Influence of the Serotonin Transporter Promoter Gene and Shyness on Children’s Cerebral Responses...

ORIGINAL ARTICLE

Influence of the Serotonin Transporter PromoterGene and Shyness on Children’s CerebralResponses to Facial ExpressionsMarco Battaglia, MD; Anna Ogliari, MD; Annalisa Zanoni, MSc; Alessandra Citterio, MSc;Uberto Pozzoli, PhD; Roberto Giorda, PhD; Cesare Maffei, MD; Cecilia Marino, MD, PhD

Background: Childhood shyness can predate social anxi-ety disorder and may be associated with biased discrimi-nation of facial expressions of emotions.

Objective: To determine whether childhood shyness, orthe serotonin transporter promoter polymorphism geno-type, can predict participants’ visual event-related poten-tials in response to expressions of children of similar ages.

Design: Study group drawn from an inception cohortof 149 subjects characterized 1 year before the presentstudy by their degree of shyness.

Setting: Third- and fourth-grade schoolchildren.

Participants: Forty-nine of the inception cohort chil-dren, randomly selected.

Main Outcome Measures: Latencies and amplitudesof the N400 waveform in response to happy, neutral, andangry expressions.

Results: Shynesspredictedsignificantly smallerN400am-plitudes in response to anger (at Pz: P�.04) and to a neu-tral expression (at Pz: P�.047). Shyness was significantlydifferent across the 3 genotypes, the SS genotype being as-sociated with higher shyness levels (analysis of variance:F2,42=4.47,P�.02;Tukeyhonestlysignificantdifference,SSvs LL, P�.01). An analysis of covariance showed that nei-therthetypeofexpressionnorthegenotypeperseinfluencedtheN400amplitudes,butasignificantexpression�genotypeinteractionwas found(F4,72=3.57,P�.01), sustainedbythedifference inamplitudeof theSSandScarrier subjectscom-pared with the LL subjects when exposed to the anger ex-pression (Tukey honestly significant difference, P�.02).

Conclusion: Children who manifest higher levels of shy-ness or have 1 or 2 copies of the short allele of the sero-tonin transporter promoter gene appear to have a differ-ent pattern of processing affective stimuli of interpersonalhostility.

Arch Gen Psychiatry. 2005;62:85-94

N EUROIMAGING STUDIES ARE

beginningtoclarify there-lationship between thebrain’scorticalandsubcor-tical activity in regulating

theemotionalandcognitive functionsofbe-havior.Weknowthat theorbitofrontal,pre-frontal, insular, temporal,cingulate,parietal,and occipital cortices are important neuralnetworknodesinhumananxiety,1-3 andthatthe amygdala plays a pivotal role in the per-ception and evocation of emotions relatedto threatanddanger, suchas fearoranger.3,4

The proportion of cortical and subcorticalactivation differs depending on the type ofemotionalandcognitive tasks.For instance,whennormal subjectsprocessperceptuallyangry or fearful faces, a bilateral amygdalaresponse is observed, whereas labeling ofthe same expressions is associated with anattenuated amygdala response and an in-creased right prefrontal cortex response.5

Moreover, the patterns of activation anddeactivation of brain regions in response

to affective stimuli or in the course ofmildly anxiogenic tasks vary quantita-tively across subjects and can be pre-dicted in part by individual differences inproneness to experience negative emo-tionality and anxiety, and by some poly-morphic genes that influence behavior.

Several studies (see Tillfors et al1 for re-view) of perceptually induced anxiety con-cord broadly by showing altered patternsof cortical activation in patients with di-verse anxiety disorders compared withcontrol subjects. Abnormal responses havebeen reported in cortical areas involved inthe emotional evaluative (eg, visuospa-tial) processes, namely, the secondary vi-sual,1,6 parietal,1 and temporal,7 in addi-tion to the prefrontal and orbitofrontal,6

cortices.In social anxiety disorder, the investi-

gation of brain activity in response to af-fective stimuli of social relevance and insimulated anxiogenic social contexts ap-pears of special interest.

Author Affiliations:Department of Psychology,Vita-Salute San RaffaeleUniversity at the Department ofNeuropsychiatric Sciences,Istituto Scientifico San Raffaele,Milan, Italy (Drs Battaglia,Ogliari, and Maffei andMss Zanoni and Citterio);Department of Child Psychiatry,Istituto Scientifico EugenioMedea, Bosisio Parini, Italy(Drs Battaglia, Pozzoli, Giorda,and Marino); and Division ofMental Health and GeneticEpidemiology, NorwegianInstitutes of Public Health,Oslo, Norway (Dr Battaglia).

(REPRINTED) ARCH GEN PSYCHIATRY/ VOL 62, JAN 2005 WWW.ARCHGENPSYCHIATRY.COM85

©2005 American Medical Association. All rights reserved.

In adult patients with social phobia, compared withcontrol subjects during simulated public speaking, a de-creased cortical (secondary visual, parietal, retrosple-nial, temporal, and insular cortex) activity occurs typi-cally in association with increased amygdala activation(regional cerebral blood flow).1,6 Likewise, exaggeratedamygdala activation has been reported in controlled stud-ies of adults with social phobia exposed to angry or con-temptuous8 and neutral9 facial expressions, which are usu-ally categorized as mildly hostile or ambiguous.10

Thereisalsoinitial,butconsistent,evidenceofbiasedpro-cessinganddiscriminationof the facial expressionsofemo-tions inadultsandchildrenatheightenedriskforsocialpho-bia. A study of visual attention showed that adults withsocial phobia avoid salient facial features when they watchfacial expressions,11 whilechildrenwithhigh indexesof so-cial anxietyorshynesshavebiasedrecognitionofemotions,particularly the neutral and the angry expressions.12,13

The brain responses to basic elements of social com-munication in children at risk of developing social anxi-ety or avoidant personality disorder can thus help clarifythe developmental pathways to social phobia.

A temperamental disposition toward the avoidance ofnovel anduncertain situations togetherwitha setofbehav-iors that indicate shyness and discomfort in social interac-tionsarecomprehensivelynamedchildhoodshyness,orbe-havioral inhibition (BI).14 Children with high indexes ofshyness-BI are at a heightened risk of developing anxietydisorders, in particular social phobia,15 and subjects whofall within the BI–social phobia developmental continuumshowspecificpatternsofneurophysiologicresponsestopic-turesof facialexpressions.Adultswhohadbeencategorizedas behaviorally inhibited at the age of 2 years exhibited ahigher amygdala activation in response to unknown vs fa-miliar faces16 compared with adults who were uninhibitedas children, consistent with the notion that novel or am-biguous environmental stimuli of potential biological rel-evance activate the amygdala.17 Turning to genes that caninfluencetheneurobiologicalbasesoftheprocessingofemo-tions, 2 common alleles, the short (S) and the long (L), ina variable repeat sequence of the serotonin transporter (5-HTT)promoterpolymorphism(5-HTTLPR)onhumanchro-mosome17q11havebeendifferentlyassociatedwithgreateramygdala activity in response to angry or fearful faces18 inhealthy adults. The L and S alleles influence 5-HTT tran-scriptionactivity,19 withtheSalleleconveyingreducedtran-scription, lower transporter levels, and diminished seroto-nin uptake. The presence of 1 or 2 copies of the S variantinfluences predisposition to anxiety, avoidant behaviors,and interpersonal negative emotionality according to sev-eral,butnotall, geneticassociationstudiesofadults,20 whilea recent study of childhood shyness-BI21 found an associa-tion in the opposite direction (ie, with the LL genotype).

