Distributed resting-state network interactions linked to the generation of local visual selectivity

Cocuzza, C.V., Sanchez-Romero, R., Ito, T., Mill, R.D., Keane, B.P., & Cole, M.W. (2022). Distributed resting-state network interactions linked to the generation of local visual category selectivity. In bioRxiv (p. 2022.02.19.481103).

Abstract
Understanding processes fundamental to brain functioning will require knowledge of how functionally-relevant localized brain processes are generated. We identified such functionally-relevant brain activations using fMRI data from N=352 human participants, focusing on select visual cortex regions given long-standing observations that they exhibit localized visual category selectivity. First, we systematically tested the hypothesis that local visual category selectivity can be generated by highly distributed, resting-state network interactions. This was accomplished using a recently developed distributed network interaction approach for mapping task-evoked brain activations by explicitly modeling the influence of activity flowing over resting-state brain connections. Next, we tested refinements to our hypothesis based on category selectivity being generated by stimulus-driven network interactions initialized in V1. We found evidence in support of the refined hypothesis that V1-initialized activity flow processes that were further shaped by fully distributed (i.e., whole-cortex), intrinsic network interactions were sufficient to generate visual category selectivity (but not when further shaped by later visual network interactions alone). Further, using null network architectures we found that each region’s unique resting-state “connectivity fingerprint” was key to category selectivity being sufficiently generated by distributed network interactions. These results generalized across regions associated with four visual categories (processing images of bodies, faces, places, and tools), and provides evidence that intrinsic network organization plays a prominent role in the generation of local, functionally-relevant responses.

pdf download: https://drive.google.com/file/d/1s_Y9f_j_uJsar8GHPnDepwE5BqXWu3p0/view?usp=sharing