Jo

AM

a

ARR1AA

KWVFA

1

1eaaaLeeg

hWvfiaW

(taa

0d

Neuropsychologia 48 (2010) 1071–1078

Contents lists available at ScienceDirect

Neuropsychologia

journa l homepage: www.e lsev ier .com/ locate /neuropsychologia

ust another face in the crowd: Evidence for decreased detectionf angry faces in children with Williams syndrome

ndreia Santos ∗, Catarina Silva, Delphine Rosset, Christine Deruelleediterranean Institute of Cognitive Neuroscience – CNRS, 31, Chemin Joseph Aiguier, 13402 Marseille cedex 20, France

r t i c l e i n f o

rticle history:eceived 11 May 2009eceived in revised form8 November 2009

a b s t r a c t

The detection of social threat is crucial for adaptive behaviour. Previous studies have shown that angryfaces capture attention and are processed more efficiently than happy faces. While this anger superiorityeffect has been found in typical and atypical development, it is unknown whether it exists in individu-als with Williams syndrome (WS), who show reduced social fear and atypical sociability. In this study,

ccepted 4 December 2009vailable online 11 December 2009

eywords:illiams syndrome

isual search

children with WS searched for angry or happy target faces surrounded by 2, 5 or 8 distracters (happy orangry faces, respectively). Performance was compared to that of mental age-matched controls. Resultsrevealed no group differences for happy faces, however for angry faces, the WS, but not the control group,showed a significant performance decrease for the 8-distracters condition, indicating the absence of ananger superiority effect, in good agreement with evidence for abnormal structure and function in brain

cessi

acial emotion detectionnger superiority effect

areas for social threat pro

. Introduction

Detection of social threat has clear adaptive value (Darwin,872/1965) and social skills are critical for survival. There is a clearvolutionary advantage for humans who can efficiently recognizend detect social threat in their environment, since this abilityllows them to anticipate danger, prepare defensive behaviournd hence to escape potentially dangerous situations (e.g., Öhman,undqvist, & Esteves, 2001; Öhman & Mineka, 2001, 2003). Fromarly in life and throughout development, we learn to use facialxpressions as powerful sources of social information, crucial foruiding behaviour towards or away from interaction.

Within this framework, a unique social-behaviour phenotype,igh sociability and reduced social fear, found in individuals withilliams syndrome (WS) is of major interest. Most of these indi-

iduals are unusually friendly and show intriguing absence of fearrom strangers, with whom they are generally at ease and keen tonteract with (e.g., Jones et al., 2000). To date, the cognitive mech-nisms underlying this consistently reported hallmark feature ofS remain unclear.In addition to providing distinctive information about people

e.g., identity or gender), faces also transmit subtle signals relatedo emotion, trustworthiness, attractiveness, as well as intention. Asconsequence, faces are strong facilitators for social interaction andcrucial site to convey signals of potential social threat (e.g., Blair,

∗ Corresponding author. Tel.: +33 (0)4 91 16 45 19; fax: +33 4 91 77 45 19.E-mail address: andreia.santos@incm.cnrs-mrs.fr (A. Santos).

028-3932/$ – see front matter © 2009 Elsevier Ltd. All rights reserved.oi:10.1016/j.neuropsychologia.2009.12.006

ng in WS.© 2009 Elsevier Ltd. All rights reserved.

2003). Faces are special stimuli for humans from the earliest stagesof development. Newborns tend to look longer and to orient pref-erentially their attention to faces than other objects (e.g., Johnson,Dziurawiec, Ellis, & Morton, 1991). This increased and early interestfor faces is thought to be related to our extraordinary abilities forprocessing faces and facial emotions quickly and accurately (for areview see Bruce & Young, 1986). Relative to typically developinginfants, those with WS tend to overuse social engagement devicessuch as eye contact and to spend significantly more time focusingon faces than on objects (Bellugi, Lichtenberger, Jones, Lai, & St.George, 2000; Mervis & Bertrand, 1997).

Given the role of face processing skills in driving the develop-ment of social behaviour, several studies have attempted to exam-ine face processing in relation to hypersociability in WS. Whilenormal or near-normal levels of performance have been foundin tasks for face recognition and discrimination (Bellugi, Wang, &Jernigan, 1994; Deruelle, Rondan, Livet, & Mancini, 2003; Tager-Flusberg, Plesa Skwerer, Faja, & Joseph, 2003; Udwin & Yule, 1991),some argue that these skills are both delayed (e.g., Karmiloff-Smithet al., 2004) and deviant (e.g., Deruelle, Mancini, Livet, Cassé-Perrot,& de Schonen, 1999) in WS. In addition, electrophysiological studieshave shown that in spite of relatively good performance on iden-tity tasks, brain activity to faces is abnormally organized in WS(Grice et al., 2001; Mills et al., 2000). At the neural level, however, a

recent study has provided fMRI and ERP evidence of increased acti-vation to positive emotional expressions in WS (Haas et al., 2009).In facial emotion processing, studies have shown indistinguish-able performance between WS and mental age-matched groups onperception (Deruelle et al., 1999; Plesa-Skewerer, Faja, Schofield,

1 cholo

V2(cht

otanditga&wpiesaft1ha

ssiESt2raCa&

scwwccgwsdfo

2

2

(ioPwt15

072 A. Santos et al. / Neuropsy

erbalis, & Tager-Flusberg, 2006) and recognition (Gagliardi et al.,003) tasks. More recently, Santos, Milne, Rosset, and Deruelle2007) and Santos, Rosset, and Deruelle (2009) have shown thathildren and adults with WS can actually categorize emotions at aigher level than that expected from their intellectual level whenhese are expressed in human (real and cartoon) faces.

While facial expressions of emotion, in general, provide a wealthf information, angry faces, in particular, provide specific informa-ion on potential social threat. Indeed, these are universally reads critical cues for interpersonal threat and potent warning sig-als. Angry faces seem thus to hold a special status for typicallyeveloping individuals and some argue that humans are biolog-

cally prepared or “hard-wired” for the recognition of anger orhreat specifically (Öhman, 1993). In agreement with this, there is areat deal of evidence supporting the idea that angry faces capturettention and are processed promptly and efficiently (Vuilleumier

Schwartz, 2001). The empirical basis for specific mechanismsith regard to anger comes mostly from studies using visual searcharadigms, such as the “face-in-the-crowd” task. In this task partic-

pants are instructed to detect the presence or absence of a targetmotional face among a crowd of distracter faces. In a pioneeringtudy using this task, Hansen and Hansen (1988) have shown thatngry faces are detected more quickly and accurately than happyaces, independent of the number of distracters. Despite the limita-ions this study may have presented (e.g., Purcell, Stewart, & Skov,996), it has launched the idea of an anger superiority effect, whichas received support from several further studies (e.g., Öhman etl., 2001).

