Elsevier

Neuropsychologia

Volume 48, Issue 7, June 2010, Pages 2110-2119
Neuropsychologia

Affiliative behavior in Williams syndrome: Social perception and real-life social behavior

https://doi.org/10.1016/j.neuropsychologia.2010.03.032Get rights and content

Abstract

A frequently noted but largely anecdotal behavioral observation in Williams syndrome (WS) is an increased tendency to approach strangers, yet the basis for this behavior remains unknown. We examined the relationship between affect identification ability and affiliative behavior in participants with WS relative to a neurotypical comparison group. We quantified social behavior from self-judgments of approachability for faces, and from parent/other evaluations of real life. Relative to typical individuals, participants with WS were perceived as more sociable by others, exhibited perceptual deficits in affect identification, and judged faces of strangers as more approachable. In WS, high self-rated willingness to approach strangers was correlated with poor affect identification ability, suggesting that these two findings may be causally related. We suggest that the real-life hypersociability in WS may arise at least in part from abnormal perceptual processing of other people's faces, rather than from an overall bias at the level of behavior. While this did not achieve statistical significance, it provides preliminary evidence to suggest that impaired social-perceptual ability may play a role in increased approachability in WS.

Introduction

Williams syndrome (WS) is a rare neurodevelopmental disorder resulting from a hemizygous deletion of 25–30 genes on chromosome 7q11.23 (Ewart et al., 1993, Korenberg et al., 2000). In addition to several physical abnormalities (e.g., Beuren et al., 1964, Morris and Mervis, 2000, Williams et al., 1961), cognitively, the majority of individuals with WS exhibit mild to moderate intellectual impairment, with an estimated mean Full-Scale IQ (FIQ) of 50–60 (Martens et al., 2008, Mervis et al., 2000). The FIQ masks an asymmetrical profile, in which Performance IQ (PIQ) is typically lower than Verbal IQ (VIQ) (Howlin et al., 1998, Udwin and Yule, 1990). Further, an unusual profile of cognitive dissociations has emerged for this population: while the mature neurocognitive phenotype is associated with relative strengths in processing specifically social stimuli, such as face, significant impairments are evident in general intellectual functioning, for example, in planning and problem solving, as well as in spatial and numerical cognition (e.g., Bellugi et al., 2000, Martens et al., 2008). Neurological studies have further shown that near-typical performance in some tasks, such as face processing, is sustained by abnormal underlying processes (e.g., Haas et al., 2009, Mobbs et al., 2004).

A highly salient behavioral feature of WS is increased sociability (e.g., Gosch and Pankau, 1994, Gosch and Pankau, 1997, Udwin and Yule, 1991, von Arnim and Engel, 1964; see Bellugi et al., 2007, Järvinen-Pasley et al., 2008, Jones et al., 2000, Martens et al., 2008, Mervis and Klein-Tasman, 2000, Meyer-Lindenberg et al., 2006, for reviews). However, the social profile of WS is poorly understood and appears to be beset by intriguing paradoxes. For example, although individuals with WS are highly social and socially fearless, they nevertheless suffer from significant anxiety (Dykens, 2003, Leyfer et al., 2006), exhibit substantial difficulties in social adjustment, and a tendency to suffer from social isolation (Udwin & Yule, 1991). Although systematic studies are relatively sparse, the increased appetitive social drive in individuals with WS has been established using a number of different paradigms, including questionnaires, behavioral observations, and experiments. For example, using two standardized temperament and personality inventories, Klein-Tasman and Mervis (2003) found that high social ratings and empathy distinguished individuals with WS from controls with other developmental disabilities. Another parent report form, The Salk Institute Sociability Questionnaire (SISQ) (Doyle et al., 2004, Jones et al., 2000, Zitzer-Comfort et al., 2007) was specifically developed by Bellugi and colleagues to elucidate the features of the social drive in WS. It gathers information regarding the individual's willingness to approach both familiar and unfamiliar people, behavior in social settings, ability to remember faces and names, eagerness to please others, empathy, and the frequency with which others approach the individual. In one study, age-related changes in social behavior in children with WS, Down syndrome (DS), and typically developing (TD) controls aged one to 13 years were investigated (Doyle et al., 2004). Consistent with earlier findings from adult participants (Jones et al., 2000), whole group analyses showed that the WS group was rated significantly higher on all aspects of sociability studied than comparison groups of individuals with various neurodevelopmental disorders and typical development. Age-related analyses showed that increased sociability was evident even among the youngest children with WS, and significantly, children with WS exceeded TD controls with respect to Global sociability and Approach strangers in every age group; similar findings were also found in relation to children with DS. These findings of significantly increased sociability in WS relative to TD have also been replicated cross-culturally (Zitzer-Comfort et al., 2007).

