Encoding of movements by Purkinje cells of the cerebellum

Execution of accurate eye movements depends critically on the cerebellum, as evidenced by the deficits observed in patients and lesion studies. These observations have suggested that Purkinje cells (P-cells) predict the state of the eye (e.g. its position or velocity) during a saccade. Yet, this encoding has remained a long-standing puzzle because P-cell firing shows little modulation with respect to saccade velocity or direction. How can the P-cells be involved in predicting or controlling the dynamics of the eye if their discharge is poorly modulated with changes in this state? Here, we re-analyzed data from P-cells in the oculomotor vermis of 5 behaving monkeys during saccades and estimated the inhibition that these cells produced at the caudal fastigial nucleus (cFN). We uncovered a time course of synaptic input at cFN that precisely predicted the real-time velocity of the eye. When we aligned the simple spikes of each P-cell to a coordinate system that depended on that cell's complex spike field (CS), the result unmasked a pattern of inhibition at cFN that encoded both saccade velocity and direction via a multiplicative coding, i.e., a gain-field. Therefore, our results suggested two new principles regarding function and anatomy of the cerebellum: the encoding of state of the eye does not occur in the firing rates of individual P-cells, but via synchronized inputs to cFN; the anatomical projections of P-cells to cFN neurons is not random, but organized by the CS fields of the P-cells.

Speaker: Reza Shadmehr, John Hopkins