Smooth-pursuit and saccadic eye movements: One clock or two?

It is commonly believed that saccadic and smooth-pursuit eye movements involve different neural and behavioural organization. However, the question of how these different movements are timed with respect to a pacing goal is less understood. In two experiments, we investigated the idea that repetitive saccadic and smooth pursuit eye movements might share the same clocking mechanisms or neural networks for timing. Eye movements exhibit predictive tracking (negative latencies) at fast frequencies and reactive (positive latencies) tracking at slow frequencies. In the first experiment, monotonic changes in pacing (speeding up and slowing down) were used to establish the point at which a transition occurs between reactive and predictive behaviours. The point of transition exhibited a strong hysteresis depending on the direction of the monotonic pacing. In subsequent trials, subjects were instructed to make smooth pursuit movements up to the point of transition after which they switched to making saccades. The saccade latencies were compared between saccade-only and switched trials. Although no uniform pattern was seen across individuals, some subjects were able to track nearly identically to baseline trials after the target switch. In the second experiment, subjects were instructed to continue making saccadic eye movements after being entrained to either a saccadic or smooth pursuit stimulus train using a synchronization-continuation design. The decomposed clock and motor variance measures and correlations at higher lags showed no differences between synchronizing conditions. Our results indicate that timing is represented globally without particular reference to the effector system for broad classes of repetitive oculomotor tasks