Does the cerebellum predict the consequences of movement before sensory feedback arrives?

Details

The motor system may use internal predictive models of the motor apparatus to achieve better control than negative feedback alone would allow. Several theories propose that the cerebellum forms these predictive representations. This article reviews those theories and attempts to unify them using an engineering control model called a Smith Predictor.

The authors propose that the cerebellum forms two types of internal model. The first is a forward predictive model of the motor apparatus, such as limbs and muscles, that rapidly predicts the sensory consequences of each movement. The second models time delays in the control loop, including receptor and effector delays, axonal conductances, and cognitive processing delays. This second model delays a copy of the rapid prediction so that it can be compared with actual sensory feedback from the movement at the matching time. In this proposed account, the comparison serves both to correct performance errors and to provide a training signal for learning the first model.

The review discusses evidence that the cerebellum could form both models and proposes that it may contain at least two separate Smith Predictors. One in the lateral cerebellum would predict movement outcomes in visual, egocentric, or peripersonal coordinates. Another in the intermediate cerebellum would predict consequences in motor coordinates.

The authors also discuss generalizing Smith Predictor theory in light of cerebellar involvement in nonmotor control systems, including autonomic functions and cognition.