Abstract
Smooth pursuit eye movements follow moving objects to prevent retinal image blurring. While modeled as motion-driven, the pursuit system is studied with a weak and unnatural motion stimulus, a small, foveal spot more ideal for inducing saccades, which shift gaze between stationary objects. We showed previously that large, random dot stimuli (10 deg) invoke fewer saccades than the spot, and increase open-loop gain (Heinen & Watamaniuk, 1998). This could occur if motion summation created a stronger pursuit signal, or, alternatively, if peripheral stimulation reduced the need to reposition the fovea. To test between these alternatives, observers pursued a single dot, or configurations of 4 and 5 dots, designed to differentially probe foveal and peripheral contributions to pursuit. The 5-dot stimulus was arranged in a 6 deg "+" configuration. The 4-dot stimulus was arranged identically, but with no central spot. Consistent with our previous results, 4 dots yielded fewer saccades and higher open-loop gain than the single spot. However, while the 5-dot stimulus further increased open-loop gain, it induced almost as many saccades as the spot, contradicting the summation hypothesis, and suggesting that the salient foveal spot was triggering saccades. In a second experiment using only 4-dot configurations with different diameters (1-12 deg), more saccades were present when the configuration was small and stimulated the fovea, but open-loop gain remained constant with configuration size. The results provide evidence that two independent mechanisms subserve pursuit, one that corrects position error between the fovea and a salient feature, and another that is driven by motion. We characterize the sensitivity of position and motion mechanisms to the different dot configurations using motion/position "Punnett squares", analogous to the characterization of the expression of phenotypes in genetics.
Meeting abstract presented at VSS 2014