Abstract
INTRODUCTION The diagnosis of Parkinson's Disease (PD) occurs after symptoms are present; therefore, predicting PD emergence is critical. PD often includes ocular symptoms: nystagmus, difficulty initiating and maintaining smooth pursuit, and more saccades during pursuit. We studied control behavior during a pursuit task for future comparison with PD patients, assessing how inter-trial interval (ITI) affected tracking of a target that deflected behind an occluding wedge. METHODS Eye tracking of eight subjects was performed using the Eyelink II (500 Hz). In the first task, a target travelled horizontally behind occluders of various sizes (2-6°, 6 trials/occluder). To guarantee saccades during occlusion in the second task, we chose the minimum wedge size resulting in saccades during occlusion at least five of six trials. In the second task, the target deflected behind the wedge diagonally up or down with unequal probability (11 versus 4 of 15 trials). After 15 trials, the more frequent direction covertly flipped and the ITI lengthened from 10 to 15 seconds. We measured saccade latency (time between target reappearance and catch-up saccade to the target) and post-saccadic error across trials. RESULTS Among more-frequent deflection trails, saccade latency decreased significantly across the first set of 15 trials (ITI = 10s) [p=0.025], but latency was constant at a higher level across the second set of 15 trials (ITI = 15s). Post-saccadic error marginally increased across the first set of 15 trials (p=0.09), but did not change across the second set of 15 trials. CONCLUSION Subjects' latency and post-saccadic error changed most across the first half of trials, indicating that subjects honed their tracking strategy during the first half of trials. The increased post-saccadic error and decreased saccadic latency implies an increasing reliance on a predictive tracking strategy, as subjects grew more confident in the higher-frequency deflection.
Meeting abstract presented at VSS 2018