Observational studies have shown that beginning in infancy, individuals with WS show a strong preference for social over non-social stimuli (Jones et al., 2000, Mervis et al., 2003, Riby and Hancock, 2008, Riby and Hancock, 2009). Experimental studies have investigated the ability of individuals with WS to make judgments regarding approach from looking at photos of unfamiliar faces (Bellugi et al., 1999, Frigerio et al., 2006). In the first study of this kind, Bellugi et al. (1999) presented participants with black-and-white photographs of unfamiliar adults, which have previously been rated in terms of approachability (both positive and negative) by typical individuals. The results showed that, while both chronological age (CA)- and mental age (MA)-matched controls performed similarly, participants with WS exhibited a positive bias by rating both positively and negatively pre-judged photographs as significantly more approachable than controls while retaining approximate rank-order. Frigerio et al. (2006) extended these findings by utilizing affective face stimuli taken from Ekman and Friesen (1976) expressing anger, disgust, fear, sadness, and happiness, in addition to neutral expressions. These stimuli had also been pre-rated for approachability. The results showed that participants with WS rated all but the most positively pre-judged happy photographs significantly more negatively than typical controls, suggesting that the social stimuli must be positive in valence in order for individuals with WS to perceive them as approachable.

Porter, Coltheart, and Langdon (2007) tested both social-perceptual abilities and approachability ratings with the same stimuli (the Diagnostic Analysis of Nonverbal Accuracy (DANVA; Nowicki & Duke, 1994)). Additionally, an extensive neurocognitive test battery was administered, to address three possible hypotheses with regard to approach behavior in WS: (1) that it reflects amygdala dysfunction; (2) that social stimuli have increased salience for individuals with WS; or (3) that it reflects frontal lobe dysfunction. The participants included individuals with WS and DS, as well as CA- and MA-matched TD comparison groups. The results from the affect identification task were consistent with the literature (Gagliardi et al., 2003, Plesa-Skwerer et al., 2006, Plesa-Skwerer et al., 2005), by showing that while the CA-matched controls significantly outperformed all other groups, those with WS performed at a similar level to the MA-matched controls across the four emotions (happy, sad, angry, scared). The WS group outperformed those with DS. The participants’ approachability ratings were analyzed twice in order to examine the effect of affect identification ability to the perception of approachability. The analysis of the data comprising only correctly identified stimulus items revealed significant between-group differences only for happy expression. The WS, DS, and CA-matched control groups performed similarly, while the MA-matched TD controls gave significantly lower ratings. When approachability ratings to all stimuli were analyzed, unlike in the studies by Bellugi et al. (1999) and Frigerio et al. (2006), CA-matched controls in this study rated the happy stimuli as significantly more approachable than did the WS and DS groups, which performed similarly. The findings were interpreted as supporting the frontal lobe hypothesis as, despite showing similar performance to the CA-matched controls in the approachability task and exhibiting non-specific affect identification deficits, the increased approachability of individuals with WS in real life (e.g., Doyle et al., 2004, Jones et al., 2000) is likely to reflect poor response inhibition.

Most recently, Martens and colleagues conducted a study linking the approachability judgments of individuals with WS and TD controls to their amygdala volumes (Martens, Wilson, Dudgeon, & Reutens, 2009). The behavioral task was the modified Adolphs Approachability Task (Bellugi et al., 1999). The behavioral results replicated those of Bellugi et al. (1999) by showing that participants with WS rated both positive and negative stimuli as significantly more approachable than the controls. Qualitatively, reports also suggested that when judging approachability, individuals with WS relied significantly less on mouth and eye regions than the controls. Interestingly, when the approachability ratings were correlated with the participants’ amygdala volumes, a positive association emerged between the right amygdala volume and approachability judgments particularly for negative stimuli, for individuals with WS only.

As discussed above, many experimental measures of sociability derived from the participants with WS themselves show some unreliability or inconsistency in the literature. Given that those measures are typically quite indirect (asking about the hypothetical approachability of a face picture), the aim of the current study will be extend the previous studies (Frigerio et al., 2006, Martens et al., 2009, Porter et al., 2007) by examining the extent to which self-ratings of approachability of individuals with WS converge with their behavioral tendencies in real life, as perceived by their parents. A further rationale is that some studies have showed that individuals with WS do not perceive all of the unfamiliar faces as more approachable than controls (e.g., Frigerio et al., 2006), while ample evidence suggests significantly increased approachability in WS (e.g., Järvinen-Pasley et al., 2008, Jones et al., 2000, Martens et al., 2008). This raises the question of the extent to which the approachability ratings of individuals with WS may generalize to real-life settings. Previous evidence has also suggested that the ability to identify facial expressions may be linked to approachability ratings in individuals with WS (Porter et al., 2007). Taken together, the conflicting evidence warrants further investigation of the specific relationship between the self-rated approachability and affect identification ability, as well as the ecological validity of self-ratings of approachability, in individuals with WS.

