DIVERGENT & CONVERGENT CEREBELLAR AND BASAL GANGLIA INPUTS STRUCTURE MOTOR THALAMIC ACTIVITY DURING SKILLED REACHING

Skilled reaching towards a target requires the generation of movements toward predicted future states while continuously evaluating ongoing motor output. Neural substrates supporting these computations include cerebellum, basal ganglia, and motor cortex, which interact via motor thalamus. Classical circuit frameworks treat cerebellar and basal ganglia influences on cortex as parallel and largely segregated, with the cerebellum associated with predictive control of movement kinematics and the basal ganglia with movement initiation and gating. However, given the anatomical convergence of cerebellar and basal ganglia outputs in motor thalamus, we hypothesise that thalamic circuits provide a substrate for integrating predictive and evaluative control signals during skilled reaching.
To test this hypothesis, we recorded large-scale single-unit activity from motor thalamus in mice performing a skilled forelimb reaching task. High-speed behavioural tracking was combined with Neuropixels 2.0 recordings to relate thalamic activity to reach initiation and execution. To identify thalamic neurons defined by their input pathways, we used dual presynaptic optogenetic stimulation of cerebellar nuclei and substantia nigra pars reticulata while recording thalamic spiking activity. We found that cerebellar- and basal ganglia-recipient thalamic neurons exhibit distinct task-related activity profiles during reaching. Neurons receiving convergent input exhibit unique activation features during the task, suggesting integrative thalamic processing. These results indicate that input-defined thalamic populations jointly shape cortical motor activity during skilled movement.