Implementation of the NucleoCortical pathways inside a Spiking Neural Network model of Cerebellar Nuclei.

The cerebellum is involved in a large number of neural processes, especially in motor control and motor learning. From a functional point of view, the Deep Cerebellar Nuclei (DCN) constitute the core of the cerebellar processing, as they exploit the spatiotemporal filtering action provided by the cerebellar cortex to generate motor corrections. To investigate the contribution of the cerebellar nuclei in motor coordination, in-silico models of the DCN are often exploited in virtual simulations of complex sensorimotor tasks. Since the outcomes of this analysis may depend on the accuracy of the DCN model reconstruction, we here advanced existing spiking neural network models of the cerebellar nuclei. Specifically, the nucle-ocortical pathways, which provide feedback signals back to the cerebellar cortex, have been implemented. This reconstruction required the development of a new neural population in the DCN (Glycinergic-Inactive neurons). Using the neural simulator NEST, Glycinergic-Inactive neurons have been modeled as Extended-Generalized Leaky Integrate and Fire models, while nucleocortical pathways have been implemented considering their anatomical organization and functional behaviour. By exploiting these improvements, the DCN models may be used for more accurate simulations of complex cerebellum-driven tasks, investigating the signal integration process between sensorimotor inputs and cerebellar output internal feedback.

Automated summary

The cerebellum plays a crucial role in motor control and learning, with the Deep Cerebellar Nuclei serving as the core of processing. By advancing spiking neural network models of the cerebellar nuclei, including nucleocortical pathways, more accurate simulations of cerebellum-driven tasks can be achieved to investigate signal integration between sensorimotor inputs and internal feedback.