B R A I N R E S E A R C H 1 1 2 2 ( 2 0 0 6 ) 1 7 1 – 1 7 8

ava i l ab l e a t www.sc i enced i r ec t . com

www.e l sev i e r. com/ l oca te /b ra in res

Research Report

An ERP investigation of emotional processing in European andJapanese individuals☆

Pascal Hota,⁎, Yasuhiko Saitob, Osamu Mandai b,Toshinori Kobayashi b, Henrique Sequeiraa

aUniversité de Lille I and Neurosciences Fonctionnelles et Pathologies, CNRS, Hôpital Salengro CHRU Lille 59037 Lille cedex, FrancebSleep Research Center, Ashikaga Institute of Technology, 268-1, Omae-cho, Ashikaga-shi, Tochigi 326-8558, Japan

A R T I C L E I N F O

☆ The present work was performed with no⁎ Corresponding author. Fax: +33 3 2044 6732.E-mail address: kutaifr@yahoo.fr (P. Hot).

0006-8993/$ – see front matter © 2006 Elsevidoi:10.1016/j.brainres.2006.09.020

A B S T R A C T

Article history:Accepted 5 September 2006Available online 10 October 2006

This article examined neural time course differences in the processing of emotional picturesin European and Asian individuals. Event-related potentials (ERPs) were recorded in 15French and 15 Japanese volunteers during the presentation of neutral and emotionalpictures in their own country. ERPs were analyzed by means of spatio-temporal PrincipalComponent Analysis (PCA) and allow to evaluate cultural influences on emotionalprocessing along three temporal windows: 105 to 140 ms, 176 to 230 ms and 255 to455 ms. Main results highlight that (1) early ERP components coded emotional charge butwere not modulated by cultural background, (2) later components showed a significantdecrease of amplitudes at parieto-occipital areas for Japanese participants compared toFrench ones in emotional condition. These findings suggest that, in both populations,similar neurocognitive processes are involved in the early stage of the emotional stimuliprocessing and effects observed on later components may reflect a poorer engagement ofparietal areas, known to be involved in emotional arousal dimension. Considering thatcognitive judgments were similar in both populations, electrophysiological findings suggestthat cultural influences on later stage of emotional processing could be related to the knownlower level of emotionally expressive behavior in Japanese than in Caucasians.

© 2006 Elsevier B.V. All rights reserved.

Keywords:EmotionEvent-related potentialCultural differenceValenceArousalSpatio-temporal PCA

1. Introduction

Numerous compelling studies pinpoint that cultural environ-ment plays a major role in orchestrating cognitive andaffective processes across many levels of influence (Li, 2003).In particular, emotional responses are modulated by social,educational and cultural factors (Ekman, 1972; Scherer andWallbott, 1994). Among cultural idiosyncrasies, several studiesparticularly support that Japanese, compared to Caucasians,are less emotionally expressive (Kitayama et al., 2000),

conflicts of interest and w

er B.V. All rights reserved

probably because they engage specific neural networks forthe processing of emotional stimuli (Moriguchi et al., 2005;Russell et al., 1993; Tsunoda, 1985). During one decade,Tsunoda (1985) produced an impressive work supporting thehypothesis that, during the childhood, the cultural back-ground of Japanese would significantly and durably modulatebrain processing of external emotional stimuli. Additionalfindings have been recently provided by neuroimagingstudies. Thus, Moriguchi et al. (2005) observed that Japaneseengage different brain structures compared to Caucasians to

ith no significant financial interest/other relationship to disclose.

.