Consistent with the idea that the visual system may be sen-itive to the emotional valence of facial stimuli, several recenttudies found more efficient search for negatively than for pos-tively valenced face target sets among neutral distracters (e.g.,astwood, Smilek, & Merikle, 2001; Gerritsen, Frischen, Blake,milek, & Eastwood, 2008), as well as more efficient search for angryarget sets among happy distracters than vice-versa (e.g., Fox et al.,000; Horstmann & Bauland, 2006). Interestingly, this anger supe-iority effect holds even for neurodevelopmental disorders, such asutism and Asperger syndrome (Ashwin, Wheelwright, & Baron-ohen, 2006; Krysko & Rutherford, 2009), and anxiety disorders (forreview see Bar-Haim, Lamy, Pergamin, Bakermans-Kranenburg,Van IJzendoorn, 2007).The current study aimed at investigating whether the anger

uperiority effect can be found in children with WS, which showlinically reduced social fear and increased sociability. To this aim,e have used a “face-in-the-crowd” task and their performanceas compared to that of typically developing mental age-matched

ontrols (MA). We hypothesized that children with WS would pro-ess angry faces differently from MA controls, whereas no suchroup differences would be found for happy faces. Specifically,hile for MA children the presence of angry, but not happy, faces

hould be reported with the same ease regardless of the number ofistracters, for WS children the detection of both angry and happyaces should present increased difficulty with increasing numberf distracters.

. Methods

.1. Participants

Twenty-one children with WS (12 female and 9 male) aged 10–17 yearsM = 13.5; SD = 2.5) participated in this study. WS diagnosis was based on both clin-cal evaluation and a FISH test (fluorescent in situ hybridization) for microdeletion

n one copy of the gene for elastin on chromosome 7. All participants fulfilled thereus (1984) criteria for WS (e.g., characteristic facial appearance, low average birtheight, feeding difficulties, musculoskeletal problems, etc.) and were positive in

he FISH test. Mental age (MA), inferred from IQ measures (WAIS-III (Wechsler,997) and WISC-III (Wechsler, 1996), according to the subject’s age), ranged fromto 15 years (M = 8.3; SD = 2.6). Full-scale IQ profile (M = 60.4; SD = 12.7; range:

gia 48 (2010) 1071–1078

47–93) was characterized by a significant dissociation between verbal (M = 67.8;SD = 14.3; range: 50–101) and performance (M = 62.9; SD = 10.5; range: 46–87) IQscores (F (1,20) = 7.62, p < .01), consistent with previous studies on WS (e.g., Jarrold,Baddeley, & Hewes, 1998). Participants were recruited via the Department of Pedi-atric Neurology at a local hospital (La Timone, Marseille, France) and via RegionalWilliams Syndrome Associations. At the time of testing, they all attended schools orspecialized centres for individuals with developmental delay.

This study also included a group of 21 typically developing individuals, aged5–15 years (M = 8.8; SD = 2.9), individually matched to WS participants on gender(12 female and 9 male) and MA (F (1,40) = .25, p > .6). Typically developing childrenwere recruited via local schools and day-care centres and they all attended normalclasses corresponding to their age level. Teachers were asked to select these childrenfrom the average level of their class thus avoiding inclusion of advanced or delayedchildren relative to their age. None of the participants had overt physical handicap,known neurological/psychiatric deficits or history of learning difficulties.

All participants were native French speakers and had normal or corrected-to-normal audition and vision. Informed consent was obtained for all subjects and theexperimental procedure was approved by the local ethics committee.

2.2. Stimuli and procedure

Stimuli comprised cartoon faces, displaying either a happy or an angry facialexpression. The same identity was used in both emotional valences. The faces con-sisted of coloured line drawings presented against a white background (see Fig. 1).Cartoon rather than real faces were used to avoid problems in equating real facesfor confounding visual features such as shadows (e.g., Purcell et al., 1996). Impor-tantly, Santos et al. (2009) have shown that individuals with WS have no difficultiesto process emotions displayed in cartoon faces. Furthermore, previous studies usingschematic faces have shown that emotional expressions are readily recognized fromsimple eyebrow and mouth line drawings (e.g., Magnussen, Sunde, & Dyrnes, 1994).

This study comprised two visual search tasks, examining the participants searchfor emotional faces in different sized arrays of faces (2, 4 or 8). Participants hadeither to detect the presence/absence of one angry (Task 1) or one happy face (Task2) within a “crowd” of angry or happy faces, respectively. All trials consisted of 3 × 3matrices (801 × 700 pixels size at the screen). In each task, half of the trials (target-present trials) consisted of one target face (discrepant face) plus other 2, 5 or 8background faces (all similar; crowd faces). In both tasks, the target could occur atany of the 9 positions in the matrix (see Fig. 1), resulting in a total of 36 differentmatrices containing a target (18 target-present trials for each task, 6 trials for eachmatrix size). The position of the target face was also randomly assigned across trialsand subjects. In the other half of the trials (36 target-absent trials, 18 trials for eachtask, 6 for each matrix size) all faces were similar (no discrepant face). The distancebetween the faces (target and the background faces) was constant across all thesecombinations (see Fig. 1).

Visual stimuli were presented on a 19-in. monitor connected to Macintosh lap-top computer. SuperLab 4.5 experiment software (Cedrus Corporation, USA) wasused to initiate trials and record responses and reaction times (RTs). Participantswere tested individually in a quiet room at the Institute of Cognitive Neuroscienceof the Mediterranean (CNRS, Marseille, France) or at their home. Participants wereseated approximately 60 cm from the computer screen and they were asked to fix-ate the centre of the screen. Participants were asked to detect a discrepant face ina matrix of faces. More precisely they were to respond whether there was a happyface (Task 1) or an angry face (Task 2) in the crowd. Responses were given by press-ing one of two coloured keys (red = no; green = yes) on a response box designed foruse with the Macintosh laptop and the Superlab program. Before the experimentalsession began, all participants were presented with a practice session to familiarisethemselves with the task (3 trials for each task, one corresponding to each matrixsize) and to ensure they understood the instructions. Each trial consisted of a pre-sentation of a fixation cross on an otherwise blank screen for 500 ms, followed bythe matrix display and the start of trial timing. The display time was terminatedby the participants’ response. Half of the participants were presented with Task 1first and the other half started with Task 2. In order to avoid interference with taskinstructions, the two tasks were separated by a 2-h interval, during which childrenwatched a movie.

2.3. Data analyses

Analyses of variance (ANOVAs) with repeated measures were used forall statistical tests, and all p-values reported below were adjusted with theGreenhouse–Geisser epsilon correction for non-sphericity. The uncorrected degreesof freedom and the probability level after correction are reported. Main and/or inter-action effects not significant (p > .05) are not be reported. Fisher LSD tests were usedfor all post hoc comparisons. Correlation analyses were computed using Pearson rtests. Only RTs for correct responses were considered in the analyses.

3. Results

In order to ensure that our paradigm could provide data consis-tent with previous studies on visual search we have first analysed

A. Santos et al. / Neuropsychologia 48 (2010) 1071–1078 1073

F 2 (anF

aivt

T(rcto(aTsoptwiMp

[bDosttnMvtSv(

ig. 1. Example of the stimuli used in Task 1 (angry face target; A, B and C) and Task) conditions.

ccuracy and RTs for all participants using three-way ANOVAsncluding Trial Type (target-present vs. target-absent), Task (happys. angry target face), and Number of Distracters (2, 5, and 8 condi-ions) as within-subjects factors.

