Abstract #0353

Identifying Common-Source Driven Correlations in Resting-State FMRI Via Laminar-Specific Analysis in the Human Visual Cortex

Jonathan Rizzo Polimeni¹, Thomas Witzel^1,2, Bruce Fischl^1,3, Douglas N. Greve¹, Lawrence L. Wald^1,2

¹Athinoula A. Martinos Center for Biomedical Imaging, Department of Radiology, Harvard Medical School, Massachusetts General Hospital, Charlestown, MA, United States; ²Harvard-MIT Division of Health Sciences and Technology, Massachusetts Institute of Technology, Cambridge, MA, United States; ³Computer Science and AI Lab (CSAIL), Massachusetts Institute of Technology, Cambridge, MA, United States

High-resolution 7T fMRI together with laminar surface-based analysis is utilized to assess the ability of laminar-specific comparisons to differentiate resting state correlations stemming from direct cortical-to-cortical connections from correlations arising from common-source input. We show that the Layer II/III outputs of human V1 are more highly correlated to the Layer IV inputs of area MT than to other layers, while each layer of V1 is maximally correlated with the same layer in the V1 of the opposite hemisphere. This suggests that laminar analysis of functional connectivity can help identify correlations that may be attributable to indirect connections through common inputs.