Insights into the development of face recognition mechanisms revealed by face aftereffects

Abstract

An important question in person perception is how we acquire the perceptual/cognitive mechanisms that characterize adult expertise. Children's performance on face recognition tests improves dramatically between age 4 and adolescence suggesting that our face recognition system may change during childhood. Yet, the source of this improvement is controversial. In this review, we consider whether changes in the way identity is represented/coded in face space could contribute to this age-related improvement. Face aftereffects have been extensively applied to studying face coding in adults and more recently they have been applied to studying the mechanisms of face coding in children. Face aftereffects are temporary distortions of perception induced by exposure to faces and are thought to reflect the mechanisms underlying face perception. Face aftereffect techniques have revealed that children as young as 4 years of age show evidence of adult-like face space organization, with opponent coding of face dimensions. These findings are consistent with an emerging picture that the key mechanisms of face perception are present early in childhood.

The left perceptual bias for adult and infant faces in adults and 5-year-old children: Face age matters

“…Some studies reported the bias to be present to at least a weak extent in 5-year-old children (Aljuhanay et al, 2010;Levine & Levy, 1986), with its strength and magnitude continuing to increase through middle childhood (Anes & Short, 2009;Levine & Levy, 1986;Watling & Bourne, 2007Workman et al, 2006). These findings would fit well with the picture emerging from current research on the development of face processing abilities, showing that, although neural and perceptual mechanisms subserving face recognition develop dramatically in the first decade, some key mechanisms of face encoding are already in place early in childhood (e.g., Jeffery & Rhodes, 2011). Given the general consensus and ample evidence that experience contributes greatly to emerging neurocognitive specialization for faces (Cohen Kadosh & Johnson, 2007), it should be expected that the LPB is more evident for those face categories for which children acquire more experience and develop greater perceptual expertise.…”

Not just the norm: Exemplar-based models also predict face aftereffects

“…These findings have been widely cited as providing perhaps the most compelling behavioral evidence in favor of norm-based models over exemplar-based models (e.g., Jeffery et al, 2010; Jeffery & Rhodes, 2011; Rhodes & Leopold, 2011; Rhodes et al, 2005; Tsao & Freiwald, 2006). Their study design was an extension of the Leopold et al (2001) paradigm, adding an additional control condition to more carefully assess the direction of aftereffects relative to the norm.…”

“…These findings are important in the context of the current face recognition literature. Many recent reviews (e.g., Jeffery & Rhodes, 2011; Leopold & Bondar, 2005; Rhodes & Leopold, 2011; Rhodes et al, 2005; Tsao & Freiwald, 2006; Tsao & Livingstone, 2008) cite the difference in adaptation for opposite versus non-opposite adaptors as compelling evidence in favor of a norm-based account and against an exemplar-based account of face space representation. Findings that the magnitude of aftereffects increase as a function of adaptor distance from the average has been taken as evidence for norms in face identification by adults (e.g., Leopold & Bondar, 2005), face identification by children (e.g., Jeffery et al, 2010), and emotion perception by adults (e.g., Skinner & Benton, 2010).…”

Face Adaptation Effects: Reviewing the Impact of Adapting Information, Time, and Transfer

“…This review focuses on empirical studies that investigate face adaptation effects on a behavioral level; this is typically realized by an overt categorization of face stimuli in a test phase. In fact, we focus on adaptation under optimal conditions in a fully developed and (more or less) optimally functioning cognitive system; i.e., we review findings of studies typically investigating adaptation effects in younger adults possessing face recognition skills that are particularly impressive, for instance the fact that normal persons can discriminate thousands of faces (Jeffery and Rhodes, 2011) when they reach so-called “face expertise” (Schwaninger et al, 2003). This focus on complex objects of the face category is realized in an exclusive and extensive way; that is, we do not relate findings in the area of face adaptation effects to other visual coding mechanisms such as color coding as realized in previous work (Webster, 2011; Webster and MacLeod, 2011).…”

“…In contrast to what we provide, we do not include results about the adaptation of neural processes to face stimuli: for instance, studies on modulations of the N170 as a result of prior adaptation (e.g., Kovács et al, 2006; Kloth et al, 2010) as questions regarding this area of research refer to further dimensions and use different theoretical frameworks, mostly based on specific brain processes and structures. In addition, we omit research from developmental and evolutionary perspectives on face adaptation effects as they were already the major aim of recent alternative review papers (Leopold and Rhodes, 2010; Jeffery and Rhodes, 2011). …”