the timecourse of a seed ROI (a voxel or group of voxels) with all voxels in the brain, revealing the spatial locations where rs-fcMRI activity is similar to the seed’s. For example, seeds in visual cortex tend to result in maps that highlight occipital cortex (Lowe et al., 1998), demonstrating high correlations in rs-fcMRI signal throughout the visual system. Component approaches (e.g. independent or principal component analysis (ICA, PCA)) employ data reduction techniques to partition voxels into components that share variance in their timecourses. Similar to the visual seed just mentioned, component analyses of rs-fcMRI data routinely detect a component of voxels in occipital cortex that share considerable portions of their variance (Damoiseaux et al., 2006). Neither a seed map, nor a component, nor a constellation of coactive regions during a task necessarily constitutes a network in the broader world of networks. When discussing data, we therefore refer to these descriptions by more neutral terms like “resting relationships” or “correlated brain activity” or the like, and generally reserve “network” for descriptions congruent with the broader graph theoretic sense. In a few specific cases, when referring to commonly understood “networks” of brain regions with particular properties, we continue to use the
