Abstract
When a saccade is directed towards a translating target, smooth pursuit movements track the target from the moment of saccade landing, indicating that motion integration occurred prior to the saccade (Gardener and Lisberger, 2001). Since, prior to saccades, perceptual performance improves at the saccade target (Kowler et al, 1995; Deubel and Schneider, 1996; Rolfs and Carrasco, 2012), we hypothesized that saccades to a motion stimulus in a stationary aperture would drive post-saccadic pursuit movements due to the pre-saccadic selection of its motion. Participants performed a saccade to one of four motion apertures, cued by a central line. Apertures consisted of random dot fields (5 deg eccentricity and diameter, 100% coherent motion) moving in one of two randomly assigned radial directions tangential to the center-out saccade. Saccades exhibited a low gain (~10%) pursuit along the target's motion direction at saccade landing. These effects were driven by motion integration prior to the saccade, as we found consistent results when the motion stimulus offset occurred during the saccade. These effects grew as we reduced the spatial certainty of the aperture location, from a well-defined ring aperture, to no ring, or to a smoothed Gaussian envelope. Pursuit velocity increased with increasing stimulus speed with gain saturating at speeds higher than 10 deg/s. To examine what period prior to the saccade contributed to motion integration, we presented stimuli with random motion (0% coherence) that transitioned to coherent motion (either permanently or for fixed 100 ms epochs) around the time before saccade onset. We found that a minimum of 100 ms motion integration was necessary to observe an effect, with 150-50 ms before the saccade providing the strongest input. These results suggest that presaccadic attention engages motion integration for the saccade target that can be observed as an involuntarily low gain pursuit upon saccade landing.
Meeting abstract presented at VSS 2018