Research On Neural Information Encoding Of Spiking Neural Network Based On Synaptic Plasticity Under External Stimulation

How brain codes external stimulation is an important research topic in the field of neuroscience.In essence,the realizations of brain's various advanced functions lie in the powerful information processing ability of nervous system.And the nature of information processing is information encoding.The study of neural information encoding is fundamental to exploring the mechanism of brain information processing.In this study,a scale of synaptic plasticity spiking neural network was constructed based on the simplification,abstraction and simulation of the brain neural network;the information encoding feature of the network under different external stimulation was investigated by using the average frequency coding method and the inter-spike interval(ISI)coding method;the information encoding features of the network were comparatively analyzed among different stimulations.This study aims at exploring the dynamic behavior specificity of the network under different external stimulating and providing experimental basis for neural network information processing mechanism research.Main work is as follows:(1)Separately used white Gaussian noise and square wave impulse noise to stimulate the network in the form of impressed current.And the characteristics of information encoding for the network were studied.The experimental results show that the pulses of the neurons in the network become less uniform and pulse synchronization between neurons are reduced under the white Gaussian noise stimulation;under the square wave impulse stimulation,the pulses of neurons in the network are more intensive in the moment of wave impulse stimulation compare to the moment that no wave impulse stimulates.And after the wave impulse stimulation,the pulses of the neurons in the network recover to the state that before the stimulation.(2)The characteristics of information encoding for the network were separately studied under the ac electric field stimulation and ac magnetic field stimulation.The experimental results show that the pulses of neurons in the network were become intensive in some period of the whole simulation time under the ac electric field simulation or ac magnetic field stimulation.And the pulses of the whole network show an increase trend.(3)The characteristics of information encoding for the network under neuron damage were studied.The experimental results show that in the case of some neurons damaged,the pulses of the undamaged neurons in the network are become sparse in some period of the whole simulation time.And the pulses of the whole network show a decrease trend.(4)The information encoding features of the network under different external stimulation were comparatively analyzed.The results show that there are different encoding features for the network under different kinds of stimulations.The network shows dynamic behavior specificity under different kinds of stimulations.