The cerebral visual event-related potentials (ERPs) arescalp potentials that occur within a few hundred milli-seconds after the presentation of a visual stimulus. TheERP waveforms contain components22-24 that span a con-tinuum between the earlier, “exogenous” potentials (re-flex responses controlled by the physical properties ofan external eliciting event) and the later, “endogenous”potentials (manifestations of information processing de-termined more by the nature of interaction between the

subject and the event). Meta-analytic evaluations of ERPstudies in twins show substantial heritability of some late(the P300 and N400 waveforms) ERP components.25

Childrenfromtheageof10yearsshowthesameERPlatecomponents26-28 found in adults when they look at picturesoffacialaffects.29Theseincludeacharacteristicnegativewave-formthatoccursatabout400milliseconds(N400)believedto reflect specific cognitive processing of human facial in-formation30,31 intheabsenceofverbal-semanticinformation.32

More broadly, the N400 has been claimed to reflect a tem-poral correlate of a corticoamygdala pathway of emotionalprocessing.33,34Inkeepingwithpositronemissiontomographicand functional magnetic resonance imaging studies thatshowedactivationofprimaryandsecondaryvisualareasandofthesomatosensorycortexduringtheactofprocessingfaceexpressions,3 ERP studies suggest that expression identifi-cation isassociatedwithacentroparietal effect, indexingat-tentionandmotivationprocessesofadaptivesignificance.29-31

The present study analyzes the N400 visual ERP re-sponses of children to different emotional facial expres-sions. We hypothesized that individual differences in shy-ness-BI and genetic variability at the 5-HTTLPR couldpredict a pattern of diminished cortical activation in re-sponse to hostile compared with neutral or prosocial ex-pressions, which could be put into a relationship with thefinding of biased recognition of expressions of emotionsin shy children,12,13 and with the patterns of central ner-vous system activation found by brain-imaging studies ofsocial phobia in tasks of perceptually induced anxiety.1

METHODS

SUBJECTS

One year before this study (time 0), an inception cohort of 149schoolchildren were characterized by their degree of shynessand ability to discriminate facial expressions of emotions.12 Re-sources were available to study a fraction of the original samplethrough interviews and electrophysiologic evaluation; 70 of the149 children were therefore randomly selected and invited toparticipate in the present study. After a complete descriptionof the study, 55 of the 70 invited children and their parentsagreed to participate, and parents signed a declaration of in-formed consent. The procedures of this study were acceptedby the ethical committee of the participating institutions.

Six of the 55 participating children were not included in thefinal sample for different reasons, such as minor physical ill-nesses on the day of the ERP recording, or unavailability of anadult person to accompany the child to the laboratory. This left49 white children of Italian ancestry with normal or corrected-to-normal visual acuity to take part in the study. Post hoc com-parisons showed no differences in demographic and psycho-metric characteristics between 49 participating vs 100nonparticipating children (Table 1). Although there were 49participants, results are shown for numbers of participants thatvary between 34 and 49 because of removal of subjects whohad artifacts like eyeblinks in some recordings, or who had anunsuccessful DNA collection (which occurred in 4 children).

At time 0, the children’s degree of shyness-BI was evaluatedby a questionnaire that was filled in by appropriately trained teach-ers, and by direct observation of the number of spontaneous com-ments made in the presence of an unfamiliar adult, based on pre-vious descriptions of children with BI.12,14 The questionnaireincluded a set of items seeking to identify temperamental dis-



position to BI and symptoms of possible social anxiety disorderproper, and included the Italian translations of the Stevenson-Hinde and Glover Shyness to the Unfamiliar,35 Cloninger andcoworkers’ Harm Avoidance Scale,36 and the Liebowitz SocialAnxiety Scale37 adapted for children. The teachers were askedto base their judgment on the instructions received from our groupat seminars and on direct observation. The scales had satisfac-tory indexes of internal consistency (Cronbach �, 0.89, 0.85, and0.88, respectively), moderate to high cross-correlations (HarmAvoidance Scale–Shyness to the Unfamiliar, 0.64; Harm Avoid-ance Scale–Liebowitz Social Anxiety Scale, 0.62; Shyness to theUnfamiliar–Liebowitz Social Anxiety Scale, 0.63; all P�.01), andnegative covariation with children’s number of spontaneous com-ments.12 Moreover, higher scores on the Liebowitz Social Anxi-ety Scale predicted a significantly higher number of misclassifi-cations of expressions.12 Therefore, we used part of these scales’items to build a more succinct index of BI-shyness.

BEHAVIORAL MEASURES

A principal component analysis of the items collected by ques-tionnaire at time 0 provided 3 factors with the following eigen-values (and percentages of explained variance): 11.99 (59.9%),2.1(9.8%),and1.2(6.1%).Amongthe itemsbelonging in the firstfactor, the following7hadmaximal factorial loading(range,0.80-0.89),maximal reciprocalcorrelations(r=0.58-0.86,allP�.001),and maximal internal consistency (Cronbach ��0.89): (1) thechild fears or avoids being at the center of attention; (2) the childfears or avoids raising his or her hand and answering questionsin frontof theclass; (3) thechild fearsoravoidsunfamiliarpeople;(4) the child speaks easily with unfamiliar adults (negative load-ing); (5) the child tends to avoid new visitors and strangers; (6)the child plays readily with new children (negative loading); and(7) when the child meets new children, it takes him or her a longtime to start talking. These items were selected to build a conciseindexofshyness-BI thatcouldbeassociatedwiththevariablesusedin the present study. The shyness-BI index ranged from 0 to 30,withameanvalueof10.86±6.6(minimum,0;maximum,26;skew-ness, 0.34 [SE, 0.33], and kurtosis, −0.76 [SE, 0.67]); it was nor-mallydistributed(Shapiro-WilkW=0.96,P=.10)withoutsignifi-cant sex differences (mean for 26 boys, 9.5 + 6.0, vs mean for 23girls, 12.8 + 5.9; t47=1.93, P=.06).

Children’s behavioral profiles were also assessed on the dayof the ERP experiment, based on 3 different measures. First, whileelectrodes were placed for the ERP recording (average time to

complete, 20 minutes; range, 17-22 minutes), the number of thespontaneous comments made by every participant was counted.Second, after the ERP recording, all mothers and children wereinterviewed individually by trained clinical psychologists withthe Italian version of the Schedule for Affective Disorders andSchizophrenia for School-age Children (K-SADS)38 interview tocollect the children’s lifetime DSM-IV symptoms of social pho-bia, simple phobia, depression, enuresis, generalized anxiety dis-order, separation anxiety disorder, panic disorder, attention-deficit/hyperactivity disorder, obsessive-compulsive disorder,conduct disorder, oppositional disorder, and tic disorder. Thepresence of symptoms of DSM-IV childhood disorders was es-tablished on the basis of blind review sessions chaired by a se-nior child psychiatrist (Dr Marino) of all the information ob-tained by both the mother and the child’s K-SADS interviewprotocols. All children were included in the study, regardless ofthe presence or absence of a full psychiatric diagnosis. Third,mothers filled in the Child Behavior Checklist 4-1839 and an-swered questions on parental occupations to calculate the so-cioeconomic status on the basis of the Hollingshead scale.40

DNA EXTRACTION AND GENOTYPING

Genomic DNA was extracted from mouthwash samples col-lected in 4% sucrose by means of a reagent for isolation of ge-nomic DNA (DNAzol Genomic DNA Isolation Reagent; Mo-lecular Research Center Inc, Cincinnati, Ohio).