Results for accuracy scores revealed a significant main effect ofrial Type [F (1,41) = 35.41, p < .001] and Number of Distracters [F2,82) = 5.76, p = .004]. However, the interaction Trial Type × Taskeached significance [F (1,41) = 10.99, p = .002]. Results of post hocomparisons computed to further track this interaction revealedhat task-related differences in accuracy, i.e., higher accuracy scoresn the angry- (M = 5.27; SD = .17) than on the happy-target taskM = 4.96; SD = .18, p = .006), were only found for target-present tri-ls (Mtask-angry = 5.53; SD = .11; Mtask-happy = 5.72; SD = .13, p > .05).his finding is unsurprising since target-absent trials included theame stimuli in the two tasks. Regarding the Trial Type × Numberf Distracters interaction [F (2,82) = 10.88, p < .001], results ofost hoc analyses showed that significant differences betweenhe number of distracters only occurred on target-present trials,here accuracy scores decreased as the number of distracters

ncreased (M2distracters = 5.44; SD = .14; M5distracters = 5.13; SD = .16;8distracters = 4.77; SD = .19; 2 vs. 5, p = .009; 2 vs. 8, p < .001; 5 vs. 8,= .003).

Results for RTs showed a significant main effect of Trial TypeF (1,41) = 34.18, p < .001], Task [F (1,41) = 12.33, p < .001] and Num-er of Distracters [F (2,82) = 51.72, p < .001]. The Task × Number ofistracters [F (2,82) = 8.57, p < .001] and the Trial Type × Numberf Distracters [F (2,82) = 17.7, p < .001] interactions were alsoignificant. Post hoc comparisons revealed that the number of dis-racters increased significantly the RTs in both tasks and trialypes. However, these differences were relatively more promi-ent in the happy-target task (M2distracters = 1600; SD = 136 ms;5distracters = 1716; SD = 144 ms; M8distracters = 1854; SD = 169 ms; 2

s. 5, p < .001; 2 vs. 8, p < .001; 5 vs. 8, p < .001) than in the angry-arget task (M2distracters = 1720; SD = 156 ms; M5distracters = 1964;D = 165 ms; M8distracters = 2195; SD = 175 ms; 2 vs. 5, p = .003; 2s. 8, p < .001; 5 vs. 8, p < .001), as well as in target-absentM2distracters = 1567; SD = 130 ms; M5distracters = 1678; SD = 146 ms;

gry face target; D, E and F) for the 2 (A and D), 5 (B and E) and 8 distracter 5 (C and

M8distracters = 1777; SD = 155 ms; 2 vs. 5, p < .001; 2 vs. 8, p < .001;5 vs. 8, p < .001) than in target-present trials (M2distracters = 1752;SD = 153 ms; M5distracters = 2002; SD = 164 ms; M8distracters = 2272;SD = 185 ms; 2 vs. 5, p = .003; 2 vs. 8, p < .001; 5 vs. 8, p = .009).

Given that the main hypothesis of the study concerned perfor-mance differences between WS and MA on target-present trials inparticular, further analyses were conducted for the two trial typesseparately.

3.1. Target-present trials

Accuracy scores were analysed using a three-way ANOVAincluding Group (WS vs. MA) as between-subjects factor andTask (happy vs. angry target face), and Number of Distracters(2, 5, and 8 conditions) as within-subjects factors. Results ofthis analysis revealed no significant main effect of Group [F(1,40) = 1.99, p > .16], but a significant main effect of the Task [F(1,40) = 6.84, p < .01], with participants being more accurate indetecting angry (M = 5.27; SD = .17) than happy faces (M = 4.96;SD = .18). The main effect of the Number of Distracters was alsofound significant [F (2,80) = 10.56, p < .001], with accuracy decreas-ing with increasing number of distracter faces (M2distracters = 5.44;SD = .14; M5distracters = 5.13; SD = .16; M8distracters = 4.77; SD = .19).Results of post hoc comparisons showed significant differences intarget face detection between sets displaying 2 vs. 5 distracters(p = .03), 2 vs. 8 distracters (p < .001) and 5 vs. 8 distracters (p < .01),indicating that the number of distracters significantly disrupteddetection performance. Most importantly, there was a significantGroup × Task × Number of Distracters interaction [F (2,80) = 3.86;p = .03]. The post hoc comparison revealed significant differencesbetween groups only for angry face detection within 8 face dis-tracters sets (p < .04). These results are illustrated in Fig. 2.

In order to further track these results, we have conducted two-way ANOVAs on the accuracy scores for each group separately withTask (happy vs. angry target face) and Number of Distracters (2, 5,and 8 conditions) as the within-subjects factors. Results of analysisfor the WS group showed a main effect of the Number of Distracters

1074 A. Santos et al. / Neuropsychologia 48 (2010) 1071–1078

Fig. 2. Response patterns (means and standard errors) observed for WS and MAga(

[ptDaD[tif(pMwisgathnimToa

yMfsT[iMRc2vtoefe(ppaf

children and adults (e.g., LoBue, 2009; Mather & Knight, 2006;

roups on target-present trials. Accuracy scores are presented for each group (WSnd MA) in each Task (happy and angry) as a function of the Number of Distracters2, 5 and 8).

F (2,40) = 6.59, p = .003], no main effect of the Task [F (1,20) = 1.75,> .20], and no interaction between the two factors, i.e., between

he emotional valence of the face to be detected and the Number ofistracters displayed [F (2,40) = .11, p > .90]. Conversely, results ofnalysis for the MA group showed a main effect of the Number ofistracters displayed [F (2,40) = 3.98, p = .03], a main effect of Task

F (1,20) = 5.66, p = .03], and a significant interaction between thewo factors [F (2,40) = 9.79, p < .001]. Results of post hoc compar-sons showed no significant differences between happy and angryace detection when the set displayed 2 (p > .36) and 5 distractersp < .17), whereas this difference was significant when the set dis-layed 8 distracters (p < .001). These results indicate that for theA group, angry faces were detected more efficiently, while thatas not the case for happy faces, which become more difficult with

ncreasing number of distracters. To further support this hypothe-is, we conducted one-way ANOVAs on the accuracy scores for eachroup in each task separately (WS happy, MA happy, WS angry, MAngry) including Number of Distracters (2, 5, and 8 conditions) ashe within-subjects factor. In line with results reported above, in theappy task the main effect of Number of Distracters was found sig-ificant for both WS (p = .04) and MA (p > .001) groups. By contrast,

n the angry task only the WS group showed a significant perfor-ance decrease with the increasing number of distracters (p = .03).

he MA group, by contrast, showed a non-significant main effectf Number of Distracters (p > .71), with target detection being asccurate in sets displaying 2, 5 or 8 distracters.

RTs were analysed using a similar procedure to that used to anal-se accuracy scores (a three-way ANOVA including Group (WS vs.A)) as between-subjects factor and Task (happy vs. angry target

ace), and Number of Distracters (2, 5, and 8 conditions) as within-ubjects factors. Results revealed a significant main effect of theask [F (1,40) = 18.67, p < .001] and of the Number of DistractersF (2,80) = 13.22, p < .001], with RTs increasing with increas-ng number of distracter faces (M2distracters = 1567; SD = 91 ms;

5distracters = 1678; SD = 103 ms; M8distracters = 1777; SD = 109 ms).esults of post hoc comparisons (Fisher LSD test) confirmed signifi-ant differences in target face detection between sets displayingvs. 5 distracters (p = .008), 2 vs. 8 distracters (p < .001), and 5

s. 8 distracters (p = .02). This indicates that the number of dis-racters significantly interfered with the target detection speedf both groups (see Fig. 3), in agreement with the idea thatach added distracter requires additional processing time (see,or instance, Gerritsen et al., 2008). Importantly, neither the mainffect of Group [F (1,40) = 1.73, p > .19] nor the Group × Task [F1,40) = .75, p > .39], Group × Number of Distracters [F (2,80) = .28,

> .76] and Group × Task × Number of Distracters [F (2,80) = .57,> .56] were found significant. These results allow us to excludespeed-accuracy trade-off as an explanation for the differences

ound between WS and MA in accuracy scores.

