Emily Levin^1,3 (), Jasmine Horton², J. Patrick Mayo³; ¹University of Pittsburgh School of Medicine, ²Hillman Academy, ³University of Pittsburgh

Eye movements help us navigate the world successfully. Saccades allow us to rapidly reorient our gaze to an object of interest, and smooth pursuit allows us to track a moving object. Understanding how these eye movements interact is important because saccadic and pursuit abnormalities are a common presentation in movement disorders. Although many studies have demonstrated the interplay between saccades and fixation in laboratory settings, far fewer have examined the interaction between saccades, fixation, and pursuit in naturalistic settings such as when driving. The goal of this experiment was to examine how eye movement dynamics vary with behavioral performance when participants played the Nintendo 64 racing videogame ExciteBike. We hypothesized that participants who performed better at ExciteBike, as measured by faster lap times and fewer crashes, would spend more time tracking the bike they were controlling and less time saccading to irrelevant obstacles. Therefore, we predicted that better performers would exhibit fewer saccades and more pursuit. To test this idea, we recorded participants’ eye movements using Tobii Pro Glasses 3 and defined individual eye movement events (saccades, pursuit, and fixation) using an automated classification algorithm. We next correlated time around the circular track with number of saccades and pursuits. We found that there was a significant correlation between track time and number of saccades, with participants that had faster track times making fewer saccades. Additionally, there was a nonsignificant correlation between track time and pursuit events, but this trended opposite to the direction we predicted: top performers instead had fewer pursuits than poorer performers. There was also a significant negative correlation between performance and fixation events, with better performers tending to have fewer fixations. These results suggest that eye movement interactions could be used as a measure of behavioral performance in motion-rich, naturalistic settings.