Related Experiment Video
Updated: Mar 23, 2026

09:03
Eye Tracking Young Children with Autism
Published on: March 27, 2012
46.7K
Development of an Objective Autism Risk Index Using Remote Eye Tracking
Thomas W Frazier1, Eric W Klingemier1, Mary Beukemann1
1Center for Autism, Cleveland Clinic, Cleveland.
Summary
A new autism risk index, based on eye tracking, accurately identifies autism spectrum disorder (ASD) in children. This objective measure shows strong correlation with symptom severity and can aid in clinical judgment.
Area of Science:
- Neuroscience
- Developmental Psychology
- Biomedical Engineering
Background:
- Abnormal eye gaze patterns are a recognized characteristic of autism spectrum disorder (ASD).
- Previous research has identified distinct attention differences in individuals with ASD.
- Objective and reliable measures for assessing autism risk are needed.
Purpose of the Study:
- To develop an objective, eye-tracking-based index for assessing autism risk.
- To evaluate the diagnostic accuracy and clinical utility of this novel autism risk index.
Main Methods:
- Children referred for ASD evaluation were recruited for initial and replication studies.
- Remote eye tracking was employed, with dwell times recorded across predefined regions of interest.
- An autism risk index was calculated by averaging dwell times, and diagnostic accuracy was assessed using ROC curve analysis.
Main Results:
- The autism risk index demonstrated high diagnostic accuracy in both study samples (AUCs = 0.91 and 0.85).
- The index strongly correlated with Autism Diagnostic Observation Schedule-Second Edition (ADOS-2) severity scores.
- No significant correlation was found between the autism risk index and language ability.
Conclusions:
- The developed autism risk index shows potential as a quantitative and objective measure for identifying autism risk.
- Further validation in larger, diverse samples is warranted for clinical application.
- This tool could support clinical decision-making in ASD diagnosis and symptom monitoring.