Fig. 3. RTs for correct responses (means and standard errors) observed for WS andMA groups on target-present trials. RTs (ms) are presented for each group (WS andMA) in each Task (happy and angry) as a function of the Number of Distracters (2, 5and 8).

3.2. Target-absent trials

Analyses of accuracy scores and of RTs were conducted fol-lowing the same procedure as that used for target-present trials(three-way ANOVA including Group (WS vs. MA) as the between-subject factor, and Task (happy vs. angry) and Number of Distracters(2, 5, and 8 conditions) as within-subject factors).

Results for accuracy scores showed a significant main effectof Task [F (1,40) = 5.40, p = .03], with participants being moreaccurate on the happy- (M = 5.72; SD = .13) than on the angry-target task (M = 5.53; SD = .11). The Group × Task × Number ofDistracters interaction [F (2,80) = 4.24, p = .02] was also foundsignificant. However, results of further post hoc comparisonsrevealed no significant differences between groups in anyof the conditions (phappy 2distracters > .12; phappy 5distracters > .60;phappy 8distracters > .86; pangry 2distracters > .73; pangry 5distracters > .09;pangry 8distracters > .60).

Results for RTs revealed a significant main effect of Task [F(1,40) = 6.18, p = .02] and of Number of Distracters [F (2,80) = 57.69,p < .001]. The Task × Number of Distracters interaction also reachedsignificance [F (2,80) = 6.05, p = .003], with significant differencesbetween 2 vs. 5 distracters, 2 vs. 8 distracters and 5 vs. 8 distracterson target-absent trials of the angry task (M2distracters = 1773 ms;SD = 117 ms; M5distracters = 2148; SD = 135 ms; M8distracters = 2451;SD = 132 ms; 2 vs. 5, p < .001; 2 vs. 8, p < .001; 5 vs. 8, p < .001),whereas only 2 vs. 8 distracters and 5 vs. 8 distracters trials weresignificantly different on the happy task (M2distracters = 1733 ms;SD = 126 ms; M5distracters = 1856; SD = 125 ms; M8distracters = 2094;SD = 151 ms; 2 vs. 5, p = .07; 2 vs. 8, p < .001; 5 vs. 8,p < .001).

Importantly, no main effect of Group nor Group × Task,Group × Number of Distracters or Group × Task × Number of Dis-tracters interactions were found (except for accuracy scores wherepost hoc comparisons failed to show group differences across all theconditions). Because the main hypotheses of this study focused ongroup differences on target-present trials in particular, no furtheranalyses were conducted on target-absent trials.

Finally, we have computed correlation analyses for each groupseparately in order to investigate whether age or low IQ lev-els in individuals with WS had an impact on the patterns ofperformance described above. Results of these analyses failedto reveal any significant correlation between age and perfor-mance for both WS and MA groups (all ps > .05). This is in linewith previous studies using similar paradigms showing simi-lar patterns of performance between typically developing young

Ruffman, Ng, & Jenkin, 2009). In addition, IQ scores were not foundcorrelated to WS performance (all ps > .05), suggesting that men-tal retardation did not account for atypical performance in thisgroup.

cholo

4

wfteaHiaSdh

WdcgnwftTegswtenqwwaWa

g

d(tahwrwtinsbtoclvM&&ofssr

A. Santos et al. / Neuropsy

. Discussion

The aim of the present study was to determine whether childrenith WS would show an anger superiority effect when looking for

aces in a crowd. In the last decades several studies have used thisype of paradigm (Eastwood et al., 2001; Fox et al., 2000; Gerritsent al., 2008; Horstmann & Bauland, 2006) attempting to replicatend further understand results of the ground-breaking study byansen and Hansen (1988). In line with previous visual search stud-

es studies, we have found that increasing number of distracters hasnegative impact on search efficiency (see Frischen, Eastwood, &

milek, 2008 for a review). Importantly, this effect was more evi-ent for target-absent than target-present trials, as well as in theappy than in the angry task, for overall RTs in particular.

Regarding the so-called anger superiority effect in children withS relative to MA controls, findings of this study revealed that chil-

ren with WS detect angry faces in a deviant fashion relative to MAhildren. In agreement with the anger superiority effect, for the MAroup, detection of angry faces was found to be independent of theumber of distracters, i.e., they detected angry target faces as easilyhen these were surrounded by 2, 5, or 8 distracters. By contrast,

or the WS group the increasing number of distracters had a nega-ive impact on performance in both happy- and angry-target trials.herefore, children with WS failed to show the anger superiorityffect; rather, they processed angry targets as they did happy tar-ets, in which they did not differ from MA children. Importantly,ignificant group differences were found when angry target facesere presented on the larger displays (8 distracters). If a particular

arget, in this case an angry face, has a special salience, its pres-nce will be readily detected when surrounded by either few orumerous distracters. By contrast, targets that do not possess suchuality require additional cognitive resources, because the subjectill need to process all items until the target is located. Consistentith this idea, group differences found in this study suggest that

ngry faces do not possess this special salience for children withS, as these were not detected more efficiently when displayed

mong distracters.But what can explain these differences between WS and MA

roups in angry face detection?One may hypothesize that task complexity accounts for these

ifferences. Individuals with WS have limited cognitive resourcestheir IQ levels usually fall about two standard deviations belowhe general population mean, e.g., Udwin, Yule, & Martin, 1987)nd may thus struggle with complex paradigms as the one usedere. Although one of the most common features of individualsith WS is mental retardation, it is important to note that no cor-

elation between IQ and performance was found for this group,hich renders unlikely the possibility that lower performance in

he WS group relative to controls was due to differences on overallntellectual ability. On the other hand, mean IQ scores in WS doesot completely capture the regularly found dissociation betweenevere visuo-spatial deficits contrasting with relatively spared ver-al abilities (e.g., Mervis et al., 2000). However, the early hypothesishat WS participants have an overall impairment across most areasf visual-cognition (Bellugi, Sabo, & Vaid, 1988) has been recentlyhallenged by studies showing that WS visuo-spatial deficits areimited to the visuo-constructive domain and mostly spare theisuo-perceptive domain (e.g., Farran & Jarrold, 2003; Rondan,ancini, Livet, & Deruelle, 2003; Rondan, Santos, Mancini, Livet,Deruelle, 2008; but see also Mobbs et al., 2007 and Pani, Mervis,Robinson, 1999, showing abnormalities in global/local responses

n non-constructive tasks in WS). In addition, a recent study usingMRI-based retinotopic mapping and cortical surface models hashown normal recruitment of primary visual cortex during visualtimulation in WS (Olsen et al., 2009). Taken together these findingsender unlikely the hypothesis that the current results are due to