Thepolymorphisminthetranscriptionalcontrolregionupstreamfrom the 5-HTT coding sequence (5-HTTLPR) was analyzed bypolymerase chain reaction according to the method reported byLeschetal19 andHeilsetal.41 Twofragmentsweregenerated:ashortvariant(S)of484basepairsandalongvariant(L)of528basepairs.

All amplification reactions were performed on a thermocy-cler (Mastercycler; Eppendorf, Milan, Italy). The amplified prod-ucts were analyzed on 2% agarose gels.

For 4 of the 49 children, the DNA extraction was unsuccess-ful, and therefore genetic data were available for only 45 subjects.

PROTOCOL

Rationale for Stimulus Selection

We chose to use standardized faces of other children of similarage (models aged 8-9 years), instead of adults, for 2 reasons: first,schoolchildren spend most of their time among other children,

Table 1. Demographic and Psychometric Features of Participants vs Nonparticipants in the ERP Experiment,Based on Evaluations Made 1 Year Before the ERP Recordings (Time 0)

Participants(n = 49)

Nonparticipants(n = 100) t Value/�2 df P Value

Age, mean ± SD, y 8.84 ± 0.72 8.67 ± 0.60 1.52 147 .13*LSAS score, mean ± SD 7.96 ± 5.23 7.71 ± 5.07 0.28 147 .78*SU score, mean ± SD 11.33 ± 5.39 11.77 ± 5.27 –0.48 147 .63*HA score, mean ± SD 3.67 ± 4.01 4.35 ± 3.97 –0.97 147 .33*No. of spontaneous comments at time 0, mean ± SD 7.42 ± 1.76 7.51 ± 1.75 –0.29 147 .77*Sex, No. (%) male 26 (53) 53 (53) 0.02 1 .86†SES, No.

Lower 6 14 0.15 2 .93†Middle 24 46Upper 19 40

Abbreviations: ERP, event-related potential; HA, Harm Avoidance Scale of Cloninger Junior Temperament and Character Inventory, Parent Version; LSAS,Liebowitz Social Anxiety Scale adapted for children; SES, socioeconomic status calculated on the basis of the Hollingshead scale (lower range, score of 1-3;middle, score of 4-6; upper, score of 7-9); SU, Stevenson-Hinde and Glover Shyness to the Unfamiliar scale.

*By t test.†By �2 test.



not adults; second, socially anxious children rate rejection andteasing from peers among the most feared situations.42 Stimuliconsisted of 6 black-and-white pictures of a boy and a girl stan-dardized for size, contrast, and luminosity, displaying 3 emo-tions: joy, anger, and a neutral expression. The boy pictures weredrawn from a collection of Linda Camras, PhD.43 The girl pic-tures were selected under the direction of Camras from a pool ofpictures that had been developed by showing the model Ekmanand Friesen’s44 prototypical pictures as a reference. In accor-dance with Camras’ standard-validation procedure, the girl pic-tures were correctly classified in more than 80% of evaluationsby 20 undergraduates of both sexes.12 Stimuli appeared on a 38-cmmonitor placed at a distance of 41 cm and were presented in anoval aperture that occluded sex-specific features. Children weretold that this was a video game in which they would get a gift ifthey carefully followed the instructions and performed well.

Joy was included as a prototype of a prosocialization ex-pression that is almost never misidentified by children12 or neu-ral networks that replicate human emotional-cognitive tasks.45

The joy expression is processed by brain networks at least par-tially independent from those involved in processing hostileexpressions3 and evokes neurofunctional responses that can bein part predicted by the extroverted, but not by the intro-verted, temperamental dimensions.46

The neutral expression was included because it is usuallycategorized as mildly hostile or ambiguous10 and was signifi-cantly more often misclassified by children with higher in-dexes of social anxiety12 at time 0.

The angry expression was included because it is signifi-cantly more often misclassified by children with higher in-dexes of social anxiety,12,13 and it has been found to elicit height-ened subcortical amygdala activity in adults with social phobia8

and in adults with 1 or 2 copies of the 5-HTT S allele.18

Trials

On each trial the children were first presented with a child’sface (total time on screen, 1300 milliseconds), which they wereinstructed to watch carefully until a blue circle appeared su-perimposed around the center of the picture. As soon as a bluecircle appeared (700 milliseconds after the appearance of thestimulus), they had to click a mouse. Thus, the ERPs relevantto this study were all generated before the motor task, whichwas merely set up to stimulate children’s participation and at-tention. The monitor screen remained dark between trials forperiods that varied randomly from 1200 to 1600 milliseconds.The stimuli were presented to all the children in a sequencethat alternated male and female pictures and that avoided closerepetition of the same expression. To simulate a real video gameand reinforce participation, a screen picture with increasing scor-

ing values appeared about every 6 pictures. Each stimulus waspresented 20 times to ensure sufficient ERP acquisition (total,120 presentations in a complete session).

Every child was exposed to a preexperiment trial of 6 pic-tures not belonging to the same set used for the experiment tomake sure she or he understood the procedure well.

Each child received a gift of a value equivalent to €30.

ERP ACQUISITION AND ANALYSIS

As in other studies of ERP and facial expressions of childrenand adults,26,27,29,47 electroencephalographic activity was re-corded at sites Fz, C3, Cz, C4, and Pz of the 10-20 system withthe use of silver–silver chloride electrodes referred to linkedmastoids with an amplifier (Neuroscan SynAmp; NeuroscanLabs, Sterling, Va), with head preamplified gain 150 and ac-quisition software (SCAN, version 4.2; Neuroscan Labs). Theground electrode was attached to the forehead. Electro-oculographic activity was recorded from electrodes placed aboveand below the right eye.

Electrode impedance was maintained below 5 k�. The elec-troencephalogram and electro-oculogram were amplified (gain500), analogically bandpass filtered (1-30 Hz), digitized, and ac-quiredata1000-Hzsamplingrate.Electroencephalogramandelec-tro-oculogram epochs between –50 and 1300 milliseconds fromthe stimulus onset were obtained by means of different triggercodes for each image, allowing for later off-line artifact rejection,sorting, and digital averaging with the Neuroscan EDIT software(Neuroscan Labs). All epochs from all electrodes were rejected ifaffected by artifacts (greater than +65 µV or less than –65 µV be-tween –50 and 700 milliseconds). The ERP averages were con-structed from artifact-free epochs for each trigger code and foreach electrode. Amplitudes were measured according to the dis-tance between peaks and troughs for each identified waveform.

RESULTS

RELATIONSHIP OF THE SHYNESS-BI INDEXTO OTHER BEHAVIORAL MEASURES

Table 2 summarizes the cross-correlations of the be-havioral indexes used in the study.

The shyness-BI index significantly predicted the num-ber of spontaneous comments made by children (mean ±SD,3.39±4.87; range,0-17)while theelectrodeswerebeingplaced on their scalps, and the number of lifetime symp-toms of social phobia (mean ± SD, 2.26±2.72; range, 0-8)collectedby theK-SADS interview,butnoother symptoms

Table 2. Cross-correlations of the Behavioral Measures Used in the ERP Experiment*

Shyness-BIIndex

No. of SpontaneousComments on ERP

Experiment Day

No. of Symptoms ofSocial Phobia by K-SADS

Direct InterviewCBCL

Withdrawn Scale

Shyness-BI index 1No. of spontaneous comments on ERP experiment day –0.52† 1No. of symptoms of social phobia by K-SADS direct interview 0.41† –0.27 1CBCL withdrawn scale 0.32‡ –0.11 0.35‡ 1

Abbreviations: BI, behavioral inhibition; CBCL, Child Behavior Checklist; ERP, event-related potential; K-SADS, Schedule for Affective Disorders andSchizophrenia for School-age Children.

