. Kravariti, F. Kane, P. Sham, E. Bramon, R.M. Murray, Genetic overlap between bipolar illness and event-related potentials, Psychol. Med. 37 (2007) 667–678.
[15] M.H. Hall, K. Schulze, F. Rijsdijk, S. Kalidindi, C. McDonald, E. Bramon, R.M. Murray, P. Sham, Are auditory P300 and duration MMN heritable and putative endophenotypes of psychotic bipolar disorder? A maudsley bipolar twin and family study, Psychol. Med. 39 (2009) 1277–1287. [16] C.J. Harmer, L. Grayson, G.M. Goodwin, Enhanced recognition of disgust in bipolar illness, Biol. Psychiatry 51 (2002) 298–304. [17] T.A. Ito, J.T. Larsen, N.K. Smith, J.T. Cacioppo, Negative information weighs more heavily on the brain: the negativity bias in evaluative categorizations, J. Pers. Soc. Psychol. 75 (1998) 887–900. [18] R. Johnson Jr., A triarchic model of P300 amplitude, Psychophysiology 23 (1986) 367–384. [19] N. Kerr, R.I. Dunbar, R.P. Bentall, Theory of mind deficits in bipolar affective disorder, J. Affect. Disord. 73 (2003) 253–259. [20] E. Kim, J. Ku, J.J. Kim, H. Lee, K. Han, S.I. Kim, H.S. Cho, Nonverbal social behaviors of patients with bipolar mania during interactions with virtual humans, J. Nerv. Ment. Dis. 197 (2009) 412–418. [21] A. Kok, On the utility of P3 amplitude as a measure of processing capacity, Psychophysiology 38 (2001) 557–577. [22] A. Lembke, T.A. Ketter, Impaired recognition of facial emotion in mania, Am. J. Psychiatry 159 (2002) 302–304. [23] B.R. Lennox, R. Jacob, A.J. Calder, V. Lupson, E.T. Bullmore, Behavioural and neurocognitive responses to sad facial affect are attenuated in patients with mania, Psychol. Med. 34 (2004) 795–802. [24] H.M. Lyon, M. Startup, R.P. Bentall, Social cognition and the manic defense: attributions, selective attention, and self-schema in bipolar affective disorder, J. Abnorm. Psychol. 108 (1999) 273–282. [25] W. Mansell, R. Pedley, The ascent into mania: a review of psychological processes associated with the development of manic symptoms, Clin. Psychol. Rev. 28 (2008) 494–520. [26] D. Matsumoto, P. Ekman, Japanese and Caucasian Facial Expressions of Emotion (JACFEE) Slides, San Francisco, CA, 1988. [27] B.F. O’Donnell, J.L. Vohs, W.P. Hetrick, C.A. Carroll, A. Shekhar, Auditory eventrelated potential abnormalities in bipolar disorder and schizophrenia, Int. J. Psychophysiol. 53 (2004) 45–55. [28] J. Polich, Updating P300: an integrative theory of P3a and P3b, Clin. Neurophysiol. 118 (2007) 2128–2148. [29] K.K. Schulze, M.H. Hall, C. McDonald, N. Marshall, M. Walshe, R.M. Murray, E. Bramon, Auditory P300 in patients with bipolar disorder and their unaffected relatives, Bipolar Disord. 10 (2008) 377–386. [30] H.V. Semlitsch, P. Anderer, P. Schuster, O. Presslich, A solution for reliable and valid reduction of ocular artifacts, applied to the P300 ERP, Psychophysiology 23 (1986) 695–703. [31] D.V Sheehan, Y. Lecrubier, K.H. Sheehan, P. Amorim, J. Janavs, E. Weiller, T. Hergueta, R. Baker, G.C. Dunbar, The Mini-International Neuropsychiatric Interview (M.I.N.I.): the development and validation of a structured diagnostic psychiatric interview for DSM-IV and ICD-10, J. Clin. Psychiatry 59 (Suppl. 20) (1998) 22–33, quiz 34–57. [32] W.K. Strik, M. Ruchsow, S. Abele, A.J. Fallgatter, T.J. Mueller, Distinct neurophysiological mechanisms for manic and cycloid psychoses: evidence from a P300 study on manic patients, Acta Psychiatr. Scand. 98 (1998) 459–466. [33] B.I. Turetsky, C.G. Kohler, T. Indersmitten, M.T. Bhati, D. Charbonnier, R.C. Gur, Facial emotion recognition in schizophrenia: when and why does it go awry? Schizophr. Res. 94 (2007) 253–263. [34] H.R. Venn, J.M. Gray, B. Montagne, L.K. Murray, D. Michael Burt, E. Frigerio, D.I. Perrett, A.H. Young, Perception of facial expressions of emotion in bipolar disorder, Bipolar Disord. 6 (2004) 286–293. [35] R.C. Young, J.T. Biggs, V.E. Ziegler, D.A. Meyer, A rating scale for mania: reliability, validity and sensitivity, Br. J. Psychiatry 133 (1978) 429–435.