Assessing Adaptive Behavior in Individuals with Intellectual Disability through Head Movement Analysis in Virtual Reality

Adaptive behavior plays a central role in independent functioning for individuals with intellectual disabilities (ID), yet existing assessments rely on caregiver reports, limiting objectivity and ecological validity. We present a virtual reality (VR)-based adaptive behavior assessment leveraging head movement features. Using a head-mounted display, we collected head movement data from 32 individuals with ID as they completed a virtual greeting test consisting of response selection, gesture selection, and bodily greeting tasks. Four head movement features (i.e., linear displacement linear velocity, angular displacement, and angular velocity) were extracted and compared between high and low adaptive behavior groups. While no significant group differences were observed in the response and gesture selection tasks, all features significantly differentiated the two groups during the bodily greeting task, where a support vector machine classifier achieved 80.95% accuracy. These findings suggest that embodied head movement in VR provides meaningful behavioral signals for assessing adaptive behavior in ID.