*Prediction of number of spontaneous comments on the ERP experiment day by the shyness-BI index: linear regression, F1,47=14.30, P=�.001, �=–0.52,adjusted R 2=0.25. Prediction of number of K-SADS lifetime symptoms of social phobia by the shyness-BI index: linear regression, F1,47=9.63, P=.003, �=0.41,adjusted R 2=0.15.

†P�.001.‡P�.03.



of mental disorders assessed with the K-SADS (the predic-tion closest to significance pertained to separation anxiety,with P=.18). Moreover, among the 9 narrow-band dimen-sions of problem behaviors measured by the Child Behav-ior Checklist, the Withdrawn scale (possible scores, 0-16;mean±SD,3.02±2.81; range,0-11)wastheonlyonetocor-relate significantly with the shyness-BI index.

ERP ANALYSES

Analyses of General Effects

Analyses of the waveforms generated by the facial expres-sions (Figure 1 and Table 3) showed an enhanced latenegativity occurring after stimulus presentation at a meanof 256 milliseconds for the Fz, 368 milliseconds for the

–10

–5

0

5

10

0 100 400200 500300 600 700Latency, ms

Ampl

itude

, µV

JoyNeutralAnger

Pz

–10

–5

0

5

10

0 400200 600

Latency, ms

Ampl

itude

, µV

Ampl

itude

, µV

EOG

–10

–5

0

5

10

0 400200 600

C3 –10

–5

0

5

10

0 400200 600

Cz –10

–5

0

5

10

0 400200 600

Pz

–10

–5

0

5

10

0 400200 600

Latency, ms Latency, ms

Fz –10

–5

0

5

10

0 400200 600

C4

Figure 1. Grand averages of waveforms generated by facial expressions at the reference electrodes, with an enlargement of the waveforms at the Pz electrode.EOG indicates electro-oculogram.



Cz, and 394 milliseconds for the Pz electrode: these 2 lat-ter waveforms were identified as N400 with an ampli-tude that followed an anteroposterior gradient midline. Ac-cording to analysis of variance (ANOVA) on 5 electrodes(Table 3), the mean N400 amplitudes at Pz and Cz werenot significantly different from each other, and the meanN400 amplitudes at C3 and C4 were not significantly dif-ferent from those obtained from the Cz and Pz elec-trodes. Such centroparietal maximum amplitudes for a face-evoked ERP N400 wave can be found in other studiesfollowing similar methods.29,48 Although the emotional ex-pressions of anger and joy tended to elicit slightly largerN400 amplitudes than the neutral expression across thedifferent electrodes, no significant difference for N400 am-plitude or latency for type of expression was disclosed byrepeated-measures ANOVA with the type of expressionas factor, carried out on the 49 children participating inthe study at the reference electrodes (Table 3). However,waveforms at Pz were characterized by an enhanced latenegativity occurring exactly at the 400-millisecond ex-pected time, with separation of 3 waveforms (correspond-ing to the 3 expressions) that was visually apparent fromthe very beginning of the ERP waves (Figure 1; see sec-tion enlargement of the Pz electrode).

To ascertain the degree of variability of measures acrosstrials, we performed a repeated-measures ANOVA with 2factors: (1) typeofexpression(on3 levels)and(2) time(on2 levels: first block, encompassing the first 50% of repeti-tionsofeachexpression;andsecondblock,withtheremain-ingrepetitions) forbothamplitudesandlatenciesatall elec-trodes.Theresults showedtheabsenceof significanteffectsof timeor time�expressionfactors for theN400waveforms(Table3:valuesandstatisticsaregivenforPz; forotherelec-trodes, data are available from authors on request).

We also found no effect of sex or sex�expression onthe N400 characteristics (Table 3), and therefore subjectsof both sexes were pooled in all subsequent analyses.

Effects of Shyness-BI on ERP Waveforms

To test the hypothesis that shyness-BI predicts N400 char-acteristicswhenchildrenarepresentedwithotherchildren’sexpressions of emotions, 2 multiple linear regression pro-cedureswereperformedseparatelyforN400amplitudesandlatencies, with the N400 amplitudes (or latencies) for eachexpression (joy, neutral, anger) at the different electrodes(Fz, Cz, Pz, C3, C4) as dependent variables, and the degreeof shyness-BI as the independent variable.

Table 3. N400 Amplitudes and Latencies Evoked by Facial Expressions at the Reference Electrodes*

Electrode

Joy Neutral Anger Mean of 3 Expressions

Amplitude, µV(r With Shyness-BI) Latency, ms

Amplitude, µV(r With Shyness-BI) Latency, ms

Amplitude, µV(r With Shyness-BI) Latency, ms Amplitude, µV Latency, ms

Fz –15.05 ± 4.50 (0.14) 254.61 ± 22.47 –14.57 ± 3.97 (0.14) 260.09 ± 25.58 –14.67 ± 5.051 (0.19) 257.33 ± 31.68 –14.83 ± 4.14 256.36 ± 22.80n 37 40 38 40 37 40 34 38

Cz –20.09 ± 6.71 (0.26) 370.00 ± 34.76 –18.99 ± 6.32a (0.43†) 364.63 ± 33.91 –19.87 ± 6.19 (0.20) 364.26 ± 39.99 –20.09 ± 5.48 367.68 ± 31.96n 47 45 46 44 46 45 44 44

Pz –18.76 ± 6.61 (0.22) 392.01 ± 28.45 –17.66 ± 6.82b (0.31†) 395.01 ± 26.83 –17.49 ± 4.94c (0.32†) 391.47 ± 28.87 –18.06 ± 5.29 393.68 ± 23.75n 43 43 43 43 42 42 42 42

C3 –19.58 ± 6.60 (0.13) 379.85 ± 27.25 –18.49 ± 5.93 (0.26) 372.81 ± 33.71 –19.43 ± 6.06 (0.17) 367.11 ± 31.17 –19.16 ± 5.54 373.21 + 25.87n 41 39 40 38 40 38 40 38

C4 –18.38 ± 6.94 (0.19) 369.90 ± 35.90 –17.80 ± 6.54d (0.31†) 368.52 ± 35.07 –19.09 ± 5.76e (0.30†) 373.17 ± 30.56 –19.08 ± 5.33 371.77 ± 29.56n 46 44 44 43 44 43 41 42

Abbreviations: BI, behavioral inhibition.*Amplitudes and latencies are mean±SD.

Differences of mean values of 3 expressions at the 5 reference electrodes:Repeated-measures general 1-way analysis of variance (ANOVA) on N400 amplitudes: F4,96=9.17, P�.001.

Post hoc Tukey honestly significant difference (HSD): Cz vs Fz, P�.001; Pz vs Fz, P�.03; Pz vs Cz, P�.07; C3 vs Fz, P�.001; C4 vs Fz, P�.003.All other post hoc comparisons: not significant (NS).

Repeated-measures general 1-way ANOVA on N400 latencies: F4,100=247.9, P�.001.Post hoc Tukey HSD: Cz vs Fz, P�.001; Pz vs Fz, P�.001; C3 vs Fz, P�.001; C4 vs Fz, P�.001; Pz vs Cz, P�.005.All other post hoc comparisons: NS.