172 B R A I N R E S E A R C H 1 1 2 2 ( 2 0 0 6 ) 1 7 1 – 1 7 8

process emotional stimuli. Whereas Caucasians engaged aclassical emotional neural network including the amygdalain the processing of fearful faces, Japanese responded in aless emotional way involving right inferior frontal area andother brain structures. However, this study did not allow toprecise how the time course of emotional processing differsbetween studied populations. Consequently, the main un-solved issue is to define if the well-described lesser emotion-al expression in Japanese corresponds to specific time courseof neural processing of expression, perception or feeling ofemotions. To unravel this issue, the Event-Related Potential(ERP) study can provide relevant findings. First, ERP compo-nents indexed neuronal changes associated to cognitiveprocesses (Rugg and Coles, 1995). Second, the high temporal

Fig. 1 – Grand average ERPs elicited in response to unpleasant, pl(A) and Japanese (B) subjects. Time courses: the red lines refer tocondition, while thin grey lines stand for neutral condition. Majofrom greater amplitudes for components elicited after 200 ms.

resolution of electrophysiological markers can be advanta-geously employed to compare the time course of emotionalprocesses. Numerous previous studies have confirmed therelevance of ERPs to index the processing of valence andarousal dimensions of emotions (e.g. Cuthbert et al., 2000;Delplanque et al., 2004, 2005, 2006; Diedrich et al., 1997;Dolcos and Cabeza, 2002; Schupp et al., 2000). It results thatdifferences in emotional processing from one cultural groupcompared to another one could be associated with additionalor modulations of ERPs components. In this frame, one study(Skrandies et al., 2004) has examined cultural influences(Chinese vs. German) on ERPs elicited by emotional words.However, one can wonder whether words, as visual stimuli,are not phylogenetically relevant to the same degree as

easant and neutral conditions at 11 electrodes sites, in Frenchunpleasant condition; the thick blue lines stand for pleasantr emotional effects seem to occur on parietal sites and stem

Fig. 1 (continued).

173B R A I N R E S E A R C H 1 1 2 2 ( 2 0 0 6 ) 1 7 1 – 1 7 8

emotional pictorial stimuli (Öhman and Mineka, 2001).Consequently, the purpose of the present study is todetermine whether differences in cultural background be-tween French and Japanese populations can be related to adifferential neural signature of emotional processing.

1 Japanese: valence dimension: U=3.04, N=4.75, P=6.54; activa-tion dimension: U=5.46, N=3.42, P=5.02. French: valence dimen-sion: U=2.98, N=5.07, P=6.34; activation dimension: U=5.34,N=3.74, P=5.41.

2. Results

2.1. Behavioral and voltage data

The analysis of affective judgment given by each subject oneach pre-selected picture after the experimental phaseshowed that all pictures were correctly assigned to the rightcategory set for all subjects, regardless of cultural group. Inaddition, mean valence and activation of the three emotionalcategory (U: unpleasant, N: neutral, P: pleasant)1 did not

reveal any significant group*emotional category difference(Valence: F(2,56)=0.53, NS; Activation: F(2.56)=1.27, NS). Fig. 1illustrates the grand average ERPs for the 30 French andJapanese participants at 11 selected electrodes sites, inresponse to unpleasant, pleasant and neutral pictures. Thevisual inspection of waveforms stresses differential brainactivity, located on the parietal late positive component (LPC)between emotional and neutral condition in both popula-tions. Unpleasant and pleasant stimuli elicited greateramplitude of the positive component, which appears around350 ms. No clear difference seems to appear between the twoemotional conditions. Cultural impact on affective process

Fig. 2 – Spatial and temporal factors extracted by the Principal Component Analysis. (A) Topographic maps of the factorloadings for the virtual electrodes: SF1, fronto-parietal areas; SF2, parieto-occipital areas; SF3, frontal areas. Percentage of totalvariance explained by each factor is indicated and the maps are viewed from above. Colored scale indicates correlation levelsbetween initial electrodes and virtual ones. (B) Loadings as a function of time for the rotated factors. The percentage of totalvariance explained by each factor is also indicated.

174 B R A I N R E S E A R C H 1 1 2 2 ( 2 0 0 6 ) 1 7 1 – 1 7 8

could occur on frontal electrodes, in particular for theunpleasant condition. In summary, classical analysis of ourdata seems to confirm the hypothesis of cultural modulationsof emotional process, probably on LPC. These observationsrequire yet statistical confirmations which we are going todevelop below.

