The Neural Mechanism Research On Decorrelation Of Neural Response By Acetylcholine

Biological neural network system controls complex behaviors of the organism such as respiration, sports and so on. Organism processes information mainly by relying on the biological neural network system, so the neural computation mechanism research of the neural network is a very important research field. With the further research on neural computation mechanism, the researchers recently found that the decorrelation of neural response can strengthen the ability of information transmission between neurons, and acetylcholine and attention can also strengthen the ability of information transmission by the decorrelation of neural response. However, so far there are still many unknown places about the underlying neural computation mechanisms of decorrelation of neural response.Previous research results demonstrated that acetylcholine input in the visual area of cerebral cortex can facilitate the expression of somatostain – expression inhibitory neurons, which then inhibit the paralbumin – expressing inhibitory neurons. As the Pv expression became weaker, it weakened Pv inhibitory on pyramidal neurons. This thesis based on the neural physiological mechanism and built a kind of neural network computing model on decorrelation of neural response by acetylcholine. Through the theoretical analysis and simulation results of the model without Som neuron, we found that the model has strong and weak winner-take-all work areas in the neurons synaptic connection parameter space. Through the theoretical analysis and simulation results of the model with Som neuron, we found when the Som gain increases in the neurons synaptic connection parameter space, the correlation of the two excitatory neurons will significantly decline. The experimental results showed that we have built a neural computation model which matched with the physiological phenomenon of decorrelation of neural response by acetylcholine.This thesis built a model on the basis of the neural computing mechanism of acetylcholine facilitating the decorrelation of neural response. This model not only achieved satisfactory results, but also made a contribution to the research of acetylcholine facilitating the decorrelation of neural response mechanisms.