Chapter 13 - Cerebellar Learning

Abstract

This chapter focuses on the issue of cerebellar learning. It describes the evidence of cerebellar involvement in a number of forms of motor learning and adaptation and reviews evidence on the mechanisms underlying cerebellar learning, based on the unique advantages of eyelid conditioning, The evidence described in the chapter suggests that cerebellar learning is mediated by plasticity in the cerebellar cortex and in its downstream targets in the deep cerebellar nuclei. Bidirectional plasticity in the cerebellar cortex and deep cerebellar nucleus are controlled by climbing fibers and Purkinje Cells, respectively. The network dynamics of the cerebellar cortex creates a temporal code in the form of granule cells firing with different temporal delays during a CS, which can help shape the timing of cerebellar output vi2i gr-Pkj plasticity. The behavioral properties of eyelid conditioning to analyze the contribution of this learning in cerebellar information processing are also examined. The temporal properties of eyelid conditioning suggests that the cerebellar learning is involved in making temporally specific feed-forward predictions, a computation that is likely across the cerebellum, given its uniform circuitry. Also, the cerebellum learns temporally specific feed-forward predictions that are used to improve the operation of any neural process occurring on a relatively short time scale.