Atypical unfamiliar face processing in Williams syndrome: What can it tell us about typical familiarity effects?

Abstract

Introduction. Familiar and unfamiliar face perception is typically dissociated by the relative use of internal and external face features. The Williams syndrome (WS) social phenotype emphasises hypersociability, with an interest in interacting with people irrespective of familiarity. The aim is to explore whether unfamiliar face processing is characterised by the typical dissociation between internal and external features in WS, or whether the social stimulus drive towards strangers is linked to atypicalities of unfamiliar face processing.

Method. The procedure replicates that previously used with typically developing children. Participants with WS (aged 10–18 years) and typically developing comparison participants determine whether two face parts are from the same person or different people, using the whole face, internal, and external features.

Results. Only participants with WS, and not typically developing participants, show greater accuracy matching unfamiliar faces from internal than external features.

Conclusions. Evidence of atypical unfamiliar face processing in WS may inform models of typical face perception, revealing the origins of the relative advantage for internal features typically associated with familiar but not unfamiliar faces. The results also have implications for understanding more clearly the social phenotype associated with WS.

Acknowledgements

The research was supported by grant PTA-030-2002-00913 to DR from the Economic and Social Research Council. We would like to thank the Williams Syndrome Foundation for their help and support, along with all the families and children who have participated in the research. We are grateful to Dr Peter Hancock for manuscript preparation advice.

Notes

¹This sample size is comparable to published research with this population, e.g., Deruelle et al. (2003), Karmiloff-Smith et al. (2004). Due to the rarity of WS and the size of the UK many published research studies use samples of this size.

²It would be interesting to incorporate familiar faces; however, it was not possible to obtain the same personally familiar faces for all participants. Previous research has used famous faces but it has been emphasised that these lack ecological validity due to an absence of interpersonal contact. Additionally, a number of parents noted that their child lacked interest in films or television and therefore even obtaining suitable famous faces would be problematic for individuals with WS.

³Reaction time analyses were not included here as they may have been particularly unreliable in the individualised testing sessions required for participants with WS. They would also have been particularly uninformative when accuracy was low. However, future research may explore this issue.