The development of face processing in infancy and early childhood: current perspectives

Infant and Child Development Inf. Child Dev. 10: 1 – 2 (2001) DOI: 10.1002/icd.238 The Development of Face Processing in Infancy and Early Childhood: Current Perspectives Olivier Pascalisa,* and Alan Slaterb a The University of Sheffield, Department of Psychology, Sheffield, UK School of Psychology, Washington Singer Laboratories, University of Exeter, Exeter, UK b The development of face perception during infancy and childhood is one of the most extensively researched topics in early development, and has resulted in hundreds of articles and many theoretical positions. There is general agreement that there is a domain-specific early face-detection and recognition system that is able to process faces, and will become adult-like with experience. The properties of this early memory and representation system are, however, still controversial, and the role of experience in the development of our ‘face perception’ abilities is little understood, and also controversial. This Special Issue presents many of these theoretical views, and the empirical papers present research that illustrate several of the current topics of interest. The first paper is a target paper by Nelson, in which he suggests that face recognition is ‘special’, and that it is subserved by discrete neural systems. He also proposes that the face processing system develops during the first years of life from a broadly-tuned non-specific complex figure recognition system to a human face-tuned discrimination system. This target paper is discussed, in the following commentaries, by several specialists in the area of face processing. Over time, infants increasingly have greater exposure to conspecifics, and the specificity of the face recognition system to the faces of one’s own species also increases. Pascalis et al. explore this species-specific bias by investigating children’s ability to process another species’ faces, in this paper, monkey’s faces. In his paper, Nelson reminds us that there is a right hemisphere bias towards processing faces in infants, a bias that is similar to that observed in the adult. In the next paper in this issue, Catherwood et al. explore this hemispheric asymmetry for colour and facial pattern in 6-month-old infants. The next two papers concern newborns’ face perception abilities. Simion et al. ask what parameters or characteristics of the face guide newborns’ attraction to faces: they argue that newborns do not respond to face-like stimuli by ‘facedness’ per se, but rather, by some more general pattern-detecting structural characteristics that best satisfy the constraints of the immature visual system. We have known for some time that newborn infants quickly learn to recognize their mothers’ faces, and in his paper, Bushnell develops this finding by asking how much exposure to the face do newborns need in order to recognize their mother, and further, can they still remember their mother after a delay? In fact, * Correspondence to: The University of Sheffield, Department of Psychology, Sheffield, S10 2TP, UK. Copyright © 2001 John Wiley & Sons, Ltd. 2 O. Pascalis and A. Slater very little exposure is needed to recognize the mother, and newborns still show a preference for their mother’s face having not seen it for 15 min. This research demonstrates both the speed of memory formation in the newborn period, and the robustness of the memories formed. Brookes et al. consider another important aspect of the face, namely, that it provides information to more than one modality! They present convincing evidence that 3-month-olds are able to associate, and remember, the pairings of strangers’ voices and faces. They compare their findings with apparently conflicting evidence from earlier research, and argue two things: (1) that findings in early face perception (as with many, if not most areas of research!) will depend upon the procedures and stimulus materials used, and (2) that the ability to associate people’s faces with their voices is likely to be present soon after birth. Several researchers have presented evidence that there is a shift in face processing in early childhood, from a reliance on ‘piecemeal’ processing (attending to the individual components of faces) to configural or holistic processing. Cohen and Cashon argue that configural processing is responsible for the striking inversion effect that is found with faces: inverted faces are less rapidly and less easily processed than upright faces. They suggest that even infants are able to process faces configurally, and in their experiment, they link the ‘inversion effect’ with holistic/configural processing. They provide evidence in favour of their hypothesis by showing that 7-month-olds process an upright face as a configuration, but an inverted face as independent features. This special issue illustrates some of the diversity and complexity of research and theorizing into face perception and processing in infancy and early childhood. Our thanks go to Charlie Lewis and Brian Hopkins for offering us the opportunity to edit the issue, and particularly, to the unpaid reviewers who willingly gave their detailed comments on the papers, and always within the deadline we set them! Copyright © 2001 John Wiley & Sons, Ltd. Inf. Child Dev. 10: 1 –2 (2001)