Effect of type of expression on N400 characteristics at each electrode:Repeated-measures ANOVA on N400 amplitudes: Fz: F2,66=0.01, P=.98; Cz: F2,86=1.61, P=.21; Pz: F2,82=1.22, P=.30; C3: F2,78=1.11, P=.34; C4: F2,80=0.98, P=.38.ANOVA on N400 latencies: all results NS.

Variability of measures across trials (habituation effects):Repeated-measures ANOVA:

N400 at Pz electrode: amplitude (time): F1,43=1.01, P=.32; amplitude (time�expression): F2,86=2.46, P=.09; latency (time): F1,43=3.78, P=.06; latency(time�expression): F2,86=0.64, P=.52.

Other electrodes: all NS for both latency and amplitude; available from the authors on request.Effect of sex and sex�expression interaction on N400 characteristics:

ANOVA on N400 amplitudes at Pz electrode: F1,40=0.27, P=.60 (sex); F2,80=0.11, P=.89 (sex�expressions).ANOVA on N400 latencies at Pz electrode: F1,40=0.24, P=.63 (sex); F2,80=0.08, P=.92 (sex�expressions).Repeated-measures ANOVA on other electrodes: all NS for both latency and amplitude; available from authors on request.

Influence of shyness-BI index on N400 by type of expression:Multiple linear regression for N400 amplitudes:

Anger expression: Pz: F1,40=4.48, superscript c indicates P�.04, �=0.32, adjusted R 2=0.07; C4: F1,39=3.87, superscript e indicates P�.056, �=0.30, adjustedR 2=0.067.

Neutral expression: Pz: F1,40=4.19, superscript b indicates P�.047, �=0.31, adjusted R 2=0.07; Cz: F1,42=9.66, superscript a indicates P�.003, �=0.43, adjustedR 2=0.17; C4: F1,39=4.15, superscript d indicates P�.048, �=0.31, adjusted R 2=0.07.

Joy expression: NS for all electrodes.Multiple linear regression for N400 latencies: NS for all expressions at all electrodes; available from authors on request.†P�.05 between the shyness-BI index and amplitudes at reference electrodes; correlations between the shyness-BI index and latencies were all NS.



Children’s shyness-BI predicted the N400 amplitudefor anger and the neutral expression (Table 3) at Pz, Cz,and C4, always in the direction of a reduction of ampli-tude, but not the amplitude for the joy expression on anyof the electrodes. Similarly, no significant prediction forany of the expressions was provided by regression ap-plied to the N400 latencies for any of the electrodes.

Genotypes and Allelic Frequencies

The allelic frequencies were 49 (54%) for the L allele and41 (46%) for the S allele (Table 4); both were in Hardy-Weinberg equilibrium, without sex-related differences indistribution (respectively, � 2

2=1.09, P=.57; and � 21=0.43,

P=.51). All ANOVAs dealing with the 5-HTTLPR geno-type were conducted with the 3 genotypes separately and

with the classification19 that combines the LS and SS geno-types into “S carriers.”

While socioeconomic status, grade, and sex wereequally distributed across genotypes and in LL vs S car-rier subjects, children with the SS genotype and S carri-ers were slightly, but significantly, younger (Table 4).

Genetic Analyses

A polynomial univariate ANOVA, where the factor was thegenotype at the 5-HTTLPR gene and the dependent vari-able was the shyness-BI index, showed that shyness wassignificantlydifferent across thegenotypes,with the SS geno-type being associated with higher shyness-BI index(Table 4); similarly, S carriers vs LL subjects showed a trendtoward significantly higher shyness-BI index (Table 4).

Table 4. Influence of 5-HTTLPR on Shyness-BI Index, N400 Waveforms, and Misclassifications of Expressions at Time 0*

5-HTTLPR Genotype

SS (n = 9) LS (n = 23) LL (n = 13)S Carriers(n = 32)

Age, mean ± SD, y† 8.44 ± 0.52 8.65 ± 0.64 9.2 ± 0.59 8.54 ± 0.56‡Sex, No. (%) male† 6 (67) 11 (48) 6 (46) 17 (55)Education, No.†

Grade 3 5 15 5 20Grade 4 4 8 8 12

SES, No. (%)†Lower 2 (22) 2 (9) 1 (8) 4 (13%)Middle 3 (33) 12 (52) 7 (54) 15 (47)Upper 4 (44) 9 (39) 5 (38) 13 (41)

Shyness-BI index, mean ± SD 15.44 ± 8.72§ 10.04 ± 5.10 7.69 ± 5.19 11.74 + 6.68N400 amplitude, µV �

JoyMean ± SD –19.76 ± 9.95 –19.83 ± 6.07 –17.5 ± 4.35 –19.81 ± 7.18n 8 20 11 28

NeutralMean ± SD –17.72 ± 7.81 –19.13 ± 6.98 –17.69 ± 5.24 –18.73 ± 7.10n 8 20 11 28

AngerMean ± SD –14.12 ± 4.43¶ –17.89 ± 3.92 –20.68 ± 4.59 –16.23 ± 4.35#n 8 20 11 28

No. of misclassified expressionsat time 0, mean ± SD

3.00 ± 1.22** 2.91 ± 1.47 1.92 ± 1.12 2.9 ± 1.43††

Abbreviations: 5-HTTLPR, serotonin transporter promoter polymorphism; SES, socioeconomic status calculated on the basis of the Hollingshead scale (lowerrange, score of 1-3; middle, score of 4-6; upper, score of 7-9).

*Time 0 refers to a trial of facial expression discrimination task taken by children 1 year before the event-related potential recordings. DNA analysis wassuccessfully completed in 45 children. The “S carriers” column represents the SS and LS genotypes combined.

†For age, by analysis of variance (ANOVA), F1,43=10.1, P=.003. Sex was equally distributed across genotypes (�22=1.11, P=.51) and in LL vs S carrier subjects

(�21=0.55, P=.46). Education by grade was equally distributed across genotypes (�2

2=2.41, P=.29) and in LL vs S carrier subjects (�21=3.24, P=.07). The SES

was equally distributed across genotypes (�22=1.71, P=.79) and in LL vs S carrier subjects (�2

1=0.68, P=.71).‡For LL vs S carrier subjects, t43=3.17, P=.02.§Association of 5-HTTLPR to shyness-BI index: ANOVA, F2,42=4.47, P�.02; 2=0.18, observed power=0.75; Tukey honestly significant difference (HSD) 95%

confidence intervals: LS vs LL, P=.5; SS vs LL, P=.01; SS vs LS, P=.07; independent samples, 2-tailed t test of LL vs S carrier subjects, t43=–1.95, P=.057.�Variability of N400 amplitude measures across trials (habituation effects) by 5-HTTLPR: repeated-measures analysis of covariance (ANCOVA) (3 5-HTTLPR

genotypes): time: F1,38=1.52, P=.20; time�expression, F2,76=1.38, P=.26; time�expression�genotype: F4,76=0.79, P=.53; repeated-measures ANCOVA (LLsubjects and S carriers): time�expression�genotype: F2,76=0.53, P=.59.

