Fig. 4.

RSFC-Boundary Mapping parcellation exhibits a high degree of correspondence with areas defined by task-evoked activity. Task-evoked activity was derived from meta-analyses of multiple studies to highlight locations exhibiting sensitivity to performance of certain tasks (e.g., reading) or certain signal types (e.g., error-related activity). The 120-subject RSFC-Boundary Mapping parcellation was thresholded (edge probability >0.15) to reveal locations exhibiting a high likelihood of being a border between areas. Many area locations defined by task-evoked activity are surrounded by RSFC-borders (e.g., the cluster of activity in the ventral medial prefrontal cortex in the task-induced deactivation meta-analytic map). In other locations RSFC-borders separate what appear to be distinct clusters of task-evoked activity, suggesting the existence of distinct areas (e.g., a cluster of activity in the inferior parietal lobule is separated from a cluster of activity in the angular gyrus in the episodic memory meta-analysis map). Parcellations are overlaid on inflated cortical surfaces; some surfaces have been tilted to facilitate viewing (i.e., the lateral surface of the right hemisphere in the motor response (button pushing) comparison and the lateral surface of the left hemisphere in the error-related activity comparison).