gia 48 (2010) 1071–1078 1075

group differences in visuo-perceptive functioning. Previous studieson facial emotion processing showing similar performance in WSand chronological (Santos et al., 2007, 2009) or mental age-matchedgroups (Deruelle et al., 1999; Gagliardi et al., 2003; Plesa-Skewereret al., 2006), also provide arguments against the idea that groupdifferences might be related to an inability to infer anger from thetarget faces. Furthermore, facial stimuli used in our visual searchtask were carefully selected so that emotional expressions could bereadily recognized from simple eyebrow and mouth line drawings(e.g., Magnussen et al., 1994) and respected the criteria typicallyuse to define a face as angry, i.e., V-shaped eyebrows and a nar-row, down-curved mouth (Lundqvist, Esteves, & Öhman, 1999;Lundqvist, Esteves, & Öhman, 2004). Using these cartoon stimuli, ifone finds differences in the detection of happy and angry face tar-gets these cannot be simply attributed to low level visual confoundsas could be the case using real faces (Purcell et al., 1996). Finally,because visual search tasks commonly implicate the allocation ofattention, one may argue that an attentional deficit in WS underliestheir poor performance on the 8-distracters anger condition. How-ever, the current findings do not indicate that WS individuals haveoverall difficulties to detect emotional faces within a crowd. Rather,surprising relatively normal performance (mental age equivalent)was found for the WS group in all but the 8-distracters anger con-dition. The fact that no group differences were found in any of thehappy conditions is indeed a major finding of this study, suggestingthat WS children do have the ability to detect happy faces withina crowd as easily as mental age-matched controls. This countersthe idea of a deficit to discriminate targets from non-targets onvisual search tasks in WS, which has, however, been proposed bya study on toddlers with WS using a different methodology (visualsearch for multiple targets on a touch-screen) than the current one(Scerif, Cornish, Wilding, Driver, & Karmiloff-Smith, 2004). Further-more, both WS and MA groups showed increased difficulties todetect the presence of happy targets as a function of the increasingnumber of distracters. This finding suggests similar effects on cog-nitive resources in both groups for this stimulus class. However, thediscrepancy to the previous study might be resolved if target dis-crimination shows a deviant developmental trajectory in WS (seeKarmiloff-Smith et al., 2004).

By contrast, on the anger detection condition, a significantly dif-ferent pattern of performance was found for MA and WS groups.While MA children showed a homogeneous detection ability acrossthe three sets size displays, children with WS rather showed anatypical pattern of performance, in which accuracy decreased asthe number of distracters increased. Because accuracy differenceswere found between the two groups on the largest size crowdcondition (8 distracters), one could argue that WS group’s poorperformance was due to a speed-accuracy trade-off. Nonetheless,analyses on RTs failed to show any significant group differences inall the conditions, including the 8-distracters anger condition. Thisrules out the hypothesis that WS group accuracy pattern in anyof the experimental conditions was influenced, either favoured orimpoverished, by differences in RTs relative to MA participants. Inaddition, and consistent with the literature, overall RTs were foundto increase with the increasing number of items in the displays(e.g., Ashwin et al., 2006; Gerritsen et al., 2008). For MA children,the fact that the ability to detect angry faces was found indepen-dent of the number of distracters is evidence for an anger superiorityeffect in typical development. This is in line with previous stud-ies, which have, for the most part, been conducted with typicallydeveloping adults (e.g., Eastwood et al., 2001; Fox et al., 2000;

Gerritsen et al., 2008; Horstmann & Bauland, 2006). Recently, how-ever, LoBue (2009) showed that young children share this samebias in the detection of emotional faces in a crowd. To our knowl-edge, the present study is the first examining the detection ofthreatening facial expressions in WS relative to typically devel-

1 cholo

obi(sadstfHheacaa(Bt

doehoampostfpfdvdiir(t(tsdtpW

uuacaocatFtdl

w

076 A. Santos et al. / Neuropsy

ping controls. Surprisingly, while an anger superiority effect aseen found even in developmental disorders characterized by atyp-

cal social behaviour and difficulties in socio-emotional processingAshwin et al., 2006; Krysko & Rutherford, 2009), our findingsuggest that for WS children angry faces are not more than justnother face in the crowd. In the case of autism spectrum disor-ers (ASD), often considered as mirror image of WS in terms ofocial behaviour (Tager-Flusberg, Plesa Skewerer, & Joseph, 2006),he anger superiority effect appears to co-exist with abnormal rapidacial reactions to anger and fear faces (Beall, Moody, McIntosh,epburn, & Reed, 2008). Specifically, research addressing this issueave found evidence for impaired recognition of negative basicmotions in autism (Ashwin et al., 2006), and for a threat-detectiondvantage in individuals with ASD, however relatively diminishedompared to controls (Krysko & Rutherford, 2009). Moreover, thisnger superiority effect appears to co-exist with and abnormalmygdala activation for facial expressions processing such as threatAshwin, Baron-Cohen, Wheelwright, O’Riordan, & Bullmore, 2007;aron-Cohen et al., 1999; Critchley et al., 2000), as well as with ges-ures expressing fear (Grèzes, Wicker, Berthoz, & de Gelder, 2009).

As mentioned above, visuo-perceptual or attentional overalleficits do not seem to provide a reliable account for the absencef an anger superiority effect in WS. The lack of significant differ-nces between the groups in happy face detection supports theypothesis that the pattern of results observed is also not due to anverall facial emotion processing deficit in WS, but possibly to antypical way of processing social threat. An alternative explanationay thus come from more social mechanisms. The human face and

articularly facial expressions are undoubtedly powerful sourcesf social information. Studies using eye-tracking techniques havehown for instance that WS individuals spend more time than isypical scanning social scenes displaying people interacting, andaces (Riby & Hancock, 2008, 2009). Happy faces are commonlyerceived as safe and motivating in social contexts, whereas angryaces are rather perceived as warning signals of potential socialanger (Öhman et al., 2001; Öhman & Mineka, 2001, 2003). Pre-ious models of face processing in WS have indeed suggested aissociation between positive and negative emotions processing

n WS. Such dissociation is most relevant when interpreting find-ngs of this study. At the behavioural level, WS individuals appear toate happy, but not angry faces as more approachable than controlsFrigerio et al., 2006). At the neural level, increased amygdala reac-ivity to positive (happy), while absent or attenuated to negativefear) facial expressions has been found in WS individuals relativeo controls (Haas et al., 2009). Findings of the current study provideupport to these models, putting forward convergent evidence forifferences in happy and angry faces detection in WS. Altogetherhese findings suggest a link between a positive bias on emotionrocessing and the well-known positive social bias characterizingS.Of major interest for research in this domain is indeed the WS

nique social behaviour profile. Together with high sociability andnusual interest towards social information, WS individuals showppetitive drive to interact with others, even those they objectivelyonsider as not approachable (Frigerio et al., 2006). Putatively, thisbsence of fear from strangers is related to atypical perceptionf social threat and decreased detection of threatening faces in arowd. On the other hand, the detection of social threat is a crucialbility for one to respond to social demands in an adaptive and pro-ective manner (Öhman et al., 2001; Öhman & Mineka, 2001, 2003).indings of this study thus suggest a link between reduced ability

o detect social threat and deviant sociability in WS, although theseo not allow clear-cut conclusions on the direction of this possible

ink.The detection of social threat is controlled by a complex net-

ork of neural structures, within which the amygdala is thought

gia 48 (2010) 1071–1078

to play an important role in producing and coordinating appro-priate responses to threatening stimuli (for reviews see Adolphs,2003, 2008; Aggleton, 2000; Armony & LeDoux, 2000). Some arguethat the amygdala operates primarily as a rapid-response “fear-module”, shaped and constrained by evolutionary contingencies,that is preferentially activated by threat-relevant stimuli in a phy-logenetic sense (Öhman & Mineka, 2001). The amygdala seemsthus to be central to socially protective neural processing throughthe monitoring of environmental events such as danger (Adolphs,2003, 2008). Early lesion studies have provided strong evidencefor a link between deficient facial threat detection and amygdaladamage (e.g., Adolphs, Tranel, Damasio, & Damasio, 1995; Adolphset al., 1999). More recently, Calder et al. (1996) have also foundimpaired recognition of fear from facial expressions following bilat-eral amygdala damage. Moreover, functional imaging studies havefound activation of the amygdala in typically developing individ-uals when viewing fearful faces (Breiter et al., 1996; Morris et al.,1996; Whalen et al., 2001).

