Eye Tracking Infant Social Attention and Autism

It is widely accepted that there is a critical need for early and reliable identification of Autism Spectrum Disorde (ASD), a neurodevelopmental condition characterized by challenges in social interaction and communication. Fu, Platt, Shic, and Bradshaw (2024) critically evaluated the utility of eye-tracking (ET) technology in identifying early markers of ASD by comparing it with traditional manual coding (MC) methods. The results of their research are detailed in their paper “Infant social attention associated with elevated likelihood for autism spectrum disorder: A multi-method comparison,” published in the Journal of Autism and Developmental Disorders. 

Traditional methods for observing infant attention, such as manual coding, while established, are often subjective and labor-intensive. Eye tracking, on the other hand, offers a high temporal and spatial resolution, enabling a more precise and direct assessment of gaze allocation, a key indicator of social attention. Previous research has hinted at the promise of ET as a biomarker for ASD, given that reduced social attention is a recognized characteristic of the disorder. However, concerns regarding data loss and quality in infant ET studies, especially with younger infants and those at elevated likelihood for ASD, have persisted.

Developmental Social Attention and Eye Tacking Methodology

The longitudinal study involved 72 infants (25 with elevated familial likelihood (EL) for ASD and 47 with low likelihood (LL) for ASD), observed at multiple time points between 3 and 24 months of age. The infants participated in a screen-based attention task featuring both social and nonsocial videos. The researchers simultaneously collected data using both ET and MC methods. Eye movements were recorded with an SR Research EyeLink 1000 eye tracker, which is the eye tracker of choice for many developmental research groups, thanks to its high levels of accuracy and low levels of data loss. The study specifically compared the two methods across three measures: percentage of time looking at the screen, percentage of time looking off-screen, and percentage of data loss. A mixed-effects model was used to analyze the data, examining the effects of method, infant age, and familial likelihood on attention measures.

Eye Tracking is a Viable Tool for Identifying Elevated Likelihood for Autism Spectrum Disorder

The research yielded several crucial findings:

• Feasibility and Success Rate: Despite concerns, ET data was obtained with a success rate of 95.83% across all ET sessions for infants aged 3 to 24 months. This demonstrates the practical applicability of ET in this young population.
• Data Loss Comparison: While ET did show a higher degree of data loss and yielded lower percentages of screen looking and off-screen looking compared to MC, this data loss was not significantly impacted by infant age or ASD likelihood status. This suggests that ET can still provide valuable insights across different age groups and ASD risk levels.
• High Concordance Between Methods: A significant positive association was found between ET and MC measures of screen looking. More importantly, both ET and MC yielded comparable findings regarding age-related changes in social attention across LL and EL infants. Specifically, both methods revealed that LL infants exhibited greater screen looking towards social videos compared to EL infants from 3 to 24 months. Furthermore, LL infants, but not EL infants, showed an age-related increase in their attention to social versus nonsocial information. This high concordance is a powerful validation of ET, indicating that it captures the same critical developmental patterns as MC.
• Identification of Early Risk Markers: The consistent findings from both ET and MC, particularly the observation that EL infants did not show the same developmental emergence of increased attention to social information as LL infants, reinforce ET’s potential as a promising approach for identifying very early markers associated with ASD likelihood. 

This study provides compelling evidence that eye tracking is a viable and valuable tool for measuring social attention in infants with elevated likelihood for ASD. As a temporally and spatially sensitive tool, ET offers a powerful avenue for identifying critical early risk markers, paving the way for timely interventions and improved support for children at risk for ASD.

For information regarding how eye tracking can help your research, check out our solutions and product pages or contact us. We are happy to help!