Do individuals with and without autism spectrum disorder scan faces differently? A new multi-method look at an existing controversy.
Autistic adults mostly scan faces like everyone else; only tiny eye-to-nose differences show up, and only with real photos.
01Research in Context
What this study did
Yi et al. (2014) watched how adults with autism, typical adults, and adults with intellectual disability looked at real faces.
They used three eye-tracking tools at once: heat maps, scan paths, and moment-to-moment fixations.
The goal was to see if autism really changes how people explore faces.
What they found
Only two small spots differed: the autism group looked away from the right eye faster and spent extra time on the nose.
Their eye-movement routes looked different from typical adults, but the same as adults with intellectual disability.
The authors say the ‘deficit’ is narrow, not a full-face problem.
How this fits with other research
Williams et al. (2002) first claimed big scanning problems in autism, but that study used looser tools; Li’s stricter methods now shrink the claim.
Shic et al. (2023) and Costa et al. (2017) still find less face time in kids and women with autism, so the tiny eye/nose gap may grow into wider social avoidance outside the lab.
van der Geest et al. (2002) saw zero group difference when children viewed cartoon people, not real photos; together the papers show stimulus and age matter—static photos in adults reveal mild quirks, while cartoons or young kids may hide them.
Why it matters
When you run social-skills programs, don’t assume clients ‘can’t’ read faces.
The subtle eye-to-nose shift Li caught is teachable: prompt longer eye contact or use live faces instead of pictures.
Also, check your eye-tracking assessments—real photos may pick up autism signs that cartoons miss.
Want CEUs on This Topic?
The ABA Clubhouse has 60+ free CEUs — live every Wednesday. Ethics, supervision & clinical topics.
Join Free →Show a real photo, count aloud to three, and praise steady eye contact to stretch those short right-eye fixations.
02At a glance
03Original abstract
Individuals with autism spectrum disorder (ASD) are known to process faces atypically. However, there has been considerable controversy regarding whether ASD individuals also scan faces differently from typical adults. Here we compared ASD individuals' face-scanning patterns with those of typically developing (TD) controls and intellectually disabled (ID) but non-ASD individuals with the use of an eye tracker and multiple approaches to analyze eye-tracking data. First, we analyzed the eye movement data with a traditional approach, measuring fixation duration on each area of interest within the face. We found that compared with TD and ID individuals, ASD individuals looked significantly shorter at the right eye. Second, we used a data-driven method that analyzes fixations on each pixel of the face stimulus and found that individuals with ASD looked more at the central nasal area than TD and ID individuals. Third, we used a novel saccade path analysis that measures frequencies of saccades between major face areas. We found that ASD individuals scanned less often between core facial features than TD individuals but did not differ from ID individuals. Findings from the multi-method approaches show that individuals with ASD appear not to have a pervasive ASD-specific atypicality in visual attention toward the face. The ASD-specific atypical face-scanning patterns were shown to be limited to fixations on the eyes and nose.
Autism research : official journal of the International Society for Autism Research, 2014 · doi:10.1002/aur.1340