Cerebellar granule cells encode the expectation of reward

Abstract

The human brain contains ~60 billion cerebellar granule cells1, which outnumber all other neurons combined. Classical theories posit that a large, diverse population of granule cells allows for highly detailed representations of sensorimotor context, enabling downstream Purkinje cells to sense fine contextual changes2–6. Although evidence suggests a role for cerebellum in cognition7–10, granule cells are known to encode only sensory11–13 and motor14 context. Using two-photon calcium imaging in behaving mice, here we show that granule cells convey information about the expectation of reward. Mice initiated voluntary forelimb movements for delayed water reward. Some granule cells responded preferentially to reward or reward omission, whereas others selectively encoded reward anticipation. Reward responses were not restricted to forelimb movement, as a Pavlovian task evoked similar responses. Compared to predictable rewards, unexpected rewards elicited markedly different granule cell activity despite identical stimuli and licking responses. In both tasks, reward signals were widespread throughout multiple cerebellar lobules. Tracking the same granule cells over several days of learning revealed that cells with reward-anticipating responses emerged from those that responded at the start of learning to reward delivery, whereas reward omission responses grew stronger as learning progressed. The discovery of predictive, non-sensorimotor encoding in granule cells is a major departure from current understanding of these neurons and dramatically enriches contextual information available to postsynaptic Purkinje cells, with important implications for cognitive processing in the cerebellum.

The Cerebellum in Social Cognition

“…As to the mechanisms, because of the similar cytoarchitecture across the cerebellum, it has been repeatedly suggested its operations for motor control and coordination may also apply to cognition (Wolpert et al, 1998;Ito, 2008;Sokolov et al, 2017). These operations are believed to involve outcome prediction based on forward models and signaling deviations from these outcomes (prediction errors) to the cerebral cortex, a hypothesis supported by recent electrophysiology data on non-sensorimotor expectations in rodents (Wagner et al, 2017) and concepts of ASD as a prediction deficit with prominent cerebellar pathology (Sinha et al, 2014). As anticipation, adaptation and learning appear indispensable for successful social behavior, extending these core cerebellar functional roles from sensorimotor models to those of social perception and behavior would seem reasonable, although several issues remain to be carefully considered and explored (Sokolov et al, 2017).…”

“…Insights on cellular mechanisms are indispensable for valid models of large-scale networks and pathophysiology, and vice versa. In a cerebellum previously conceptualized as rather uniform, rodent electrophysiology has already unveiled different rules for synaptic plasticity (Wadiche and Jahr, 2005;Zhou et al, 2014;Suvrathan et al, 2016), and recently demonstrated some granule cells encode non-sensorimotor predictions and their unexpected violations (Wagner et al, 2017). These data offered novel perspectives on how the cerebellum may be equipped to contribute to diverse cognitive processes, but assessing higher cognition in animals and particularly rodents with a truly translational potential poses a significant challenge.…”

Dopamine D1 Receptor–Positive Neurons in the Lateral Nucleus of the Cerebellum Contribute to Cognitive Behavior

“…In primates, there is robust distinction between Dopamine tranporter (DAT)-positive and Dopamine-β-Hydroxylase-positive axons in different parts of the cerebellar cortex(47). Retrograde mapping with Cav2-Cre in a tomato reporter line failed to reveal any tomato labeling in the VTA of mice(81). In rats, VTA sends glutamatergic projections to LCN, but not dopaminergic projections(13).…”

Participation of the pedunculopontine tegmental nucleus in arousal-demanding functions

“…Among these stimuli, we can also include reward and reinforcement learning information that the basal ganglia could provide to the cerebellum. These aspects deserve future investigations given a recent demonstration that the discharge of granule cells in the cerebellum encode reward expectation [130].…”