Mathematical modeling of brain circuitry during cerebellar movement control

Reconstruction of movement control properties of the brain could result in many potential advantages for application in robotics. However, a hampering factor so far has been the lack of knowledge of the structure and function of brain circuitry in vivo during movement control. Much more detailed information has recently become available for the area of the cerebellum that controls arm-hand movements. In addition to previously obtained extensive background knowledge of the overall connectivity of the controlling neuronal network, recent studies have provided detailed characterizations of local microcircuitry connectivity and physiology in vivo. In the present study, we study one component of this neuronal network, the cuneate nucleus, and characterize its mathematical properties using system identification theory. The cuneate nucleus is involved in the processing of the sensory feedback evoked by movements. As a substrate for our work, we use a characterization of incoming and outgoing signals of individual neurons during sensory activation as well as a recently obtained microcircuitry characterization for this structure. We find that system identification is a useful way to find suitable mathematical models that capture the properties and transformation capabilities of the neuronal microcircuitry that constitute the cuneate nucleus. Future work will show whether specific aspects of the mathematical properties can be ascribed to a specific microcircuitry and/or neuronal property.