Do Individuals with Autism Spectrum Disorder Process Own- and Other-Race Faces Differently?

Background:  It has been well documented that individuals with autism spectrum disorders (ASD) exhibit atypical eye scanning patterns when processing human faces. For example, they display reduced attention to faces and their core features (eyes, nose, mouth), especially the eye region (e.g., Klin et al., 2002; Pelphrey et al., 2002). Most of the previous studies on atypical face processing in ASD, however, have used only faces of individuals from the same racial group as the participants.

A growing literature has demonstrated a robust “other-race effect” (ORE) in typically developing (TD) individuals. That is, TD observers recognize and discriminate own-race faces more accurately than faces from other racial groups. This effect reflects the role of visual experience in shaping face recognition expertise. We investigated whether individuals with ASD scan own- and other-race faces differentially using eye tracking. Considering that there is inconsistency in the evidence regarding the existence of a behavioral ORE in ASD (Chien et al., 2014; Wilson, 2011), we investigated face scanning patterns using eye tracking, to provide a different measure of processing race information from faces.

It has been shown that both adults and infants display differential scanning of own- and other-race faces. For example, Asian observers tend to focus on the central region (i.e., the nose) of Chinese faces and the eye region of Caucasian faces (Fu et al., 2012; Liu et al., 2011). 

Objectives:  The present study examined whether individuals with ASD would, like typical individuals, show differential patterns of visual scanning when viewing own- and other-race faces. 

Methods:   The study included 15- to 25-year-old Chinese adolescents and young adults with ASD, age-matched typically developing (TD) individuals, and individuals with intellectual disability (ID). Participants completed a face recognition task with both own- (Chinese) and other-race (Caucasian) faces, while their eye movements were tracked. We analyzed fixation durations within each area of interest (AOI) and compared them between groups and races.

Results:  Results, as shown in Figure 1, indicated that (a) in terms of recognition, the ASD and ID groups, although not the TD group (due to a ceiling effect), displayed superior recognition of own-race faces relative to other-race faces; (b) different from TD individuals and individuals with ID, individuals with ASD showed atypical face processing patterns regardless of face race; (c) similar to TD and ID individuals, individuals with ASD scanned own- and other-race faces differentially: they fixated on the eyes of other-race faces longer than those of own-race faces, whereas they looked at the nose and mouth of own-race faces longer than those of other-race faces. 

Conclusions:  The results suggest that there are similar cross-race differences in the face scanning patterns of ASD and non-ASD individuals. The findings corroborate the behavioral evidence of an other-race effect in children with ASD (Wilson et al., 2011) and suggest that similar to TD individuals, individuals with ASD are sensitive to face race information: their visual scanning and recognition of faces are both influenced by asymmetrical experience with different types of faces.