A Study on Patterns of Neural Activity Generation from A Bio-realistic Cerebellum Neural Network

摘要

The biological neural network signal flow is a complex process and difficult to interpret. The purpose of this study is to build a model of a simplified cerebellum neural network and investigate its working behavior in generating neuronal patterns. The model is constructed by referring to the neural anatomic structure of the mammal cerebellum and then simulated using Matlab Simulink software. The artificial neural network dimension is much smaller than the size of the real cerebellar neural network in mammal due to hardware simulation resource limitation. Thus, the utilization of extensivescale spiking neural networks is limited due to the grown of computational costs associated with propagating bio-realistic neural models. The parameters correlated with the neuron function also affect the biological plausibility of the spiking neural network. At the conclusion, we present the pattern generation of a cerebellum- like neural network and briefly explained why such patterns are generated.