Datensatz

Zusammenfassung

Microsaccades (MSs) are the largest “fixational” eye movements, but their role in visual perception is not fully understood. Some recent studies suggest that in natural vision, there may be a direct link between MS activity and perceptual improvement. Other evidence demonstrates that MS dynamics are influenced by cognitive factors including spatial attention, working memory, and ocular and manual motor preparation, suggesting that their occurrence relates to the active sampling of the environment. We used a visual illusion to examine the effect of perceptual state on MS patterns. We analyzed the eye movements of three monkeys performing the Generalized Flash Suppression (GFS) task (Wilke et al, 2006). In GFS the presentation of a salient target is followed by the abrupt onset of a surround pattern, at which point the target either remains subjectively visible (“visible” trials) or seems disappeared completely (“invisible” trials). Eye movements were monitored using scleral search coils, and were processed offline to detect MSs within the first second, in which the target was either visible or invisible. We found that when fixating subjects experienced GFS, MSs after the presentation of the surrounding patterns occurred in about 5% and 35% of the “invisible” and “visible” trials respectively, suggesting a large influence of perceptual state on the occurrence of microsaccades. Immediately after the surround onset, the occurrence rate of MSs consistently dropped to near-zero for both visible and invisible trials. For the visible trials, the MS rate promptly rebounded to the pre-onset level, whereas for the invisible trials the rate remained low, reaching pre-onset levels hundreds of milliseconds later. In addition, MS direction was affected, in that it showed different distributions, reflecting a bias of MS direction to the target positions in the “visible” condition. In contrast, amplitude, duration and peak velocity of MS showed no significant correlation with perception. The results show that the measured MS dynamics are influenced by physical stimulus changes (i.e. the onset of the surround), but also correlated with the subjective perceptual interpretation of a single physical stimulus that is either subjectively visible or invisible on alternate trials.