Rapid visual stability for action

点击翻译按钮获取中文摘要

About 3-5 times each second we make fast eye movements called saccades in order to examine the world around us. Saccades are necessary to bring the fovea, the part of the retina with the highest acuity, to different areas of a visual scene. However, such manner of exploration poses significant challenges since the locations of objects on the retina change every time we move our eyes. Nevertheless, typically we do not experience that objects shift their positions during saccades and the world is perceived as a coherent and stable whole. More than a century scientists have been working on this intriguing and at the same time fundamental problem in vision. Some of the theories put more weight on the role of extra-retinal signals (e.g. the efferent copy), while others emphasize the role of visual information after saccade. Some theories propose that spatiotopic representations are not default and emerge only if relevant for the task. The way to understand these controversial findings is to assume a difference in visual stability for action and visual stability for perception. This is similar to a previously made distinction between representations for perception and for action. The current project proposes a novel theoretical approach, suggesting that rapid spatiotopic representations are primarily formed and used by the motor system, since they are critical for interaction with the objects despite the intervening saccades and are less critical for perception.