The human visual system masks the perceptual consequences of retinal or cortical lesion-induced scotomas by predicting what is missing from nearby regions of the visual field. To reveal the neural mechanisms underlying this remarkable capacity, known as predictive masking, we used fMRI and neural modeling to track changes in cortical population receptive fields (pRFs) and connectivity in response to the introduction of an artificial scotoma (AS). Consistent with predictive masking, we found that extrastriate areas increased their sampling of the V1 region outside the AS projection zone. Moreover, throughout the visual field and hierarchy, pRFs shifted their preferred position towards the AS border. A gain field model, centered at this border, accounted for these shifts, especially for extrastriate areas. This suggests that a system-wide reconfiguration of neural populations in response to a change in visual input is guided by extrastriate signals and underlies the predictive masking of scotomas.

Predictive masking is associated with a system-wide reconfiguration of neural populations in the human visual cortex