Information Transmission And Energy Efficiency In A Neuron Model

Through the perception of environmental changes,the neurons in the brain transmit information in the form of action potentials to other neurons after being stimulated by external signals.A nervous system constantly receives information,analyzes information,makes decisions and controls the entire body directly or indirectly.According to the mechanism of information transmission,the information carried by action potentials can be quantified to judge the transmitted information capacity and encoding efficiency by using the concept of “entropy” about the information theory.In addition,there exists energy consumption in the process of transmission and expression of neuronal firing sequences.In order to deal with the complex system environment,on the one hand,the effective information capacity transmitted by the system must be maximized,and on the other hand,the energy consumption of the system should be minimized during the transmission process.It is a key principle for a neural system to obey during the evolution process.In this paper,two aspects of the research work are carried out.One hand,the dynamical properties,the effective information capacity and the encoding efficiency are studied in a single neuron under different firing and coding modes based on twodimensional mapping Courbage-Nekorkin-Vdovin(CNV)model.On the other hand,dynamical response,information transition capacity and energy efficiency are studied under the inhibitory chemical autaptic feedback based on the autaptic HH model.Based on CNV model,the firing process of a neuron is numerically simulated.In the temporal coding mode,the information capacity and coding efficiency are quantified in transmission process using the time bin discretization method on the random initial condition of 300 neurons.Firstly,the bursting and tonic firing modes are simulated by the parameter setting and the trajectory of the phase plane is analyzed.Secondly,in the bursting mode,the total entropy rate,the noise entropy rate,the information entropy rate and the coding efficiency in temporal coding mode and the information entropy rate in rate coding mode are calculated without regard to the noise influence.Then,also in the bursting mode and the rate coding mode,the two information entropy rates are compared regarding the noise influence or not.Finally,in tonic mode,the total entropy rates,the noise entropy rates,the information entropy rates and the encoding efficiencies are compared in different coding modes between the CNV model and the stochastic HH model.The experimental results show that the two-dimensional CNV mapping model can simulate rich firing modes and dynamical characteristics in a single neuron with high computational efficiency and feasibility without considering the ion channel voltage gating mechanism about action potentials.In Bursting firing mode,the effective information transmitted by the system increases with the increase of the external stimulus current,and the coding efficiency is slightly different under different coding modes and noise environment.In Tonic mode,the CNV model can simulate the transmission process of the stochastic HH model.Therefore,CNV can express rich neuron firing behaviors in a relatively simple model,and has higher information transmission capacity and coding ability.Autapses are a class of special synapses of neurons,whose axons are not connected to dendrites of other neurons but attached to their own cell bodies.Output signal of a neuron is fed back to itself so that the neuronal firing behaviors are affected.Autapses can adjust the accuracy of a firing potentials,achieve local self-control,and regulate the synchronization of a neuronal system.Firstly,we studied the dynamical effects of excitatory and inhibitory chemical autapse based on the HH neuron model with autaptic feedback.Secondly,we studied the information entropy rate and the coding efficiency of neurons on the condition of inhibitory chemical autapse at different coupling strengths and different delay times in a noisy environment.Finally,we studied the energy efficiency of inhibitory chemical synapse on the transmission of information under the same parameters setting.The results show that the transmitted effective information is the most,information encoding capability is the highest,and the energy efficiency is optimal at a given autaptic coupling strength,when the delay time is half of the input signal period.In addition,at the same delay time,with the increasing of inhibitory strength of autapse,this maximization is more and more obvious.Therefore,inhibitory chemical autapse tends to regulate the neuron firing regularity.On the action of autaptic feedback,neurons have higher information transmission capacity and energy efficiency on the process of information processing.According to the above research work,a summary and a prospect for the future work is made about the paper.