2.2. Spatio-temporal PCA

The spatial Principal Component Analysis (PCA) performedon electrode sites yielded 17 spatial factors (SF) or virtualelectrodes, which describe spatial variance in the data set.The number of selected SF is determined by two criteria: thetotal number of factors must account for at least 90% of thevariance, each factor must account for 70% of the variance inone of the original variables (electrodes). Such criteria allowto identify, in the current study, three SF (Fig. 2A) whichhave the highest contribution to the spatial variance (SF1:64%; SF2: 21%; SF3: 14%). SF1 corresponds to a fronto-central(FC) activity (Fz, F4, Fc4, C3/4, Cz, Cp4, T8) and SF2corresponds to a bilateral parieto-occipital (PO) activity (P3/4, Pz, Tp7/8, O1/2). The last factor is quite restricted to afrontal area (F; Fp1/2, F8).

The temporal PCA reduced temporal dimensions of thedata set from 225 time-points to 5 temporal factors (TF, seeFig. 2B). TF1 (21% of the temporal variance) and TF3 (17%)were excluded from the further analysis because theirextractions are related to the auto-correlation of the data(see Spencer et al., 1999 for more details). TF2 (19%) has highloadings on a large epoch between 255 and 455 ms. Inaddition, temporal PCA detected two earlier short epochs(TF4 (18%): 176–230 ms; TF5(10%): 105–140 ms).

On the earliest window (105 to 140 ms), a mainemotional effect appears only at FC and PO areas (respec-tively F(2,56) =3.93, p<0.05 and F(2,56) =6.63, p<0.05) andstems from an arousal emotional effect on both areas(quadratic trend: FC, F(1,28)=4.27, p<0.05; PO, F(1,28)=4.65,p<0.05). During the second epoch (176 to 230 ms), asignificant emotional effect appears on frontal areas(respectively FC: F(2,56)=3.98, p<0.05 and F: F(2,56) =3.74,p<0.05) and corresponds to an arousal emotional effect(quadratic trend respectively for FC and F: F(1,14)=5.98,p<0.05 and F(1,14)=9.33, p<0.01). In addition, an interactionbetween groups and emotional conditions is observed onPO area (F(2,56)=3.74, p<0.05) and stems from the fact thatemotional effects are restricted to French group (French

Fig. 3 – Illustration of temporal/spatial factor scores related to emotional conditions for the 255–455 ms window, in bothpopulations. (A) Mean amplitude in each emotional condition on parieto-occipital and frontal areas. Emotional impact on ERPselicited on frontal and parieto-occipital sites did not differ whereas a valence effect occurs at frontal site and stems from greateramplitude in unpleasant condition compared to pleasant one. (B) Greater ERPs in emotional conditions compared to neutralone occur only for the French group. Note that the unitless of Y axis is due to the combination of temporal/spatial factors.

175B R A I N R E S E A R C H 1 1 2 2 ( 2 0 0 6 ) 1 7 1 – 1 7 8

population: F(2,28) = 5.68, p<0.05; Japanese population:F(2,28)=0.53, NS). On the late temporal window (255 to 455 ms;Fig. 3A), an emotional sensitivity appears at F and PO areas(respectively F(2,56)=10.01, p<0.001; F(2,56)=6.76, p<0.01) whereemotional pictures elicit greater components (quadratic trendsfor F and PO: F(1,28)=12.25, p<0.01, F(1,28)=4.93, p<0.05). Inaddition, F area is sensitive to valence effect (linear trend:F(1,28)=9.94, p<0.01). Moreover, a group*condition interactionis significant at PO area (F(2,56)=7.33, p<0.001) and stems fromthe fact that the emotional arousal effect is restricted to Frenchpopulation (Fig. 3B; F(1,14)=20.44, p<0.001 for French group andF(1,14)=1.14, NS for Japanese group).

3. Discussion

The main objective of the current study was to unravel ifdifferences in cultural background between French andJapanese populations could be related to a differential neuralsignature of emotional processing. Thus, our findings showedinteractions between cultural background and emotionalprocesses. Our main finding yielded cultural differences inemotional ERPs, between French and Japanese populations,topographically restricted to parieto-occipital componentsoccurring after 170 ms.

