Abstract
During a saccade, a copy of the oculomotor command updates spatial representations and suppresses motion perception in order to yield perceptual stability. A similar need for a stabilization mechanism also exists during visual fixation, when microscopic eye movements incessantly shift the retinal image. In a recent study, we have shown that motion judgments can be predicted on the basis of retinal image motion irrespective of the amount of ocular drift (Poletti et al., J. Neurosci. 2010).These findings suggest that, unlike during saccades, visual stability during fixational drift relies on the motion of the stimulus on the retina rather than on an extraretinal signal. However, the possibility remains that an extraretinal drift signal which eludes awareness is available in the motor system. Discrepancies between action and perception have been previously reported in the literature. Here, we examined whether an extraretinal signal associated with ocular drift influences the execution of a later saccade. Human observers were asked to maintain fixation in complete darkness and saccade toward a briefly cued location. The saccade was executed after an auditory signal which came 2 s after the display of the target cue. In this interval, drift moved the eye away from the position assumed at the time of display of the target cue. Across subjects, ocular drift was on average larger than 40′. If motor knowledge of this drift is available to the brain, saccadic amplitude and direction should be properly adjusted so that the saccade lands on the target. We show, instead, that saccades do not compensate for the preceding drifts, and that the distance between the target and the landing position increases with the amount of drift. These results show that the visual system does not use a motor signal proportional to drift to compensate for possible shifts in the location of gaze.
NIH R01 EY18363 grant and NSF BCS-0719849 grant.