Eye-tracking based social visual engagement may predict autism among kids: JAMA

This double-blind, multisite study of 475 children aged 16 to 30 months, conducted in six speciality clinics, found that measuring social visual engagement had a sensitivity of 71.0% and a specificity of 80.7% in determining autism diagnosis. In the subgroup of children whose autism diagnosis was definite (n=335), the test had a sensitivity of 78.0% and a specificity of 85.4%.

This Original Investigation was published on September 5, 2023, entitled “Eye-Tracking–Based Measurement of Social Visual Engagement Compared With Expert Clinical Diagnosis of Autism by a team of researchers led by Warren Jones and colleagues.

The concern here is: Can eye-tracking be used to accurately measure social visual engagement in young children, potentially aiding in the early diagnosis and assessment of autism?

In the United States, children displaying symptoms of autism often experience a delay of over a year in receiving a diagnosis, and this delay is even longer for those from racially, ethnically, or economically disadvantaged backgrounds. Furthermore, most diagnoses are not made using standardized diagnostic tools. To address this issue, researchers in the present study used eye-tracking technology for early autism diagnosis. This study was completed in a 1-day protocol for each participant, and test sensitivity and specificity were measured as the primary outcomes.

The key results of the study are:

• Eye-tracking measurement of social visual engagement was successful in 475/499 enrolled children.
• The mean age of participants was 24.1 months.
• The number of Asian, Black, White, Hispanic and other participants were 38, 37, 352, 68 and 44, respectively.
• Two hundred twenty-one children, constituting 46.5%, had autism, and 254, comprising 53.5%, did not have a diagnosis (by expert clinical diagnosis).
• In all children, the measurement of social visual engagement had sensitivity and specificity of 71.0% and 80.7%, respectively.
• In the subgroup of 335 children whose autism diagnosis was certain, researchers recorded sensitivity and specificity of 78.0% and 85.4%, respectively.
• There was a correlation between Eye-tracking test results and expert clinical assessments of individual levels of social disability, verbal ability and nonverbal cognitive ability.

Concluding further, eye-tracking–based measurement of social visual engagement is predictive of autism diagnoses.

Further research is needed to determine the effectiveness of this test in the early diagnosis and assessment of autism in routine speciality clinic practice.

Further reading:

https://jamanetwork.com/journals/jama/article-abstract/2808996