Of great interest within this framework, amygdala reactivity tothreatening faces appears significantly diminished in individualswith WS (Meyer-Lindenberg et al., 2005). Meyer-Lindenberg et al.(2005) have conducted an fMRI study to investigate amygdala func-tioning and regulation in WS relative to controls. The authors usedtwo tasks requiring processing of threatening stimuli, either facesor scenes, known to engage amygdala (Hariri, Tessitore, Mattay,Fera, & Weinberger, 2002). Interestingly, these two tasks broughtforth significant group differences in amygdala activation, with WSshowing reduced amygdala activation for threatening faces only.These results were found to be clearly in agreement with the WSatypical social profile: diminished amygdala reactivity to threaten-ing faces being putatively related to absence of fear from strangersand consequent social disinhibition. Similar reactivity differencesbetween groups were also found in the prefrontal cortex, wherecontrols differentially activated orbitofrontal, dorsolateral- andmedial-prefrontal cortex for threatening faces, while WS individu-als showed no such specific pattern of activation. These findingsshowing a dysregulation of amygdala-prefrontal systems in WSprovide important insights for the interpretation of our findings.Since this system is involved in social fear signalling (e.g., Schultz,2005) it might be its dysregulation that underlies the absence of theanger superiority effect found here for children with WS. In addition,a recent analysis found that interactions between the fusiform facearea (FFA) and amygdala are abnormally decreased in WS (Sarpalet al., 2008). Therefore, even if primary processing of faces, whichis supported by ventral stream and specifically FFA function, isspared in WS (Meyer-Lindenberg et al., 2004), decreased saliencesignals from amygdala for angry faces in WS (Haas et al., 2009;Meyer-Lindenberg et al., 2005) fed back into FFA could be a neu-ral mechanism through which the anger superiority effect in WSis suppressed. In addition to this face-specific mechanism, alloca-tion of attentional resources in visual pop-out priming has beenfound related to activity in the intraparietal sulcus (Kristjánsson,Vuilleumier, Schwartz, Macaluso, & Driver, 2007), a region in whichprofound structural and functional abnormalities have been foundin WS (Kippenhan et al., 2005; Meyer-Lindenberg et al., 2004).Although these neurobiological explanations require confirmationfrom further neuroimaging studies, findings reported here mayprovide a point of departure for future research on the brain andcognitive mechanisms underlying these intriguing aspects of theWS social profile.

Acknowledgements

We are grateful to all the participants, their parents andthe Regional Williams Syndrome Associations. A. Santos and C.

cholo

SSc

R

A

A

A

A

A

A

A

A

B

B

B

B

B

B

B

B

B

C

C

D

D

D

E

F

F

F

F

A. Santos et al. / Neuropsy

ilva were supported by a grant from the FCT-MCTES (Portugal,FRH/BD/18820/2004 and SFRH/BD/30355/2006, respectively) toonduct this study.

eferences

dolphs, R. (2003). Cognitive neuroscience of human social behaviour. NatureReviews Neuroscience, 4(3), 165–178.

dolphs, R. (2008). Fear, faces, and the human amygdala. Current Opinion in Neuro-biology, 18(2), 166–172.

dolphs, R., Tranel, D., Damasio, H., & Damasio, A. R. (1995). Fear and the humanamygdala. The Journal of Neuroscience, 15(9), 5879–5892.

dolphs, R., Tranel, D., Hamann, S., Young, A. W., Calder, A. J., Phelps, E. A., et al.(1999). Recognition of facial emotion in nine individuals with bilateral amygdaladamage. Neuropsychologia, 37(10), 1111–1117.

ggleton, J. (2000). The amygdala: A functional analysis (2nd ed.). New York: OxfordUniversity Press.

rmony, J. L., & LeDoux, J. E. (2000). How danger is encoded: Toward a systems,cellular, and computational understanding of cognitive-emotional interactionsin fear. In M. S. Gazzaniga (Ed.), The cognitive neurosciences (pp. 1067–1079).Cambridge: The MIT Press.

shwin, C., Baron-Cohen, S., Wheelwright, S., O’Riordan, M., & Bullmore, E. T. (2007).Differential activation of the amygdala and the ‘social brain’ during fearful face-processing in asperger syndrome. Neuropsychologia, 45(1), 2–14.

shwin, C., Wheelwright, S., & Baron-Cohen, S. (2006). Finding a face in the crowd:Testing the anger superiority effect in Asperger Syndrome. Brain and Cognition,61(1), 78–95.

ar-Haim, Y., Lamy, D., Pergamin, L., Bakermans-Kranenburg, M. J., & Van IJzen-doorn, M. H. (2007). Threat-related attentional bias in anxious and non-anxiousindividuals: A meta-analytic study. Psychological Bulletin, 133(1), 1–24.

aron-Cohen, S., Ring, H. A., Wheelwright, S., Bullmore, E. T., Brammer, M. J., Sim-mons, A., et al. (1999). Social intelligence in the normal and autistic brain: AnfMRI study. European Journal of Neuroscience, 11(6), 1891–1898.

eall, P. M., Moody, E. J., McIntosh, D. N., Hepburn, S. L., & Reed, C. L. (2008). Rapidfacial reactions to emotional facial expressions in typically developing childrenand children with autism spectrum disorder. Journal of Experimental Child Psy-chology, 101(3), 206–223.

ellugi, U., Lichtenberger, L., Jones, W., Lai, Z., & St. George, M. (2000). I. The neu-rocognitive profile of Williams syndrome: A complex pattern of strengths andweaknesses. Journal of Cognitive Neuroscience, 12(Suppl. 1), 7–29.

ellugi, U., Sabo, H., & Vaid, V. (1988). Spatial defects in children with Williamssyndrome. In J. Stiles-Davis, M. kritchevsky, & U. Bellugi (Eds.), Spatial cognition:Brain bases and development (pp. 273–298). Hillsdale, NJ: Lawrence ErlbaumAssociates.

ellugi, U., Wang, P., & Jernigan, T. (1994). Higher cortical functions: Evidence fromspecific genetically based syndromes of disorder. In S. Broman, & J. Graffman(Eds.), Cognitive deficits in developmental disorders: Implications for brain function(pp. 23–56). Hillsdale, NJ: Lawrence Erlbaum.

lair, R. J., (2003). Facial expressions, their communicatory functions and neuro-cognitive substrates. Philosophical Transactions of the Royal Society B:Biological Sciences, 358 (1431), 561-572.

