Phase Synchronization Motion and Neural Coding in Dynamic Transmission of Neural Information
Rubin Wang,Zhikang Zhang,Jingyi Qu,Jianting Cao
DOI: https://doi.org/10.1109/tnn.2011.2119377
2011-01-01
IEEE Transactions on Neural Networks
Abstract:In order to explore the dynamic characteristics of neural coding in the transmission of neural information in the brain, a model of neural network consisting of three neuronal populations is proposed in this paper using the theory of stochastic phase dynamics. Based on the model established, the neural phase synchronization motion and neural coding under spontaneous activity and stimulation are examined, for the case of varying network structure. Our analysis shows that, under the condition of spontaneous activity, the characteristics of phase neural coding are unrelated to the number of neurons participated in neural firing within the neuronal populations. The result of numerical simulation supports the existence of sparse coding within the brain, and verifies the crucial importance of the magnitudes of the coupling coefficients in neural information processing as well as the completely different information processing capability of neural information transmission in both serial and parallel couplings. The result also testifies that under external stimulation, the bigger the number of neurons in a neuronal population, the more the stimulation influences the phase synchronization motion and neural coding evolution in other neuronal populations. We verify numerically the experimental result in neurobiology that the reduction of the coupling coefficient between neuronal populations implies the enhancement of lateral inhibition function in neural networks, with the enhancement equivalent to depressing neuronal excitability threshold. Thus, the neuronal populations tend to have a stronger reaction under the same stimulation, and more neurons get excited, leading to more neurons participating in neural coding and phase synchronization motion.