Abstract
Functional connectivity Magnetic Resonance Imaging (fcMRI) has assumed a central role in neuroimaging efforts to understand the changes underlying brain disorders. Current models of the spatial and temporal structure of fcMRI based connectivity contain strong a priori assumptions. We report that low temporal frequency fMRI signal synchrony within the local (3 mm radius) neighborhood of a location on the cortical surface strongly predicts the scale of its global functional connectivity. This relationship is tested vertex-wise across the cortex using Spearman’s rank order correlation on an individual subject basis. Furthermore, this relationship is shown to be dynamically preserved across repeated within session scans. These results provide a model free data driven method to visualize and quantitatively analyze patterns of connectivity at the imaging voxel resolution across the entire cortex on an individual subject basis. The procedure thus provides a tool to check directly the validity of spatial and temporal prior assumptions incorporated in the analysis of fcMRI data.
