Functional gradients of the cerebellum: a fundamental movement-to-thought principle

Abstract

A central principle for understanding the cerebral cortex is that macroscale anatomy reflects a functional hierarchy from primary to transmodal processing. In contrast, the central axis of motor and nonmotor macroscale organization in the cerebellum remains unknown. Here we applied diffusion map embedding to resting-state data from the Human Connectome Project dataset (n=1003), and show for the first time that cerebellar functional regions follow a gradual organization which progresses from primary (motor) to transmodal (DMN, task-unfocused) regions. A secondary axis extends from task-unfocused to task-focused processing. Further, these two principal gradients reveal functional properties of the well-established cerebellar double motor representation, and its relationship with the recently described triple nonmotor representation. These interpretations are further supported by data-driven clustering and cerebello-cerebral functional connectivity analyses. Importantly, these descriptions remain observable at the individual subject level. These findings, from an exceptionally large and high-quality dataset, provide new and fundamental insights into the functional organization of the human cerebellum, unmask new testable hypotheses for future studies, and yield an unprecedented tool for the topographical, macroscale interpretation of cerebellar findings.