Comparing Automatic Eye Tracking and Manual Gaze Coding Methods in Young Children with Autism Spectrum Disorder

Area of Science:

• Neuroscience
• Developmental Psychology
• Clinical Psychology

Background:

• Eye-gaze methods are valuable for studying cognitive processes in children with autism spectrum disorder (ASD).
• Data loss in eye-gaze studies can compromise the validity and generalizability of research findings.
• Empirical investigation into data loss rates across different eye-gaze systems is lacking.

Purpose of the Study:

• To compare data loss rates between automatic eye-tracking and manual gaze coding in toddlers with ASD.
• To determine if different eye-gaze systems yield divergent results in ASD research.
• To inform the selection of appropriate eye-gaze methodologies for autism research.

Main Methods:

• Simultaneous data collection from 51 toddlers with ASD using both automatic eye-tracking and manual gaze coding.
• Analysis of data loss based on the number of usable trials and proportion of looks to images per trial.
• Comparison of results patterns obtained from the two eye-gaze measurement systems.

Main Results:

• Manual gaze coding demonstrated significantly lower data loss compared to automatic eye-tracking.
• The proportion of usable trials and looks to images was higher with manual gaze coding.
• Distinct patterns of results emerged between automatic eye-tracking and manual gaze coding.

Conclusions:

• The choice of eye-gaze system significantly influences data loss and research outcomes in studies involving children with ASD.
• Methodological decisions regarding eye-gaze data acquisition can impact the interpretation and conclusions drawn from autism research.
• Researchers should carefully consider the strengths and limitations of different eye-gaze systems for their specific research questions and populations.