The Critical Role of V2 Population Receptive Fields in Visual Orientation Crowding

Crowding, the identification difficulty for a target in the presence of nearby flankers, is an essential bottleneck for object recognition and visual awareness [1, 2]. As suggested by multitudes of behavioral studies, crowding occurs because the visual system lacks the necessary resolution (e.g., small receptive field or high resolution of spatial attention) to isolate the target from flankers and therefore integrates them mistakenly [3-12]. However, this idea has rarely been tested with neuroscience methods directly. Here, using the fMRI-based population receptive field (pRF) technique [13, 14], we found that, across individual subjects, the average pRF size of the voxels in V2 responding to the target could predict the magnitude of visual orientation crowding. The smaller the pRF size, the weaker the crowding effect. Furthermore, we manipulated the magnitude of the crowding effect within subjects. The pRF size in V2 was smaller in a weak crowding condition than in a strong crowding condition, and this difference was attention dependent. More importantly, we found that perceptual training could alleviate the orientation crowding and causally shrink the pRF size in V2. Taken together, these findings provide strong and converging evidence for a critical role of V2 pRFs in visual orientation crowding. We speculate that, synergistic with spatial attention, the dynamic and plastic nature of the V2 pRFs serves to prevent interference from the flankers through adjusting their size and consequently reduces visual crowding.