Initial Eye Gaze to Faces and Their Functional Consequence on Face Identification Abilities in Autism Spectrum Disorder

Background: Two areas of active research in autism spectrum disorder (ASD) include visual social attention and face processing. The visual social attention literature demonstrates that individuals with ASD exhibit decreased attention to salient social stimuli. Research on face processing suggests deficits in this domain in ASD, but includes many mixed findings regarding the nature and extent of these differences. The existing literature in these areas points to the need for additional research that carefully characterizes visual social attention to faces, optimizing precision in both the spatial and temporal domains, and assessing its functional role in a specific task or context (e.g., face identification). One way to achieve this is to focus on the most important, or critical moments, for a given context. Highly complex contexts, such as typical social interactions, are comprised of a series of critical moments that contain brief, specific, and often unpredictable cues. A simpler context, such as face identification, is also comprised of critical moments of processing. Specifically, the first moment one looks at a face has been shown to be highly informative and sufficient to identify faces in neurotypical adults.

Objectives: The current study focused on comprehensively characterizing the first moment one looks to a face and its functional consequence on face identification abilities in adolescents with and without ASD.

Methods: 42 adolescents with and without ASD (n = 21 per group) completed an adapted version of an established eye-tracking and face identification paradigm. Specifically, a series of faces were presented briefly, and we observed where participants naturally look first, while simultaneously measuring their face identification abilities. Then, the location at which individuals look when presented a face was experimentally manipulated, and we observed how face identification performance varied as a function of that location. Participants also completed the Dartmouth Face Perception Test (DFPT), a more traditional measure of face identification.

Results: Adolescents with ASD showed strikingly similar patterns of behavior related to initial eye gaze to faces (average first look location: t(40)=.61, p=.54; variability (SD) of landing location of the first look across trials: t(40)=-.32, p=.75) and face identification performance (t(40)=.60, p=.55) for briefly presented faces when naturally viewing faces. Face identification accuracy varied as a function of experimentally manipulated first look location (F(4,124)=26.77, p<.001), but there were no differences between groups (F(1,31)=.10, p=.75). Although there were no group differences related to first look, adolescents with ASD performed significantly worse on the DFPT compared to TD controls (t(40)=2.56, p=.01).

Conclusions: Results suggest that the initial look to faces and face identification for briefly presented faces are intact in ASD, ruling out the possibility that deficits in face perception, at least in adolescents with ASD, begin at the initial look. However, individuals with ASD showed impairments on the more traditional measure of face identification, pointing to the possibility that atypicalities in face processing in ASD appear after the first look.