Poster F5, Tuesday, March 28, 8:00 – 10:00 am, Pacific Concourse
Eye Movement Patterns During Scene Viewing Predict Clinical Individual Difference Measures
Taylor R. Hayes1, John M. Henderson1,2; 1Center for Mind and Brain, University of California, Davis, 2Department of Psychology, University of California, Davis
An understudied area is the degree to which clinical individual differences associated with attentional deficits influence real-world viewing behaviors. The present study investigated this issue by predicting individual differences in attention-deficit disorder, autism spectrum disorder, and dyslexia measures using eye movement scan patterns during real-world scene viewing. Forty participants completed a free-view memorization task for 40 real-world scenes while their eye movements were recorded. Three different area of interest grids were used to represent each observers' tendencies to shift their attention vertically, horizontally, and from central to peripheral scene regions. Successor Representation Scanpath Analysis (SRSA, Hayes, Petrov, & Sederberg, 2011) was used to extract the regularities in each participant's scan patterns across these 3 area of interest grids. SRSA quantifies regularities in scan patterns using temporal-difference learning to construct a fixed-size matrix called a successor representation (Dayan, 1993) that captures the underlying structure in temporally-extended scan patterns. A principal component analysis of participant successor representations was performed for each individual difference measure, and these components were then used to predict clinical individual difference scores. The best leave-one-out cross validation was achieved with the vertical area of interest grid which demonstrated significant prediction across all 3 clinical individual difference measures: attention-deficit disorder (r2=0.53), autism spectrum disorder (r2=0.33), and dyslexia (r2=0.23). These results suggest that individual differences in attention-deficit disorder, autism spectrum disorder, and dyslexia significantly influence vertical scan patterns when viewing real-world scenes. The results also suggest that scan patterns might provide useful diagnostic information about individual differences in attentional control.
Topic Area: ATTENTION: Spatial