Transsaccadic vision is typically described as the ability to retain and integrate visual information across rapid eye movements (saccades) for the purpose of continuous perception. In this review, we break this down further into two fundamental processes: the cortical mechanisms that maintain extrinsic spatial properties (i.e., location, orientation) versus intrinsic identity cues (i.e., spatial frequency, shape) of objects across saccades. First, we summarize the computational problems associated with transsaccadic vision, emphasizing the need to retain and integrate both extrinsic and intrinsic feature information between fixations for predictive vision. Next, we briefly review psychophysical, neurophysiological, stimulation, and neuroimaging evidence that specific cortical mechanisms exist for these processes. Then, we focus on our recent experiments combining functional magnetic resonance imaging adaptation-inspired and functional connectivity approaches. These experiments suggest that parietal cortex (specifically, right supramarginal gyrus) contributes to spatial updating of object orientation, whereas dorsomedial occipital cortex (cuneus) supports updating of spatial frequency and object shape. Taken together with the previous literature, this supports the notion that an ‘extended parietal eye field’ is involved in the updating of extrinsic spatial cues, whereas dorsomedial occipital cortex is involved in updating intrinsic cues to object identity, both in communication with the broader functional networks for perception and action. Finally, we synthesize these results in a simple conceptual model for cortical updating and integration of object features and consider the implications for future studies of transsaccadic integration, the ‘binding problem’, and possible clinical applications.
Baltaretu et al. (Mon,) studied this question.