Eye Tracking Area of Interest in the Context of Working Memory Capacity Tasks

Abstract

Adults diagnosed with Attention-Deficit / Hyperactivity Disorder (ADHD) have reduced working memory capacity, indicating attention control deficits. Such deficits affect the characteristic movements of human gaze, thus making it a potential avenue to investigate attention disorders. This paper presents a converging operations approach toward the objective detection of neurocognitive indices of ADHD symptomatology that is grounded in the cognitive neuroscience literature of ADHD. The development of these objective measures of ADHD will facilitate its diagnosis. We hypothesize that the characteristic movements of human gaze within specific areas of interests (AOIs) may be used to estimate psychometric measures and that distinct eye movement scan patterns can be used to better understand ADHD. The results of this feasibility study confirm the utility of a combination of fixation and saccade feature set captured within specific AOIs indexing Working Memory Capacity (WMC) as a predictor of a diagnosis of ADHD in adults. Tree-based classifiers performed best in-terms of predicting ADHD with 86% percent accuracy using physiological measures of sustained visual attention during a WMC task.