As previously described (Schupp et al., 2003), the earliestcomponent sensitive to emotional dimension appears be-tween 100 and 150 ms. The scalp distribution suggests thatthis component corresponds to the P1/N1 complex whichindexes early stages of attentional orienting (Hopf and Man-gun, 2000). Thus, our results would indicate that earlyattentional resource allocations to the emotional stimuliindexed by P1/N1 (Keil et al., 2001; Schupp et al., 2000) arenot affected by cultural factors. On the contrary, cultural

modulations of emotional ERPs were observed on the compo-nents elicited beyond 170 ms after the stimulus onset andparticularly in the 250–450 ms window.

Thecomponentselicitedinthiswindowhadbeenpreviouslydescribed as a LPC probably (i) to be distinguished from P3component which generally refers to a positive componenttypically elicited through the oddball paradigm and thought toreflect specific cognitive processes like immediate memorymechanisms (DonchinandColes, 1988), (ii) topinpoint that thiscomponent occurs generally in a larger and later temporalwindow than the previous one. According to Palomba et al.(1997) and Cuthbert et al. (2000), greater LPCs in emotionalcondition reflect a greater cortical engagement associated to adeeper and more complete processing of emotional stimuli.CorrelationanalysesbetweenamplitudeofLPCsandbehavioraland autonomic indices confirm such a hypothesis. Indeed,Cuthbert et al. (2000)demonstrated that subject's ratingson thearousal dimension of emotional stimuli are positively correlat-ed with the amplitude of LPCs components. Moreover, emo-tional arousing stimuli that elicited highest amplitude valuesalso prompted increased electrodermal responses amplitudes,an autonomic choice index of emotional arousal (Hot et al.,2005). Consequently, our main findings, clearly related to thislatetemporalwindow,suggestthatcorticalactivity,knowntobeassociated with emotional arousal was reduced in Japanesesubjects.

Considering temporal windows pointed out, we assumethat cultural factors clearly modulate neural time course ofemotional processing. ERP differences between both popula-tions did not involve additional components or delayedlatency but topographical differences, which means that dif-ferential neural generators are involved during emotionalprocessing in French and Japanese individuals. Consideringthat spatial factors extracted by PCA are orthogonal (i.e.

176 B R A I N R E S E A R C H 1 1 2 2 ( 2 0 0 6 ) 1 7 1 – 1 7 8

activity on each spatial area is independent of the others),this study provides indirect evidence of a neural activity de-crease at parieto-occipital area, which can been associated tothe arousal dimension of emotions as proposed by theHeller's (1993) model. This model lied on dimensionalapproach of emotions (valence, arousal) and supports thatvalence dimension preferentially engages frontal areas (seealso Davidson, 2004) whereas arousal dimension, associatedto behavioral and autonomic expressions, mobilizes corticalparietal areas.

Consequently, this study did not afford to unravel sourcelocalization, but raises the question of a putative less parietalareas engagement. The lack of significant difference inaffective judgment between French and Japanese groupsprovides interesting lead in favor of this hypothesis. Since itcould be assumed that the both groups obtained comparablevalence and arousal ratings, late ERPs components areprobably more related to processes controlling descendingpatterns of emotional behavioral expression. To furtherinvestigate this hypothesis, we currently develop a newprotocol including EEG sources localization and autonomicmeasures. However, the current study is the first contribution,which reports a brain electrophysiological signature of cul-tural specificity in emotional processing. Finally, in a globalcross-cultural perspective, the current study contributes tofurther understand appraisal mechanisms in emotion withrespect to social norms and values (Sander et al., 2005).

4. Experimental procedures

4.1. Subjects

Fifteen French right-handed subjects (6 males) and fifteenJapanese right-handed subjects (8males) aged between 18 and26 years (mean 22.6±3.3 years) living in their own nativecountry were included in the study. All subjects were healthyand had no history of neurological disorder, or drug abuse, andnone was being treated with medications at the time of thestudy. They gave their written consent to the study afterdetailed informationwas provided to them, and the study wasdone in-line with the Declaration of Helsinki followingapproval by the Regional Ethics Committees (France, Japan).

