Structures and characteristics of a neural network model for generating circadian rhythm

Abstract

The biological subject contains a circadian rhythm, which is considered as an autonomous phenomenon originating from its internal structure. the rhythm is known as a biological activity observed in any creature except for prochordata. In most animals, it is found that the circadian rhythm originates from the nervous system. We have discussed the neural network model based on the functions of the neural system, which generate the circadian rhythm. the aim of this paper is to clarify the network structure of the model and to discuss in detail the characteristics through analysis and simulation. It is shown first that the model is composed of two neural oscillators and a neural network with nonlinear transformation characteristics. Then each of the subnetworks is analyzed, describing their basic behaviors. By computer simulation, the entrainment and the change of period are discussed, comparing the result with that of analysis. Those results are compared further with the actual properties of the circadian rhythms of creatures.