Cerebellar implementation of movement sequences through feedback

By A. Khilkevich, Juan Zambrano, Molly-Marie Richards, M. Mauk. 2018

35 citations recorded by the source corpus.

Most movements are not unitary, but are comprised of sequences. Although patients with cerebellar pathology display severe deficits in the execution and learning of sequences (Doyon et al., 1997; Shin and Ivry, 2003), most of our understanding of cerebellar mechanisms has come from analyses of single component movements. Eyelid conditioning is a cerebellar-mediated behavior that provides the ability to control and restrict inputs to the cerebellum through stimulation of mossy fibers. We utilized this advantage to test directly how the cerebellum can learn a sequence of inter-connected movement components in rabbits. We show that the feedback signals from one component are sufficient to serve as a cue for the next component in the sequence. In vivo recordings from Purkinje cells demonstrated that all components of the sequence were encoded similarly by cerebellar cortex. These results provide a simple yet general framework for how the cerebellum can use simple associate learning processes to chain together a sequence of appropriately timed responses.

Subfields: Cognitive Neuroscience, Neurology

Citations in the corpus

• Re-evaluating circuit mechanisms underlying pattern separation (2019, neutral): “One possibility is that feedback projections may have no role in pattern separation, and may simply extend the timescale over which sequences of movements can be learned (Khilkevich et al., 2018).”