reiter, H. C., Etcoff, N. L., Whalen, P. J., Kennedy, W. A., Rauch, S. L., Bruckner, R. L.,et al. (1996). Response and habituation of the human amygdala during visualprocessing of facial expression. Neuron, 17(5), 875–887.

ruce, V., & Young, A. (1986). Understanding face recognition. British Journal of Psy-chology, 77(Pt 3), 305–327.

alder, A. J., Young, A. W., Rowland, D., Perrett, D. I., Hodges, J. R., & Etcoff, N. L.(1996). Facial emotion recognition after bilateral amygdala damage: Differen-tially severe impairment of fear. Cognitive Neuropsychology, 13(5), 699–745.

ritchley, H. D., Daly, E. M., Bullmore, E. T., Williams, S. C. R., Van Amelsvoort, T.,Robertson, D. M., et al. (2000). The functional neuroanatomy of social behaviour:Changes in cerebral blood flow when people with autistic disorder process facialexpressions. Brain, 123(11), 2203–2212.

arwin, C. (1872/1965). The expression of emotions in man and animals. Chicago:University of Chicago Press.

eruelle, C., Mancini, J., Livet, M. O., Cassé-Perrot, C., & de Schonen, S. (1999). Con-figural and local processing of faces in children with Williams syndrome. Brainand Cognition, 41(3), 276–298.

eruelle, C., Rondan, C., Livet, M. O., & Mancini, J. (2003). Exploring face processingin Williams syndrome. Cognition, Brain, Behavior, 7, 157–172.

astwood, J. D., Smilek, D., & Merikle, P. M. (2001). Differential attentional guidanceby unattended faces expressing positive and negative emotion. Perception andPsychophysics, 63(6), 1004–1013.

arran, E. K., & Jarrold, C. (2003). Visuospatial cognition in Williams syndrome:Reviewing and accounting for the strengths and weaknesses in performance.Developmental Neuropsychology, 23(1–2), 173–200.

ox, E., Lester, V., Russo, R., Bowles, R. J., Pichler, A., & Dutton, K. (2000). Facialexpressions of emotion: Are angry faces detected more efficiently? Cognition

and Emotion, 14(1), 61–92.

rigerio, E., Burt, D. M., Gagliardi, C., Cioffi, G., Martelli, S., Perret, D. I., et al. (2006).Is everybody always my friend? Perception of approachability in Williams syn-drome. Neuropsychologia, 44(2), 254–259.

rischen, A., Eastwood, J. D., & Smilek, D. (2008). Visual search for faces with emo-tional expressions. Psychological Bulletin, 134(5), 662–676.

gia 48 (2010) 1071–1078 1077

Gagliardi, C., Frigerio, E., Burt, D. M., Cazzaniga, I., Perrett, D. I., & Borgatii, R. (2003).Facial expression recognition in Williams Syndrome. Neuropsychologia, 41(6),733–738.

Gerritsen, C., Frischen, A., Blake, A., Smilek, D., & Eastwood, J. D. (2008). Visual searchis not blind to emotion. Perception & Psychophysics, 70(6), 1047–1059.

Grèzes, J., Wicker, B., Berthoz, S., & de Gelder, B. (2009). A failure to grasp the affec-tive meaning of actions in autism spectrum disorder subjects. Neuropsychologia,47(8–9), 1816–1825.

Grice, S. J., Spratling, M. W., Karmiloff-Smith, A., Halit, H., Csibra, G., de Haan, M., etal. (2001). Disordered visual processing and oscillatory brain activity in autismand Williams syndrome. Neuroreport, 12, 2697–2700.

Haas, B. W., Mills, D., Yam, A., Hoeft, F., Bellugi, U., & Reiss, A. (2009). Geneticinfluences on sociability: Heightened amygdala reactivity and event-relatedresponses to positive social stimuli in Williams syndrome. The Journal of Neuro-science, 29(4), 1132–11349.

Hansen, C. H., & Hansen, R. D. (1988). Finding the face in the crowd: An angersuperiority effect. Journal of Personality and Social Psychology, 54(6), 917–924.

Hariri, A. R., Tessitore, A., Mattay, V. S., Fera, F., & Weinberger, D. R. (2002). Theamygdala response to emotional stimuli: A comparison of faces and scenes.Neuroimage, 17(1), 317–323.

Horstmann, G., & Bauland, A. (2006). Search asymmetries with real faces: Testingthe anger-superiority effect. Emotion, 6(2), 193–207.

Jarrold, C., Baddeley, A. D., & Hewes, A. K. (1998). Verbal and nonverbal abilitiesin the Williams syndrome phenotype: Evidence for diverging developmentaltrajectories. Journal of Child Psychology and Psychiatry, 39(4), 511–523.

Johnson, M. H., Dziurawiec, S., Ellis, H., & Morton, J. (1991). Newborns’ preferentialtracking of face-like stimuli and its subsequent decline. Cognition, 40(1–2), 1–19.

Jones, W., Bellugi, U., Lai, Z., Chiles, M., Reilly, J., Lincoln, A., et al. (2000). Hyperso-ciability in Williams syndrome. Journal of Cognitive Neuroscience, 12(Suppl. 1),30–46.

Karmiloff-Smith, A., Thomas, M., Annaz, D., Humphreys, K., Ewing, S., Brace, N., et al.(2004). Exploring the Williams syndrome face-processing debate: The impor-tance of building developmental trajectories. Journal of Child Psychology andPsychiatry, 45(7), 1258–1274.

Kippenhan, J. S., Olsen, R. K., Mervis, C. B., Morris, C. A., Kohn, P., Meyer-Lindenberg, A.,et al. (2005). Genetic contributions to human gyrification: Sulcal morphometryin Williams syndrome. Journal of Neuroscience, 25(34), 7840–7846.

Kristjánsson, A., Vuilleumier, P., Schwartz, S., Macaluso, E., & Driver, J. (2007). Neuralbasis for priming of pop-out during visual search revealed with fMRI. CerebralCortex, 17(7), 1612–1624.

Krysko, K. M., & Rutherford, M. D. (2009). A threat-detection advantage in those withautism spectrum disorders. Brain and Cognition, 69(3), 472–480.

LoBue, V. (2009). More than just a face in the crowd: Detection of emotional facialexpressions in young children and adults. Developmental Science, 12(2), 305–313.

Lundqvist, D., Esteves, F., & Öhman, A. (1999). The face of wrath: Critical features forconveying facial threat. Cognition and Emotion, 13, 691–711.

Lundqvist, D., Esteves, F., & Öhman, A. (2004). The face of wrath: The role of fea-tures and configurations in conveying social threat. Cognition and Emotion, 18(2),161–182.

Magnussen, S., Sunde, S., & Dyrnes, S. (1994). Patterns of perceptual asymmetry inprocessing facial expression. Cortex, 30, 215–229.

Mather, M., & Knight, M. R. (2006). Angry faces get noticed quickly: Threat detec-tion is not impaired among older adults. The Journals of Gerontology Series B:Psychological Sciences and Social Sciences, 61(1), 54–57.

Mervis, C., & Bertrand, J. (1997). Developmental relations between cognition andlanguage: Evidence from Williams Syndrome. In L. B. Adamson, & M. A. Rom-ski (Eds.), Research on communication and language disorders: Contributions totheories of language development (pp. 75–106). New York: Brookes.

