Remote Eye Tracking Could Effectively Measure Autism Risk

Objective measures of autism symptom severity are necessary to accurately diagnose autism and measure the effectiveness of treatment interventions.

Researchers from the Center for Autism at the Cleveland Clinic have found that an objective autism risk index (ARI) created using remote eye tracking may be an effective measurement of autism risk.

The findings were published in the Journal of the American Academy of Child & Adolescent Psychiatry.

“Deficits in eye gaze are a hallmark feature of autism spectrum disorder (ASD) and are included in gold-standard diagnostic instruments. More than a decade of research into abnormalities of eye gaze has confirmed social attention deficits as a key feature of ASD,” Thomas W Frazier, PhD, of the Cleveland Clinic, and colleagues wrote.

“This implies that eye gaze patterns, particularly based on dynamic temporal analysis, may be a promising objective risk marker of ASD as well as a quantitative measure of autism symptoms spanning the full continuum of behavior.”

The researchers noted that objective measures of autism symptom severity are necessary not only to accurately diagnose autism, but also to provide a quantitative assessment for tracking the effectiveness of treatment interventions.

To attempt to create an objective, eye tracking-based autism risk index, the researchers recruited children ages 3 to 8.11 years who were referred for comprehensive multidisciplinary evaluation of ASD. They grouped the children into 2 groups, one for the initial study sample, and the second for the replication study sample. In the initial study group, 25 children with ASD and 20 controls without ASD were studied, and in the replication study group, 15 children with ASD and 19 controls were studied.

As the children watched images on a television presenting stimuli with various regions of interest (ROI), the researchers collected eye tracking data and found that in both the initial and replication groups, the autism risk index had high diagnostic accuracy (area under the curve [AUC] = 0.91 and 0.85, 95% CIs = 0.81–0.98 and 0.71–0.96), was strongly associated with Autism Diagnostic Observation Schedule–Second Edition (ADOS-2) severity scores (r = 0.58 and 0.59, p < .001), and was not significantly correlated with language ability (r ≤ |–0.28|, p > .095).

“The ARI, based on eye gaze to social and nonsocial information, may be a useful quantitative and objective measure, informing clinical judgment regarding the presence of an ASD diagnosis in at-risk settings,” the researchers wrote. “The ARI and, by extension, eye tracking-based attention measures, may supplement existing clinical observation measures for grading the severity of autism symptom levels.”

The researchers concluded that more research is needed with larger samples to cross-validate these findings—but if these findings are validated and scaled for clinical use, this measure could inform clinical judgment of ASD diagnosis and serve as a way to track improvements in symptoms.


Frazier TW, Klingemier EW, Buekemann M, et al. Development of an Objective Autism Risk Index Using Remote Eye Tracking. Am Acad Child Adolesc Psychiatry. 2016;55(4):301-309. doi:10.1016/j.jaac.2016.01.011.