4.2. Materials

Stimuli consisted in ninety pictures taken from the Interna-tional Affective Picture System (IAPS; Lang et al., 1997) tocompose three groups of 30 pictures2 (unpleasant, U; neutral,N; pleasant, P). IAPS pictures are rated on two 9-point scalesfor valence (1=very unpleasant to 9=very pleasant) and

2 Unpleasant set: 1120, 1200, 1220, 1300, 2730, 2750, 2800, 2900,3053, 3062, 3063, 3071, 3080, 3102, 3110, 3140, 3150, 3170, 3250,3280, 3350, 6250, 6370, 6570, 9008, 9042, 9140, 9300, 9570, 9571;Pleasant set: 1440, 1560, 1610, 1720, 1811, 2000, 2057, 2311, 4001,4005, 4180, 4232, 4275, 4310, 4611, 4652, 4659, 4664, 4669, 4670,4680, 5260, 5480, 5626, 5780, 7281, 7283, 7285, 7325, 7340; Neutralset: 2840, 2850, 5130, 5300, 5390, 5900, 7000, 7002, 7004, 7006, 7009,7010, 7025, 7034, 7035, 7050, 7080, 7090, 7100, 7130, 7140, 7160,7175, 7182, 7187, 7190, 7205, 7224, 7233, 7235.

arousal (from 1=calmed to 9=aroused). Our pictures wereselected in such a way they differed significantly in thevalence dimension (means: U=2.75, N=4.86, P=7.00; linearcontrast F(2,87)=486, p<0.001). In addition, emotional picturesdiffered from neutral ones in arousal dimension (means:U=5.87, N=2.97, P=5.62; quadratic trend: F(2,87)=47, p<0.001).The validity of emotional ratings for selected IAPS pictures hasbeen confirmed in a previous investigation on 35 French and25 Japanese participants.

4.3. Procedure

The experiment took place in a sound attenuated, electricallyshielded, dimly lighted room. Each subject performed anidentical standardized protocol under the control of the firstauthor. All pictures were randomly presented centrally on acomputer screen for 1.5 s each, occupying about 15° ofhorizontal visual angle, with an ISI which varied randomlybetween 3 and 5 s. During the ISI, subjects were told tomaintain gaze on the white centered cross, occurring betweenthe picture presentations, to avoid blinking and to move theireyes as less as possible during the stimuli presentations. Thesubjects were instructed only to look at the pictures and to tryto experience feelings which could be induced by pictures.After the experimental phase, each picture was presentedagain during 6 s and subjects had to give an affective judgmentfor each one.

4.4. ERP recording

Electroencephalographic (EEG) activity was recorded at 25electrode sites of the extended 10–20 system (Nuwer et al.,1998, FP1/2, F3/4, F7/8, FZ, FC3/4, C3/4, CZ, CP3/4, P3/4, PZ, O1/2,T7/8, TP7/8, P7/8), using tin electrodes inserted in an elasticnylon cap (Electrocap International) referenced to linked-earlobes with a forehead ground. Four additional electrodeswere placed at the outer canthus and supraorbitally to theright eye with a bipolar recording for electro-ocular activity.The impedance for all the electrode sites was kept below 5 kΩ.The EEG filter bandpasswas 0.016 to 70 Hz andwas digitized at300 Hz in continuous. Eye-movement artifacts were correctedfrom the EEG by a dynamic regression analysis in thefrequency domain (Woestenburg et al., 1983). Remaining trialswith artifacts exceeding ±100 μV were excluded from theanalysis in all channels.

4.5. Spatio-temporal analyses

Data reductions by means of a spatio-temporal PCA wereconducted for two reasons. First, this analysis is particularlyrelevant in this kind of paradigm to disentangle overlappingportions of ERP waveforms (see Spencer et al., 1999, 2001 forfurther details). Second, although EEG recording was per-formed in similar conditions, differences in materials sensi-tivity cannot be excluded. Since PCA was performed oncorrelation scores, data recorded by French and Japanesesystems are consequently transformed in a similar scale ofvalues.

This analysis consists in performing multiple PCA on ERPsto reduce its dimensionality and disentangle overlapping

177B R A I N R E S E A R C H 1 1 2 2 ( 2 0 0 6 ) 1 7 1 – 1 7 8

components. The methodological analysis is reported indetails in Spencer et al. (2001). Authors underlined that theuse of PCA allows capturing the relevant variance in neworthogonal factors.

