Visualization and analysis of eye movement data from children with typical and atypical development

Looking at other children's interactions provides rich learning opportunities for a small child. How children with autism look at other children is largely unknown. Using eye tracking, we studied gaze performance in children with autism and neurotypical comparison children while they were watching videos of semi-naturalistic social interactions between young children. Using a novel, bottom-up approach we identified event-related measures that distinguished between groups with high accuracy. The observed effects remained in a subset of the total sample matched on IQ, and were replicated across several different stimuli. The described method facilitates the detection of meaningful patterns in complex eye tracking data. Also, the approach significantly improves visualization, which will help investigators understand, illustrate, and generate new hypotheses.