Abstract
How the visual world remains stable across the frequent movements of the eyes is one of the long-standing mysteries in neuroscience. We tested whether an active remapping process constructs a spatiotopic neural representation, which is reflected in behavior and might thereby enable visual stability. We combined behavioral visual adaptation and fMRI - adaptation. This combination allowed tracking spatiotopic and retinotopic specificity across eye movements in both, neural activations and behavioral performance. Subjects saw an oriented gabor patch for 3 seconds, followed by a saccade target. Subjects were required to keep fixation for further 1000 ms to give the perceptual system time to build up a spatiotopic representation. Following the saccade, a probe gabor patch was flashed and subjects had to judge its orientation. The probe gabor patch was shown either in the same spatiotopic position as the adapter, in the retinotopically matched position or in a neutral control condition. We contrasted neural activation and behavioral responses to baseline trials in which no adapter was shown. We found behavioral adaptation in both the retinotopic and the spatiotopic condition. Significant clusters of neural adaptation were found in both conditions in contralateral visual areas V1-V4. In order to establish adaptation in the hemisphere which was not adapted before the saccade, adapter activity must have been actively remapped contingent to the saccade. No adaptation was found in the control condition, neither in behavior nor in neural activations, thus ruling out global adaptation which spreads unspecifically across the visual cortex. The remapping to be behaviorally relevant needed a certain duration to build-up. The results show that visual features are actively remapped in early visual areas across saccade eye movements.
Meeting abstract presented at VSS 2014