Looking to the Left or Looking to the Right? Revisiting Visual Scanning of Faces in Young Children with ASD

Background: Behavioral and neuroimaging studies have demonstrated a right hemisphere advantage in face processing, manifested by a left gaze bias (LGB) when participants look at faces (i.e. visual input come from left visual field). This tendency to look first and for longer periods at the left hemiface (from the viewer’s perspective) appears to develop between 6 and 12 months of age. The LGB has been observed not only for human faces, but also for dog faces in a group of 4-year-old typically developing children. To date, much of the studies investigating visual scanning of faces in ASD have focused on looking-time towards the mouth and eyes regions. However, two studies have recently found that adults with ASD as well as children at risk for ASD spend less time looking at the left side of a face than control participants. Whether this lack of LGB is also present for the direction of the first fixation, a measure reflecting the right hemisphere bias, and overall, whether a lack of LGB is specific to human faces in young children with ASD remains unknown.  

Objectives: To determine if young children with ASD demonstrate the typical LGB for the first fixation while looking at human faces and to determine the specificity of a lack of LGB.  

Methods: Eye-tracking data were collected while young children with ASD (24- to 60-months old) and typically developing children (24- to 60-months old) viewed picture of human faces and of dog faces. For each participant and each picture category presented, the direction of the first fixation and the total fixation time on each side of the face were analyzed, and Laterality Index (LI) was computed. LI (first look) and LI (total fixation time) were compared between diagnostic groups for each picture category separately.

Results: Preliminary results suggest that the ASD group differ significantly from the TD group. As expected, typically developing children demonstrate the typical LGB for both human and dog faces. To the contrary, ASD children do not show such an effect for any of the face category presented. At an individual level, some of the young children with ASD even present an opposite bias, and notably exhibit a higher probability of first gaze towards the right side of the face.  

Conclusions: These results suggest that a lack of LGB for the direction of the first fixation may be considered as an eye-tracking marker for ASD. Overall, these results are discussed in terms of their potential use in a screening program for ASD in a general population setting.