In a first step, a spatial PCA is performed with electrodesites (25) as variables and time-points, subjects and conditionsas observations. Spatial PCA identifies groups of electrodes,which are highly correlated and replaces them by a reducednumber of linear combinations. Each new extracted factor canbe considered as one specific spatial configuration of brainactivation, called by the same authors “virtual electrode”(Spencer et al., 2001). The virtual electrode can be visualized bya topographic map of the factor scores, which correspond tothe correlation between original data and the new factors.Factors that fit better the global spatial variance are consid-ered as « virtual ERPs » and can be represented in temporalsets for each subject and each condition. In a second step,these times series (225 points) were submitted to a similar PCAwith new spatial factors, using subjects and conditions asobservations and reduced in a new space of several temporalfactors.

4.6. Statistical analyses

Effects of experimental conditions are statistically prospectedby themeans of GLM procedure with a group factor (French vs.Japan group), and a repeated measure factor performed oneach conjunction between « spatial » and « temporal » factors.The Greenhouse–Geisser ε correction was employed to adjustthe degrees of freedom of the F ratios. Significant level was setat 0.05 (two-tailed). Linear and quadratic contrasts are used toevaluate respectively valence (unpleasant vs. pleasant condi-tion) and arousal (emotional vs. neutral) effects.

Acknowledgments

This work was supported by grants from the Sleep ResearchCenter (Ashikaga Institute of Technology, Japan) and theCentre International de Lille I (France). P.H. was granted by thepostdoctoral fellowship of the Egide/Ministère des AffairesEtrangères (France). Special thanks go to V. Matuszewski forher useful suggestions.

R E F E R E N C E S

Cuthbert, B.N., Schupp, H.T., Bradley, M.M., Birbaumer, N., Lang,P.J., 2000. Brain potentials in affective picture processing:covariation with autonomic arousal and affective report. Biol.Psychol. 52, 95–111.

Davidson, R.J., 2004. What does the prefrontal cortex “do” in affect:perspectives on frontal EEG asymmetry research. Biol. Psychol.67, 219–233.

Delplanque, S., Lavoie, M.E., Hot, P., Silvert, L., Sequeira, H., 2004.Modulation of cognitive processing by emotional valencestudied through event-related potentials in humans. Neurosci.Lett. 356, 1–4.

Delplanque, S., Silvert, L., Hot, P., Sequeira, H., 2005. Event-relatedP3a and P3b in response to unpredictable emotional stimuli.Biol. Psychol. 68, 107–120.

Delplanque, S., Silvert, L., Hot, P., Rigoulot, S., Sequeira, H., 2006.Arousal and valence effects on event-related P3a and P3bduring emotional categorization. Int. J. Psychophysiol. 60,315–322.

Diedrich, O., Naumann, E., Maier, S., Becker, G., Bartussek, D., 1997.A frontal positive slow wave in the ERP associated withemotional slides. J. Psychophysiol. 11, 71–84.

Dolcos, F., Cabeza, R., 2002. Event-related potentials of emotionalmemory: encoding pleasant, unpleasant, and neutral pictures.Cogn. Affect. Behav. Neurosci. 2, 252–263.

Donchin, E., Coles, M.G.H., 1988. Is the P300 component amanifestation of context updating? Behav. Brain Sci. 11,357–374.

Ekman, P., 1972. Universals and cultural differences in facialexpressions of emotion. In: Cole, J. (Ed.), Nebraska Symposiumon Motivation. University of Nebraska Press,Lincoln, pp. 207–282.

Heller, W., 1993. Neuropsychological mechanisms of individualdifferences in emotion, personality, and arousal.Neuropsychology 7, 476–489.

Hopf, J.M., Mangun, G.R., 2000. Shifting visual attention in space:an electrophysiological analysis using high spatial resolutionmapping. Clin. Neurophysiol. 111, 1241–1257.

Hot, P., Leconte, P., Sequeira, H., 2005. Diurnal autonomicvariations and emotional reactivity. Biol. Psychol. 69, 261–270.

