Abstract
background: Our eyes are always in motion, even during fixation, but tend to “freeze” in response to stimulus onset. This phenomenon is known as “oculomotor inhibition” (OMI) and its time-course and magnitude depend on stimulus parameters, attention and expectation. It applies to microsaccades and eye-blinks in strict fixation, with some evidence for OMI in drift. In previous experiments, we found that a flashed Gabor patch during smooth pursuit induced a velocity slowdown (“pursuit-inhibition”) followed by acceleration, as well as inhibition of catch-up saccades (< 2°), with stronger inhibition and faster release for more salient stimuli, demonstrating that OMI applies to smooth pursuit. The purpose of the current study was to re-examine ocular drift inhibition and its generalization to conditions of smooth pursuit. Methods: In two experiments, observers (N=18) followed a small circle surrounded by a bright circular envelope, which moved back and forth horizontally at 6.2°/sec. While tracking, a Gabor patch was briefly flashed (100ms) at 0.5Hz rate, at the display center with varied spatial-frequency and contrast. We computed drift velocity and drift “box-counting” across time, after filtering out all saccade traces and the pursuit movement itself. Results: Drift during pursuit, like microsaccades and the pursuit itself, showed initial slowdown flowed by increase in amplitude and speed depending on the stimulus parameters. This occurred even when all saccades and the pursuit movement were carefully filtered out. Furthermore, we found no correlation between pupil size and the filtered horizontal trace, and the results persisted even in one case of artificially dilated pupil, evidence against a pupil-drift artifact of the eye-tracker. Finally, we obtained a similar drift inhibition for a static target in comparison. Conclusion: The OMI phenomenon applies to drift during pursuit as well as during fixation. This may indicate a general mechanism to attenuate incoming information while processing previous stimuli.