¶ANCOVA on N400 amplitudes at Pz by expression (3 levels), with genotype as independent variable and shyness-BI index and age as covariates: type ofexpression: F2,72=1.25, P�.29, mean square (MS) effect=18.3, MS error=14.59; genotype: F2,35=0.11, P�.89, MS effect=8.49, MS error=76.64;expression�genotype interaction: F4,72=3.57, P�.01, MS effect=52.0, MS error=14.58; 2=0.18, observed power=0.88. Post hoc comparison: SS vs LL subjectson anger, Tukey HSD, P�.02.

#ANCOVA on N400 amplitudes at Pz by expression (3 levels), with LL or S carrier as independent variable and shyness-BI index and age as covariates: type ofexpression: F2,74=0.17, P=.84; LL or S carrier: F1,35=0.11, P=.74. Expression by LL or S carrier interaction: F2,76=5.70, P�.005. Post hoc comparison: S carrier vsLL subjects on anger, Tukey HSD, P�.05.

**Linear regression on number of misclassified expressions at time 0 by 5-HTTLPR genotype (0, 1, or 2 S alleles): F1,44=4.08, P�.05, �=0.29, adjustedR 2=0.065.

†† t Test for number of misclassifications of LL vs S carrier subjects: t43=−2.20, P=.03.



In light of the foregoing relationships, and after check-ing for the homogeneity of covariance matrices (Bartlett �2

not significant on all dependent variables and the covari-ate; box M=19.98, � 2

20=16.15, P=.70), we performed analy-sis of covariance where the independent variables were thegenotypes at the 5-HTTLPR, and the dependent variableswere the N400 amplitudes evoked by facial expressions atthe Pz electrode, the repeated-measures factor was the ex-pression (joy, neutral, anger), and the covariates were theshyness-BI index and age. Neither the type of expressionnor genotype per se influenced the N400 amplitudes, buta significant expression�genotype interaction was found(Table 4). The post hoc comparison showed that the sig-nificance was sustained by smaller amplitude elicited byanger in the SS subjects compared with the LL subjects(Figure 2, Figure 3, and Table 4). The analysis of co-variance, contrasting LL subjects with S carriers, providedvery similar results, ie, the absence of genotype or expres-sion effects, but a significant expression�genotype inter-action effect, sustained by smaller N400 elicited by angerin S carriers (Table 4).

These results for the anger expression could not beexplained by habituation effects on the different expres-sions that could have occurred between the first and thesecond block of repetitions, as shown by repeated-measures analysis of covariance on amplitudes with 3 fac-tors: (1) time (on 2 levels: first and second blocks of rep-etitions), (2) expression, and (3) genotype, withshyness-BI index and age as covariates (Table 4).

Because the results of the ERP analyses suggested a rolefor both shyness-BI index and the 5-HTTLPR genotype indetermining different patterns of information processingof hostile (angry) and/or neutral facial expressions, wetested post hoc whether the 5-HTTLPR genotype could pre-dict a biased discrimination of expressions in the trial com-pleted 1 year before the ERP recordings.12 A linear regres-sion was then performed on the total number ofmisclassifications made at time 0 by the 45 children whosegenotyping was available, with the 5-HTTLPR genotype(0, 1, or 2 S alleles) as the independent variable. The num-ber of misclassifications was predicted significantly, in keep-ing with the previous findings. Similarly, an independentsamples, 2-tailed t test of the number of misclassifica-tions between S carriers and LL subjects showed a trendtoward significantly worse performance for the former(Table 4).

To evaluate whether the 5-HTTLPR genotype hadgreater power to predict the phenotype of shyness-BI in-dex, or the endophenotype of N400 amplitude at Pz inresponse to anger, we specified a linear trend, with weightsof –1, 0, and 1 for the 0, 1, and 2 S alleles, and a qua-dratic trend, with weights of –1, 2, and −1 for the 0, 1,and 2 S alleles, respectively, and then entered these codedvectors simultaneously in the regression equations49,50 topredict shyness-BI index and the N400 amplitude. Re-sults showed that the linear solution was the only oneretained by the analyses, with a 25% greater R2 for theN400 amplitude (Table 5).

–10

–12

–14

–16

–18

–20

–22

–24

LS

SS

5-HTTLPR

N400

Am

plitu

de, µ

V

LLAnger

Neutral

Joy

Figure 2. Amplitudes of N400 generated at the Pz electrode by serotonintransporter promoter polymorphism (5-HTTLPR) genotype and type of facialexpression.

28

22

16

10

4

L/S

SS

5-HTTLPR

Shyn

ess-

BI

LL

–2

–6

–10

–14

–18

–22

–26

–30

N400

Am

plitu

de, µ

V

Figure 3. Scatterplot of subjects in the experiment ordered by serotonintransporter promoter polymorphism (5-HTTLPR) genotype. The graphshows the degree of shyness or behavioral inhibition (BI) and the N400amplitude evoked by anger at the Pz electrode for each subject.

Table 5. Regressions of 5-HTTLPR Genotype on N400Amplitude Elicited by the Anger Expression at Pz andon the Shyness-BI Index

Anger N400 Amplitudeat Pz Model:

F2,36 = 5.55, P = .008

Shyness-BIIndex Model:

F2,42 = 4.57, P = .02

� t P Value � t P Value

Linear trend 3.30 3.33 .002* 3.88 2.99 .005†Quadratic trend –0.16 –0.34 .73 –0.51 –0.85 .40

Abbreviations: BI, behavioral inhibition; 5-HTTLPR, serotonin transporterpromoter polymorphism.

*R 2=0.24; adjusted R 2=0.19.†R 2=0.18; adjusted R 2=0.14.



COMMENT

Shyness-BI index and the presence of 1 or 2 copies of the5-HTTLPRSallelepredictedsmallerERPN400amplitudesin response toovertlyhostile andneutral facial expressionsatcentroparietalregions,whichareinterconnectedwithemo-tion processing networks encompassing the amygdala andthe prefrontal cortex.3,51-53 Inasmuch as the N400 reflectsthe neural processes involved in recognition of facialexpressions,29-31 asmalleramplitudeincarriersof1or2cop-ies of the 5-HTTLPR S allele suggests diminished corticalinvolvement and predicts partially impaired reading29,54 inresponse to some facial affects, consistentwith the findingsof biased discrimination of the hostile and ambiguous ex-pressionsattime0bythesamesubjects.12 SmallerERPwave-forms in response to angry faces have been reported in acontrolled study of posttraumatic stress disorder and havebeen interpreted as a relative reduction of cortical activity,co-occurring with heightened subcortical activity,54 in re-sponsetohostilesocialstimuli. Indeed,arelativelydecreasedactivity in thecortices that are involved inemotional evalu-ativeprocesses(includingthesecondaryvisualandparieto-occipital areas)hasbeendemonstrated toco-occurwithex-aggerated activation of the amygdala in adults with socialphobia during simulated social interaction investigated byfunctional brain imaging techniques.1,6,7 A decreasedcortical—asopposed tosubcortical—perfusion is contraryto the pattern of activation in normal controls1 in condi-tions of simulated social interaction. Likewise, adults withsocialphobia showheightenedsubcortical, andspecificallyamygdala, activation in response to neutral9 and angry8 fa-cial expressions, and carriers of 1 or 2 copies of the S allelehave an increased amygdala response18 to anger. The in-creasedsubcortical, anddecreasedcortical, activation inre-sponse toemotional stimuliof social relevancehasbeen in-terpretedas theexcessive involvementofaphylogenetically“older” system1 of recognition of dangerous, novel, or am-biguous stimuli.17 Heightened amygdala activation in re-sponse to novel faces in adults who had been categorizedasshy inchildhood16 maybeseenasbroadlyconsistentwiththese data.

