We leverage virtual reality to collect gaze and position data to create a predictive model and a mixed effects model to show gaze orientation precedes other features