Keil, A., Muller, M.M., Gruber, T., Wienbruch, C., Stolarova,M., Elbert, T., 2001. Effects of emotional arousal in thecerebral hemispheres: a study of oscillatory brain activityand event-related potentials. Clin. Neurophysiol. 112,2057–2068.

Kitayama, S., Markus, H.R., Kurokawa, M., 2000. Culture, emotion,and well-being: good feelings in Japan and the United States.Cogn. Emot. 14, 93–124.

Lang, P.J., Bradley, M.M., Cuthbert, B.N., 1997. InternationalAffective Picture System [pictures]. NIMH Center for the Studyof Emotion and Attention, Gainesville.

Li, S.C., 2003. Biocultural orchestration of developmental plasticityacross levels: the interplay of biology and culture in shapingthe mind and behavior across the life span. Psychol. Bull. 129,171–194.

Moriguchi, Y., Ohnishi, T., Kawachi, T., Mori, T., Hirakata, M.,Yamada, M., Matsuda, H., Komaki, G., 2005. Specific brainactivation in Japanese and Caucasian people to fearful faces.NeuroReport 16, 133–136.

Nuwer, M.R., Comi, G., Emerson, R., Fuglsang-Frederiksen, A.,Guerit, J.M., Hinrichs, H., Ikeda, A., Luccas, F.J., Rappelsburger,P., 1998. IFCN standards for digital recording of clinical EEG.International Federation of Clinical Neurophysiology.Electroencephalogr. Clin. Neurophysiol. 106, 259–261.

Öhman, A., Mineka, S., 2001. Fears, phobias, and preparedness:toward and evolved module of fear and fear learning. Psychol.Rev. 108, 483–522.

Palomba, D., Angrilli, A., Mini, A., 1997. Visual evoked potentials,heart rate responses and memory to emotional pictorialstimuli. Int. J. Psychophysiol. 27, 55–67.

Rugg, M.D., Coles, M.G.H., 1995. The ERP and cognitive psychology:conceptual issues. In: Rugg, M.D., Coles, G.H. (Eds.),Electrophysiology of Mind: Event-related Brain Potentials andCognition. Oxford Univ. Press, Oxford, pp. 27–39.

Russell, J., Suzuki, N., Ishida, N., 1993. Canadian, Greek andJapanese freely produced emotion labels for facial expressions.Motiv. Emot. 17, 337–351.

Sander, D., Grandjean, D., Scherer, K.R., 2005. A systemsapproach to appraisal mechanisms in emotion. Neural Netw.18, 317–352.

Scherer, K.R., Wallbott, H.G., 1994. Evidence for universality andcultural variation of differential emotion response patterning.J. Pers. Soc. Psychol. 66, 310–328.

Schupp, H.T., Cuthbert, B.N., Bradley, M.M., Cacioppo, J.T., Ito, T.,

178 B R A I N R E S E A R C H 1 1 2 2 ( 2 0 0 6 ) 1 7 1 – 1 7 8

Lang, P.J., 2000. Affective picture processing: the late positivepotential is modulated by motivational relevance.Psychophysiology 37, 257–261.

Schupp, H.T., Junghofer, M., Weike, A.I., Hamm, A.O., 2003.Emotional facilitation of sensory processing in the visualcortex. Psychol. Sci. 14, 7–13.

Skrandies, W., Chiu, M.J., Lin, Y., 2004. The processing of semanticmeaning in Chinese words and evoked brain topography. BrainTopogr. 16, 255–259.

Spencer, K.M., Dien, J., Donchin, E., 1999. A componential analysis

of the ERP elicited by novel events using a dense electrodearray. Psychophysiology 36, 409–414.

Spencer, K.M., Dien, J., Donchin, E., 2001. Spatiotemporal analysisof the late ERP responses to deviant stimuli. Psychophysiology38, 343–358.

Tsunoda, T., 1985. The Japanese Brain: Uniqueness andUniversality. Taishukan Publishing Company, Tokyo.

Woestenburg, J.C., Verbaten, M.N., Slangen, J.L., 1983. The removalof the eye-movement artifact from the EEG by regressionanalysis in the frequency domain. Biol. Psychol. 16, 127–147.