Consistent with the finding that the neurofunctionalresponses to happiness expressions can be partly pre-dicted by the extraverted, but not by the introverted, tem-peramental dimensions,46 childhood shyness-BI and/or the5-HTTLPR genotype did not predict the N400 potentialsevoked in response to a prosocial (joy) expression.

The greater R2 found by regressing the 5-HTTLPR geno-type on the endophenotype of N400 than on the shy-ness-BI phenotype is consistent with the suggestions55 thatthe effect of functional genetic polymorphisms is morerobustly assessed by assays of brain physiology than withbehavioral phenotypes. However, while the associationbetween the SS 5-HTTLPR genotype or S carrier statusand shyness-BI is consistent with most studies of adulthuman emotionality,20 a study found the LL genotype as-sociated with childhood shyness-BI.21

Although these results need replication, they show thatgenetic variation at 5-HTTLPR contributes to shape theN400 waveform, a possible index of complex neuronalactivity that occurs when expressions of emotions are

being processed.29-32,56 A pattern of decreased cortical ac-tivation in response to some specific social cues (includ-ing peers’ nonverbal signs of refusal or neutrality) maythen constitute a heritable basis for biased discrimina-tion of some forms of socially relevant information, whichin turn may hamper social interactions and ultimately re-inforce a child’s disposition to shyness-BI.

Shy children have been shown to provide relativelydistinct physiologic responses in a variety of con-texts.57,58 These data suggest that a biased pattern of pro-cessing emotional information of social relevance can berecognized and characterized psychophysiologically earlyin life in children with heightened indexes of shy-ness-BI or 1 or 2 copies of the 5-HTTLPR S allele.

Submitted for Publication: February 24, 2003; final re-vision received May 27, 2004; accepted June 9, 2004.Correspondence: Marco Battaglia, MD, Department ofNeuropsychiatric Sciences, Istituto Scientifico San Raf-faele, 20 via Stamira d’Ancona, 20127 Milan, Italy ([email protected]).Funding/Support: This study was supported by the Ital-ian Ministry of University and Research, Rome (Co-Fingrant 11/2001-113555_004 awarded to Dr Battaglia), andthe National Alliance for Research in Schizophrenia andDepression Independent Investigator Award, Great Neck,NY (Dr Battaglia).Acknowledgment: We thank Massimo Molteni, MD, FloraBinaghi, MSc, Luca Bigotta, PhD, Linda Camras, PhD, An-drea Fossati, MD, Frederica Villa, MSc, Livia Brenna, NinoStillitano, Silvana Villa, MSc, Raffaella Braga, MSc, KimSommerschield, PhD, and all of the children, parents, andteachers who took part in this study.

REFERENCES

1. Tillfors M, Furmark T, Marteinsdottir I, Fischer H, Pissiota A, Langstrom B, Fred-rikson M. Cerebral blood flow in subjects with social phobia during stressful speak-ing tasks: a PET study. Am J Psychiatry. 2001;158:1220-1226.

2. Hariri AR, Mattay VS, Tessitore A, Fera F, Weinberger DR. Neocortical modulationof the amygdala response to fearful stimuli. Biol Psychiatry. 2003;53:494-501.

3. Adolphs R. Neural systems for recognizing emotion. Curr Opin Neurobiol. 2002;12:169-177.

4. Nomura M, Ohira H, Haneda K, Iidaka T, Sadato N, Okada T, Yonekura Y. Func-tional association of the amygdala and ventral prefrontal cortex during cognitiveevaluation of facial expressions primed by masked angry faces: an event-relatedfMRI study. Neuroimage. 2004;21:352-363.

5. Hariri AR, Bookheimer SY, Mazziotta JC. Modulating emotional responses: ef-fects of a neocortical network on the limbic system. Neuroreport. 2000;11:43-48.

6. Van Ameringen M, Mancini C, Farvolden P, Oakman JM. The neurobiology ofsocial phobia: from pharmacotherapy to brain imaging. Curr Psychiatry Rep. 2000;2:358-366.

7. Reiman EM. The application of positron emission tomography to the study ofnormal and pathologic emotions. J Clin Psychiatry. 1997;58(suppl 16):4-12.

8. Stein MB, Goldin PR, Sareen J, Zorrilla LT, Brown GG. Increased amygdala ac-tivation to angry and contemptuous faces in generalized social phobia. Arch GenPsychiatry. 2002;59:1027-1034.

9. Birbaumer N, Grodd W, Diedrich O, Klose U, Erb M, Lotze M, Schneider F, WeissU, Flor H. fMRI reveals amygdala activation to human faces in social phobics.Neuroreport. 1998;9:1223-1226.

10. Morris JS, Frith CD, Perrett DI, Rowland D, Young AW, Calder AJ, Dolan RJ.A differential neural response in the human amygdala to fearful and happy facialexpressions. Nature. 1996;383:812-815.

11. Horley K, Williams LM, Gonsalvez C, Gordon E. Social phobics do not see eye toeye: a visual scanpath study of emotional expression processing. J Anxiety Disord.2003;17:33-44.



12. Battaglia M, Ogliari A, Zanoni A, Villa F, Citterio A, Binaghi F, Fossati A, Maffei C.Children’s discrimination of expressions of emotions: relationship with indicesof social anxiety and shyness. J Am Acad Child Adolesc Psychiatry. 2004;43:358-365.

13. Simonian SJ, Beidel DC, Turner SM, Berkes JL, Long JH. Recognition of facialaffect by children and adolescents diagnosed with social phobia. Child Psychia-try Hum Dev. 2001;32:137-145.

14. Kagan J, Reznick JS, Snidman N. Biological bases of childhood shyness. Science.1988;240:167-171.

15. Biederman J, Hirshfeld-Becker DR, Rosenbaum JF, Herot C, Friedman D, Snid-man N, Kagan J, Faraone SV. Further evidence of association between behav-ioral inhibition and social anxiety in children. Am J Psychiatry. 2001;158:1673-1679.

16. Schwartz CE, Wright CI, Shin LM, Kagan J, Rauch SL. Inhibited and uninhibitedinfants “grown up”: adult amygdalar response to novelty. Science. 2003;300:1952-1953.

17. Whalen PJ. Fear, vigilance, and ambiguity; initial neuroimaging studies of thehuman amygdala. Curr Dir Psychol Sci. 1998;7:177-187.

18. Hariri AR, Mattay VS, Tessitore A, Kolachana B, Fera F, Goldman D, Egan MF,Weinberger DR. Serotonin transporter genetic variation and the response of thehuman amygdala. Science. 2002;297:400-403.

19. Lesch KP, Bengel D, Heils A, Sabol SZ, Greenberg BD, Petri S, Benjamin J, MullerCR, Hamer DH, Murphy DL. Association of anxiety-related traits with a polymor-phism in the serotonin transporter gene regulatory region. Science. 1996;274:1527-1531.

20. Munafo MR, Clark TG, Moore LR, Payne E, Walton R, Flint J. Genetic polymor-phisms and personality in healthy adults: a systematic review and meta-analysis.Mol Psychiatry. 2003;8:471-484.

21. Arbelle S, Benjamin J, Golin M, Kremer I, Belmaker RH, Ebstein RP. Relation ofshyness in grade school children to the genotype for the long form of the sero-tonin transporter promoter region polymorphism. Am J Psychiatry. 2003;160:671-676.

