Is the Cerebellum a Smith Predictor?

Abstract

ABSTRACT. The motor system may use intemal predictive models of the motor apparatus to achieve better control than would be possible by negative feedback. Several theories have proposed that the cerebellum may form these predictive representations, In this article, we review these theories and try to unify them by reference to an engineering control model known as a Smith Pre dictor. We suggest that the cerebellum forms two types of internal model. One model is a forward predictive model of the motor apparatus (e.g. , limb and muscle), providing a rapid prediction of the sensory consequences of each movement. The second model is of the time delays in the control loop (due to receptor and effector delays, axona] conductances, and cognitive processing delays). This model delays a copy of the rapid prediction so that it can be compared in temporal register with actual sensory feedback from the movement. The result of this comparison is used both to correct for errors in perf.ormance and as a training signal to learn the first model We discuss evidence that the cerebellum could form both of these models and suggest that the cerebellum may hold at least two separate Smith Predictors. One, in the lateral cerebellum, would predict the movement outcome in visual, egocentric, or peripersonaJ coordinates. Another, in the intermediate cerebellum, would predict the consequences in motor coordinates. Generalization of the Smith Predictor theory is discussed in ]ight of cerebellar involvement in nonmotor control systems, including autonomic functions and cognition

The perception of self-produced sensory stimuli in patients with auditory hallucinations and passivity experiences: evidence for a breakdown in self-monitoring

“…We propose that this inhibition of anterior cingulate and somatosensory cortex activity by self-generated movements could result from the comparison between predicted and actual sensory feedback, which results in less sensory discrepancy when a tactile stimulus is selfproduced relative to when it is externally produced. There is evidence that this comparison might take place in the cerebellum (Ito, 1970 ;Paulin et al 1989 ;Miall et al 1993 ;Leiner et al 1995 ;Wolpert et al 1998), a proposal supported by the functional imaging data (Blakemore et al 1998). We have suggested that this network of brain areas is involved in predicting the consequences of actions in order to distinguish selfproduced and externally produced sensations.…”

Changes in predictive motor control in drop-jumps based on uncertainties in task execution

“…action(s) (Miall, Weir, Wolpert, & Stein, 1993;Wolpert & Miall, 1996). Consequently, to change prediction would mean to modify the internal model.…”

Agency, gait and self-consciousness

“…In the forward model, this happens before any sensory feedback is available. One advantage of the forward model is that movements can be planned and adjusted without the need to wait for sensory feedback which can be delayed by up to 250 ms (Miall et al, 1993), which is one of the reasons why such models have first been proposed for ballistic movements, such as eye movements (Sperry, 1950;von Holst and Mittelstaedt, 1950). Simultaneously, the sensory consequences of the action are predicted using a sensory forward model (lower branch, Fig.…”