Abstract: PURPOSE: When one’s central vision is deprived, a spared part of the peripheral retina acts as a pseudo fovea for fixation, which is termed as preferred retinal locus (PRL). Previously, we demonstrated that oculomotor training with simulated central vision loss not only induced a PRL in normally sighted adults, but also reduced crowding at the PRL (Chen et al., 2019). Does this compensatory adjustment involve changes in information communication in the visual processing network? Here we addressed this question by performing functional connectivity analyses on the BOLD fMRI signals recorded before and after training.
METHODS: During the scan, crowded letters were displayed at the PRL while subjects were engaged in a central fixation task. Background connectivity was computed based on residual timeseries after removing stimulus-evoked signals. Voxels in the extrastriate cortex and in the intraparietal sulcus (IPS) that showed a stronger response to stimuli were identified as seeds for computing correlation with vertexwise V1. We compared background connectivity around the region covering the crowded letters before and after training.
RESULTS: After training, the connectivity was enhanced between V1 and V2, and between V1 and IPS. This effect was observed at the retinotopic regions representing the crowded target.
CONCLUSIONS: PRL training enhanced the background connectivity between V1 and V2, and between V1 and IPS. These results suggest that learning enhances peripheral vision by reweighting information transmission in the visual processing hierarchy in the human visual cortex.

Key words: fMRI, peripheral vision, functional connectivity, preferred retinal locus, crowding