Intra-V1 functional networks predict observed stimuli

Abstract

A recent fMRI study found that the patterns of fMRI co-fluctuations within V1 changed with variations in the perceptual organization of observed stimuli. Intra-V1 correlation matrices associated with modified Global and Local Navon letters (that allowed dissociation of levels over time) were examined with univariate tests. Links between left/right V1 areas were found for Global letters, whereas links for Local letters were concentrated within the left dorsal V1. Here we used multivariate pattern analysis on the same data to predict the level and shape of the observed Navon letter from V1 correlation matrices to ascertain the stability across subjects of network topologies and if they contained invariant information about shape or level. We found that inter-subject classification was accurate for both level and letter shape. Intra-participant cross-classification of levels across shape was accurate but failed for shape across levels. Furthermore, cross-classification weight maps evinced asymmetries of link strengths across the visual field that mirrored perceptual asymmetries. These results indicate that the association of V1 topologies and perceptual states is stable across participants. We hypothesize that feedback (that differs between level and shape) to V1 drives the intra-V1 networks and that the topology of intra-V1 networks can shed light on the neural basis of perceptual organization.