This question is of both clinical and theoretical interest as parents of individuals with WS commonly report worrying about their children placing themselves at risk for harm by approaching unfamiliar people (Doyle et al., 2004). This design will thus allow for the examination of the relationship between self-rated approachability of individuals with WS and their real-life social tendencies, albeit indirectly, as well as the role of facial expression processing in these behaviors. Participants with WS and a TD comparison group were administered the SISQ (Jones et al., 2000), the Adolphs Approachability Task (Bellugi et al., 1999), and a task assessing the identification of facial affect. No MA-equivalent comparison group to participants with WS was included here as numerous previous studies have robustly established that, while individuals with WS are consistently rated significantly higher than MA-matched controls on all aspects of sociability as measured by the SISQ (Doyle et al., 2004, Jones et al., 2000), individuals with WS perform at a similar level to MA-matched controls in tasks of affect identification (Gagliardi et al., 2003, Plesa-Skwerer et al., 2005, Plesa-Skwerer et al., 2006, Porter et al., 2007). Finally, evidence from two previous studies indicates that MA- and CA-matched controls perform similarly in various versions of the Adolphs Approachability Task (Bellugi et al., 1999, Frigerio et al., 2006), while those with WS show significantly different pattern of performance. In the study of Porter et al. (2007), between-group differences in approachability ratings only emerged for the happy expression, and MA-matched controls (mean age = 4.9 years) rated happy stimuli as significantly less approachable than the other groups. However, this result indicates that the performance profile of individuals with WS is distinctly different from that of MA-matched TD controls, and in fact is more similar to that observed in CA-matched TD participants. Moreover, it is generally agreed that the social disposition of individuals with WS is at least in part genetically based (e.g., Deutsch et al., 2007, Doyle et al., 2004, Meyer-Lindenberg et al., 2005), and thus is not simply a result of general developmental delay (e.g., Doyle et al., 2004, Gosch and Pankau, 1994, Jones et al., 2000, Klein-Tasman and Mervis, 2003).

Based upon the literature reviewed above, it was hypothesized that while participants with WS will exhibit increased sociability as measured both with the SISQ and the Adolphs Approachability Task, they will show impaired facial affect identification, as compared to TD. As the findings of Frigerio et al. (2006) and Porter et al. (2007) suggest that individuals with WS can discriminate people on the basis of approachability, it was hypothesized that individuals with higher facial affect identification abilities will be more discriminative of unfamiliar people, and will thus be less inclined to approach them, as reflected by their self-ratings. As anecdotal and parental evidence nevertheless suggests that individuals with WS show increased approachability towards unfamiliar people (e.g., Jones et al., 2000), it was further predicted that self-rated approachability in WS may lack ecological validity, as reflected by parental perceptions of their children's social behavior.

Section snippets

Participants

Twenty individuals with WS (11 males), who had completed the three experimental measures, were identified from a pre-existing database at the Salk Institute. Genetic diagnosis of WS was established using fluorescent in situ hybridization (FISH) probes for elastin (ELN), a gene invariably associated with the WS microdeletion (Ewart et al., 1993, Korenberg et al., 2000). In addition, all participants exhibited the medical and clinical features of the WS phenotype, including cognitive, behavioral,

Results

The means and standard error means for the ratings of the two groups of participants across the SISQ subscales, as well as for the Global Sociability score, in experiment 1 are shown in Fig. 1.

The data were analyzed by a repeated-measures analysis of variance (ANOVA), with Global sociability (Approach strangers/Approach familiars/Social-emotional) entered as the within-participants factor, and group (WS/TD) as the between-participants factor. This analysis revealed a significant effect of

Participants

See experiment 1.

Stimuli and procedures

The modified version of the Adolphs Approachability Task (see Bellugi et al., 1999) included 42 black-and-white photographs of faces of unfamiliar people. Out of the original 100 stimuli (Adolphs, Tranel, & Damasio, 1998), 21 photographs that had been most consistently pre-rated positively, and 21 photographs that had been most consistently pre-rated negatively, and that together spanned as wide a range of ratings as possible, were selected as test stimuli. Participants were

Results

Consistent with the analyses reported in Bellugi et al. (1999), Frigerio et al. (2006), and Martens et al. (2009), the participants’ ratings were analyzed in relation to norms derived from the pre-ratings of the stimuli (Adolphs et al., 1998). Fig. 2 shows that, relative to the TD group, participants with WS tended to use higher approachability ratings for the positively pre-judged faces, but lower approachability ratings for most of the negatively pre-judged face stimuli. A repeated-measures

Participants

See experiment 1.