22. Picton TW, Bentin S, Berg P, Donchin E, Hillyard SA, Johnson R Jr, Miller GA,Ritter W, Ruchkin DS, Rugg MD, Taylor MJ. Guidelines for using human event-related potentials to study cognition: recording standards and publication criteria.Psychophysiology. 2000;37:127-152.

23. Pizzagalli D, Lehmann D, Koenig T, Regard M, Pascual-Marqui RD. Face-elicitedERPs and affective attitude: brain electric microstate and tomography analyses.Clin Neurophysiol. 2000;111:521-531.

24. Munte TF, Brack M, Grootheer O, Wieringa BM, Matzke M, Johannes S. Brainpotentials reveal the timing of face identity and expression judgments. NeurosciRes. 1998;30:25-34.

25. van Beijsterveldt CE, van Baal GC. Twin and family studies of the human elec-troencephalogram: a review and a meta-analysis. Biol Psychol. 2002;61:111-129.

26. Batty M, Taylor MJ. Visual categorization during childhood: an ERP study.Psychophysiology. 2002;39:482-490.

27. Pollak SD, Klorman R, Thatcher JE, Cicchetti D. P3b reflects maltreated chil-dren’s reactions to facial displays of emotion. Psychophysiology. 2001;38:267-274.

28. Coch D, Maron L, Wolf M, Holcomb PJ. Word and picture processing in chil-dren: an event-related potential study. Dev Neuropsychol. 2002;22:373-406.

29. Eimer M. Effects of face inversion on the structural encoding and recognition offaces: evidence from event-related brain potentials. Brain Res Cogn Brain Res.2000;10:145-158.

30. Schweinberger SR, Burton AM. Covert recognition and the neural system for faceprocessing. Cortex. 2003;39:9-30.

31. Posamentier MT, Abdi H. Processing faces and facial expressions. Neuropsy-chol Rev. 2003;13:113-143.

32. Olivares EI, Iglesias J, Rodriguez-Holguin S. Long-latency ERPs and recogni-tion of facial identity. J Cogn Neurosi. 2003;15:136-151.

33. Le Doux JE. The Emotional Brain. New York, NY: Simon & Schuster; 1996.34. Williams LM, Liddell BJ, Rathjen J, Brown KJ, Gray J, Phillips M, Young A, Gor-

don E. Mapping the time course of non-conscious and conscious perception of

fear: an integration of central and peripheral measures. Hum Brain Mapp. 2004;21:64-74.

35. Stevenson-Hinde J, Glover A. Shy girls and boys: a new look. J Child PsycholPsychiatry. 1996;37:181-187.

36. Cloninger CR, Przybeck TR, Svrakic DM, Wetzel RD. The Temperament and Char-acter Inventory. St Louis, Mo: Dept of Psychiatry, Washington University; 1994.

37. Dummit ES III, Klein RG, Tancer NK, Asche B, Martin J, Fairbanks JA. System-atic assessment of 50 children with selective mutism. J Am Acad Child AdolescPsychiatry. 1997;36:653-660.

38. Ambrosini PJ. Historical development and present status of the Schedule for Af-fective Disorders and Schizophrenia for School-age Children (K-SADS). J AmAcad Child Adolesc Psychiatry. 2000;39:49-58.

39. Achenbach TM. Manual for the Child Behavior Checklist 4-18 and Profile. Burl-ington: Dept of Psychiatry, University of Vermont; 1991.

40. Hollingshead AB. Four-Factor Index of Social Status. New Haven, Conn: Dept ofSociology, Yale University; 1975.

41. Heils A, Teufel A, Petri S, Stober G, Riederer P, Bengel D, Lesch KP. Allelic varia-tion of human serotonin transporter gene expression. J Neurochem. 1996;66:2621-2624.

42. Compton SN, Nelson AH, March JS. Social phobia and separation anxiety symp-toms in community and clinical samples of children and adolescents. J Am AcadChild Adolesc Psychiatry. 2000;39:1040-1046.

43. Camras LA, Allison K. Children’s understanding of emotional facial expressionsand verbal labels. J Nonverbal Behav. 1985;9:84-94.

44. Ekman P, Friesen WV. Pictures of Facial Affect. Palo Alto, Calif: Consulting Psy-chologists Press; 1976.

45. Dailey MN, Cottrell GW, Padgett C, Adolphs R. EMPATH: a neural network thatcategorizes facial expressions. J Cogn Neurosci. 2002;14:1158-1173.

46. Canli T, Sivers H, Whitfield SL, Gotlib IH, Gabrieli JD. Amygdala response to happyfaces as a function of extraversion. Science. 2002;296:2191.

47. Kestenbaum R, Nelson CA. Neural and behavioral correlates of emotion recog-nition in children and adults. J Exp Child Psychol. 1992;54:1-18.

48. Bentin S, Deouell LY. Structural encoding and identification in face processing:ERP evidence for separate mechanisms. Cogn Neuropsychol. 2000;17:35-54.

49. Waldman ID, Rowe DC, Abramowitz A, Kozel ST, Mohr JH, Sherman SL, Cleve-land HH, Sanders ML, Gard JMC, Stever C. Association and linkage of the do-pamine transporter gene and attention-deficit hyperactivity disorder in children:heterogeneity owing to diagnostic subtype and severity. Am J Hum Genet. 1998;63:1767-1776.

50. Marino C, Vanzin L, Giorda R, Frigerio A, Lorusso ML, Nobile M, Molteni M, Batta-glia M. An assessment of transmission disequilibrium between quantitative mea-sures of childhood problem behaviors and DRD2/TaqI and DRD4/48bp-repeatpolymorphisms. Behav Genet. 2004;34:495-502.

51. Dolan RJ. Emotional processing in the human brain revealed by functional imaging.In: Corbetta MD, Corbetta M, eds. The New Cognitive Neurosciences. New York,NY: Plenum Press; 2000:858-879.

52. Lang PJ, Bradley MM, Cuthbert BN. Emotion, motivation, and anxiety: brain mecha-nisms and psychophysiology. Biol Psychiatry. 1998;44:1248-1263.

53. Sprengelmeyer R, Rausch M, Eysel UT, Przuntek H. Neural structures associ-ated with recognition of facial expressions of basic emotions. Proc R Soc LondB Biol Sci. 1998;265:1927-1931.

54. Felmingham KL, Bryant RA, Gordon E. Processing angry and neutral faces in post-traumatic stress disorder: an event-related potentials study. Neuroreport. 2003;14:777-780.

55. Hariri AR, Weinberger DR. Imaging genomics. Br Med Bull. 2003;65:259-270.56. Haxby JV, Hoffman EA, Gobbini MI. The distributed human neural system for

face perception. Trends Cogn Sci. 2000;4:223-233.57. Kagan J, Reznick S, Snidman N, Johnson MO, Gibbons J, Gersten M, Bieder-

man J, Rosenbaum J. Origins of panic disorder. In: Ballenger JC, ed. Neurobi-ology of Panic Disorder. New York, NY: Alan R Liss Inc; 1990:71-87.

58. Battaglia M, Bajo S, Ferini-Strambi L, Brambilla F, Castronovo C, Vanni G, Bel-lodi L. Physiological and behavioral responses to minor stressors in offspringof patients with panic disorder. J Psychiatr Res. 1997;31:365-376.



Influence of the Serotonin Transporter Promoter Gene and Shyness on Children’s Cerebral Responses...

Documents

Transcript of Influence of the Serotonin Transporter Promoter Gene and Shyness on Children’s Cerebral Responses...