Control of the Tongue by the Primate Cerebellum

About This Grant

Project Summary/Abstract The tongue mediates behaviors fundamental to survival, such as vocal communication and food manipulation. Cerebellar dysfunction disrupts normal tongue muscle activation patterns, leading to significant clinical pathologies in swallowing and speech that severely impact quality of life and can be fatal. Despite this clinical significance, the neural mechanisms of cerebellar tongue control in non-human primates remain largely unexplored, with no standardized behavioral tasks for measuring lingual motor learning in these models. Pathologically, cerebellar dysfunction is characterized by deficits in Purkinje cells (P-cells), the sole output neurons of the cerebellar cortex. P-cells produce two types of action potential: the simple spike, which is their primary output, and the complex spike, which is triggered by climbing fiber inputs to the P-cells. Complex spikes are often in response to error, and are always followed by complete suppression of the simple spikes. Though we know that P-cells sculpt the computational output of the cerebellar cortex, the way in which they enact population-level control of tongue movements remains mysterious. Here, I propose to establish the marmoset (Callithrix jacchus) as a model for cerebellar tongue control. Marmosets present several key advantages: 1) their long and protrusive tongue is easy to track with computer vision, 2) their cerebellum demonstrates developmental alignment with non-human primates while remaining accessible with recording instruments designed for mice, and 3) their survival depends on harvesting tree sap via skilled licks, suggesting robust cerebellar involvement in control of targeted tongue movements. Our recent work in marmosets has yielded a breakthrough: when a P-cell is briefly but completely suppressed by a complex spike, the result is a small displacement which serves as a proxy for that cell's causal contribution to behavioral output. I hypothesize that grouping P-cells by their downstream effects on licking will reveal population-level computations that the cerebellum performs to aid in control of the tongue. However, testing this hypothesis requires mapping downstream effects of P-cell suppression across cerebellar lingual regions, as it is a prerequisite for functional classification of P-cells and subsequent analysis of population-level dynamics during targeted licking. Here, my aim is to first develop an experimental setup that will allow for a wide range of errors during licking, thus driving complex spikes. Next, I will record from various cerebellar lingual regions and use spike-triggered averaging to measure the change in tongue kinematics caused by the complex spike-induced suppression of each P-cell. To test specificity of the effects due to suppression, and not error, I will consider movements in which there was no error, but a suppression occurred. This will test whether downstream effects of P-cells are regionally organized in the cerebellum. Overall, my proposed work will bridge the gap in our understanding of cerebellar control over tongue movements and will aid lingual neuroscience by developing a new non-human primate model in marmosets.