Stimuli and procedures

The stimuli consist of an image board comprising six black-and-white photographs of a female face expressing extreme prototypical examples of various affective states (happy, sad, angry, afraid, surprised, and neutral; see Fig. 3). The stimuli were taken from the Affective Judgment Questionnaire for Children (Delehanty, 1993, Reilly and Delehanty, 1991). The stimuli were developed using Ekman and Friesen's Facial Action Coding System (FACS; Ekman & Friesen, 1978), and were

Results

The means and standard error means for the affect-labeling scores of the two groups of participants are shown in Fig. 4.

As the data were not normally distributed, two-tailed Mann–Whitney-U tests were carried out on the affect identification data in order to compare performance between the two groups. This analysis showed that relative to typical individuals, participants with WS exhibited poorer ability overall (Z = −5.45, p < 0.001), with significantly lower performance with the surprised (Z = 

General discussion

The present experiments were designed to address the question of how self-measures of approachability are mediated by affect identification ability, and how such ratings correspond to real-life affiliative behavior, in individuals with WS. The main findings from the experiments showed that, consistent with previous data (Doyle et al., 2004, Jones et al., 2000, Zitzer-Comfort et al., 2007), relative to the TD comparison group, individuals with WS were perceived as significantly higher on aspects

Acknowledgement

This study was supported by a grant P01 HD033113-13 from the National Institute of Child Health and Human Development.

References (60)

  • J. Atkinson et al.

    Neurobiological models of visuospatial cognition in children with Williams syndrome: Measures of dorsal-stream and frontal function

    Developmental Neuropsychology

    (2003)
  • U. Bellugi et al.

    Towards the neural basis for hypersociability in a genetic syndrome

    NeuroReport

    (1999)
  • U. Bellugi et al.

    I. The neurocognitive profile of Williams syndrome: A complex pattern of strengths and weaknesses

    Journal of Cognitive Neuroscience

    (2000)
  • U. Bellugi et al.

    Affect, social behavior, and the brain in Williams syndrome

    Current Directions in Psychological Science

    (2007)
  • A.L. Benton et al.

    Contributions to neuropsychological assessment

    (1983)
  • A.J. Beuren et al.

    The syndrome of supravalvular aortic stenosis, peripheral pulmonary stenosis, mental retardation and similar facial appearance

    American Journal of Cardiology

    (1964)
  • Delehanty, S. G. (1993). Thinking about feelings: Emotional development in the preschool years. Unpublished Ph.D....
  • S.I. Deutsch et al.

    Williams syndrome: A genetic deletion disorder presenting clues to the biology of sociability and clinical challenges of hypersociability

    CNS Spectrums

    (2007)
  • T.F. Doyle et al.

    “Everybody in the world is my friend”. Hypersociability in young children with Williams syndrome

    American Journal of Medical Genetics

    (2004)
  • E.M. Dykens

    Anxiety, fears, and phobias in persons with Williams syndrome

    Developmental Neuropsychology

    (2003)
  • P. Ekman et al.

    Pictures of facial affect

    (1976)
  • P. Ekman et al.

    Facial action coding system (FACS): A technique for the measurement of facial action

    (1978)
  • A.K. Ewart et al.

    Hemizygosity at the elastin locus in a developmental disorder, Williams syndrome

    Nature Genetics

    (1993)
  • A.M. Galaburda et al.

    V. Multi-level analysis of cortical neuroanatomy in Williams syndrome

    Journal of Cognitive Neuroscience

    (2000)
  • A.M. Galaburda et al.

    Williams syndrome: Neuronal size and neuronal-packing density in primary visual cortex

    Archives of Neurology

    (2002)
  • A. Gosch et al.

    Social-emotional and behavioral adjustment in children with Williams-Beuren syndrome

    American Journal of Medical Genetics

    (1994)
  • A. Gosch et al.

    Personality characteristics and behavioral problems in individuals of different ages with Williams syndrome

    Developmental Medicine and Child Neurology

    (1997)
  • B.W. Haas et al.

    Genetic influences on sociability: Heightened amygdala reactivity and event-related responses to positive social stimuli in Williams syndrome

    Journal of Neuroscience

    (2009)
  • P. Howlin et al.

    Cognitive functioning in adults with Williams syndrome

    Journal of Child Psychology and Psychiatry

    (1998)
  • A. Järvinen-Pasley et al.

    Defining the social phenotype in Williams syndrome: A model for linking gene, brain, and cognition

    Development and Psychopathology

    (2008)
  • Cited by (0)

    View full text