Mervis, C. B., Robinson, B., Bertrand, J., Morris, C. A., Klein-Tasman, B., & Armstrong,S. (2000). The Williams syndrome cognitive profile. Brain and Cognition, 44(3),604–628.

Meyer-Lindenberg, A., Hariri, A. R., Munoz, K. E., Morris, C. A., Mervis, C., Mattay,V. S., et al. (2005). Neural correlates of genetically abnormal social cognition inWilliams syndrome. Nature Neuroscience, 8(8), 991–993.

Meyer-Lindenberg, A., Kohn, P., Mervis, C. B., Kippenhan, J. S., Olsen, R. K., Morris, C.A., et al. (2004). Neural basis of genetically determined visuospatial constructiondeficit in Williams syndrome. Neuron, 43(5), 623–631.

Mills, D. L., Alvarez, T. D., St George, M., Appelbaum, L. G., Bellugi, U., & Neville, H.(2000). III. Electrophysiological studies of face processing in Williams syndrome.Journal of Cognitive Neuroscience, 12(Suppl. 1), 47–64.

Mobbs, D., Eckert, M. A., Menon, V., Mills, D., Korenberg, J., Galaburda, A. M., et al.(2007). Reduced parietal and visual cortical activation during global process-ing in Williams Syndrome. Developmental Medicine & Child Neurology, 49(6),433–438.

Morris, J. S., Frith, C. D., Perrett, D. I., Rowland, D., Young, A. W., Calder, A. J., et al.(1996). A differential neural response in the human amygdala to fearful andhappy facial expressions. Nature, 383(6603), 812–815.

Öhman, A. (1993). Fear and anxiety as emotional phenomena: Clinical phenomenol-ogy, evolutionary perspectives, and information processing mechanisms. In M.Lewis, & J. M. Haviland (Eds.), Handbook of emotions (pp. 511–536). New York,

NY: The Guilford Press.

Öhman, A., Lundqvist, D., & Esteves, F. (2001). The face in the crowd revisited: Athreat advantage with schematic stimuli. Journal of Personality and Social Psy-chology, 80(3), 381–396.

Öhman, A., & Mineka, S. (2001). Fears, phobias, and preparedness: Toward an evolvedmodule of fear and fear learning. Psychological Review, 108(3), 483–522.

1 cholo

Ö

O

P

P

P

P

R

R

R

R

R

S

078 A. Santos et al. / Neuropsy

hman, A., & Mineka, S. (2003). The malicious serpent: Snakes as a prototypicalstimulus for an evolved module of fear. Current Directions in Psychological Science,12(1), 5–9.

lsen, R. K., Kippenhan, J. S., Japee, S., Kohn, P., Mervis, C. B., Saad, Z. S., et al.(2009). Retinotopically defined primary visual cortex in Williams syndrome.Brain, 123(Pt3), 634–644.

ani, J. R., Mervis, C. B., & Robinson, B. F. (1999). Global spatial organiza-tion by individuals with Williams syndrome. Psychological Science, 10(5),453–458.

lesa-Skewerer, D., Faja, S., Schofield, C., Verbalis, A., & Tager-Flusberg, H. (2006).Perceiving facial and vocal expressions of emotion in individuals with Williamssyndrome. American Journal of Mental Retardation, 111(1), 15–26.

reus, M. (1984). The Williams syndrome: Objective definition and diagnosis. ClinicalGenetics, 25(5), 422–428.

urcell, D. G., Stewart, A. L., & Skov, R. B. (1996). It takes a confounded face to popout of a crowd. Perception, 25(9), 1091–1108.

iby, D. M., & Hancock, P. J. (2008). Viewing it differently: Social scene perceptionin Williams syndrome and autism. Neuropsychologia, 46(11), 2855–2860.

iby, D. M., & Hancock, P. J. (2009). Do faces capture the attention of individuals withWilliams syndrome or autism? Evidence from tracking eye movements. Journalof Autism and Developmental Disorders, 39(3), 421–431.

ondan, C., Mancini, J., Livet, M. O., & Deruelle, C. (2003). Perceptual and visuo con-structive performance in children with Williams syndrome. Cognition, Brain,Behavior, 7, 149–157.

ondan, C., Santos, A., Mancini, J., Livet, M. O., & Deruelle, C. (2008). Global and localprocessing in Williams syndrome: Drawing versus perceiving. Child Neuropsy-chology, 14(3), 237–248.

uffman, T., Ng, M., & Jenkin, T. (2009). Older adults respond quickly to angry facesdespite labeling difficulty. The Journals of Gerontology Series B: Psychological Sci-ences and Social Sciences, 64(2), 171–179.

antos, A., Milne, D., Rosset, D., & Deruelle, C. (2007). Challenging symmetry on men-tal retardation: Evidence from Williams syndrome. In E. B. Heinz (Ed.), Mentalretardation research advances (pp. 147–174). NY: Nova Science Publishers.

gia 48 (2010) 1071–1078

Santos, A., Rosset, D., & Deruelle, C. (2009). Human versus non-human face process-ing: Evidence from Williams Syndrome. Journal of Autism and DevelopmentalDisorders, doi:10.1007/s10803-009-0789-1

Sarpal, D., Buchsbaum, B. R., Kohn, P. D., Kippenhan, J. S., Mervis, C. B., Morris, C. A.,et al. (2008). A genetic model for understanding higher order visual process-ing: Functional interactions of the ventral visual stream in Williams syndrome.Cerebral Cortex, 18(10), 2402–2409.

Scerif, G., Cornish, K., Wilding, J., Driver, J., & Karmiloff-Smith, A. (2004). Visualsearch in typically developing toddlers and toddlers with Fragile X or Williamssyndrome. Developmental Science, 7(1), 116–130.

Schultz, R. T. (2005). Developmental deficits in social perception in autism: The roleof the amygdala and fusiform face are. International Journal of DevelopmentalNeuroscience, 23(2–3), 125–141.

Tager-Flusberg, H., Plesa Skwerer, D., Faja, S., & Joseph, R. M. (2003). People withWilliams syndrome process faces holistically. Cognition, 89(1), 11–24.

Tager-Flusberg, H., Plesa Skewerer, D., & Joseph, R. M. (2006). Model syndromesfor investigating social cognitive and affective neuroscience: A comparison ofautism and Williams syndrome. Social Cognitive and Affective Neuroscience, 1(3),175–182.

Udwin, O., & Yule, W. (1991). A cognitive and behavioural phenotype in Williamssyndrome. Journal of Clinical and Experimental Neuropsychology, 13(2), 232–244.

Udwin, O., Yule, W., & Martin, N. (1987). Cognitive abilities and behavioural char-acteristics of children with idiopathic infantile hypercalcaemia. Journal of ChildPsychology and Psychiatry and Allied Disciplines, 28(2), 297–309.

Vuilleumier, P., & Schwartz, S. (2001). Emotional facial expressions capture atten-tion. Neurology, 56(2), 153–158.

Wechsler, D. (1996). Manual for intelligence scale for children (3rd ed.). New York:

The Psychological Corporation.

Wechsler, D. (1997). Wechsler adult intelligence scale (3rd ed.). New York: The Psy-chological Corporation.

Whalen, P. J., Shin, L. M., McInerney, S. C., Fischer, H., Wright, C. I., & Rauch, S. L. (2001).A functional MRI study of human amygdala responses to facial expressions offear vs. anger. Emotion, 1(1), 70–83.