Study On The Neural Coding Of Learning And Memory

Learning and memory,as the advanced cognitive function of the brain,has been widely concerned in neuroscience.In fact,the essence of cognition is the process of information processing,and the essence of information processing is neural coding.Neural coding is designed to study the relationship among the stimulus and the response of a single or a population of neurons.The selection of neural coding model plays an important role in the determination of brain function,which helps to analyze the relationship between information integration and the learning and memory function.However,the scope of application of different neural coding patterns are limited,which requires us to find the effective neural coding for the specific field through the comparison of the coding performance.Therefore,the research topic of this paper is how to select appropriate neural encoding and decoding theory to explore the function determination method and mechanism of learning and memory.This paper first summarizes the current situation of neural coding according to the information processing,and explores the effective neural coding modes including receptor coding,rate coding,and temporal coding,population coding,and neural energy coding.And then by using the idea of phase encoding within population coding,a new method to study the synchronization of cerebral cortex is proposed which is called Spike-LFP coherence.The application process of this method is that first get the distribution of all frequency components in LFP through the spectrum analysis;and then through the filter to obtain the frequency band,so as to ensure the effective analysis of LFP and spikes;and then do the coherence analysis to the data after pre-processed.Spike-LFP coherence plays influence on neural encoding,neural competition,neural plasticity,which can further reveal the dynamic mechanism of neural information processing and cognitive function.The function of learning and memory is achieved by interaction among multiple brain regions,therefore encoding method we proposed can be used to identify the relevant information-exchanging regions related to the function,and helps to study the physiological mechanism of the function.After locating the brain regions associated with memory function by phase coding,the mechanism of transformation of memory systems is studied by rate coding.We study the maintenance mode of high frequency firing rate of neurons from the improved C-W model with Ca2+ subsystem-induced bi-stability.TBS and HFS were simulated to act as the initial stimuli of this working memory model,and the influence of two kinds of initial stimulus(period,amplitude,duty ration)on the memory mechanism of the model was evaluated by the control variable method.Study shows that increasing the cycle,amplitude and duty ration would increase the firing rate of network activities and make memory last longer.And it shows that both stimuli(TBS,HFS)could activate the model to transform from working memory to long-term memory.Comparing the critical values of stimuli properties for the two types of stimuli when they could induce long-term memory,it shows that the two types of stimuli were of the same pulse number and duty ration but the amplitude of HFS was smaller leading to a lower stimulation cost.Two LTP induced protocols were adopted to successfully produce long-term memory in working memory model and memory mode is shown by rate coding,which provided theoretical basis for the study of neural dynamics mechanism of long-term memory formation.Based on the above research,the idea of energy coding was used to study the interaction between working memory and long-term memory.During the process of long-term memory formation,the change in energy consumption of TBS and HFS were studied.The energy index can more specifically demonstrate the effect of a change in a specific stimulus element(number of pulses,amplitude,duty ration)on the comprehensive stimulus properties,so that the conclusion got from the control variable method can be shown more intuitively through the energy index.The two types of stimuli could both induce similar effects of long-term memory in the working memory model,but the energy use rate of HFS was higher.The results demonstrated the transformation from working memory to long-term memory in response to the two types of LTP-induced protocols in a quantitative way,which reflects the physiological mechanisms of neural activity in the interaction of the two types of memory system.After studying the memory mechanism,we explore the navigation under the guidance of Learning and memory function.Based on the locating navigation model related to the interaction between place cells and grid cells,the idea of energy coding is introduced to define the differences in energy consumed between place cell and grid cell after learning so as to reflect the mechanism of navigation.It is found that with the increase of the mean of cell distribution,the energy consumed after learning of the two types of cells increase significantly.During the study process,the high-energy ratio of place cell is more than that of grid cell.However,the high-energy ratio of grid cell exceeds that of place cell at last.The changing rate of high-energy ratio of position cell decreases with the increase of high-energy ratio.However,the changing rate of high-energy ratio of grid cells is increasing.The high-energy ratio reaches a certain value balancing the principle of energy minimization and the maximization of the signal transmission efficiency will be beneficial to the navigation.Analyzing the high-energy ratio of learning process of navigation with energy coding can reveals the mechanism of the two types of navigating cells,which helps to promote the efficiency of navigation.