Phase Constancy in a Ladder Model of Neural Dynamics

Abstract

This paper presents a novel concept of modeling biological systems by means of preserving the natural rules governing the system's dynamics, i.e., their intrinsic fractal (recurrent) structure. The purpose of this paper is to illustrate the capability of recurrent ladder networks to capture the intrinsic recurrent anatomy of neural networks and to provide a dynamic model which shows typical neuronal phenomena, such as the phase constancy. As an illustrating example, the simplified model for a neural network consisting of motor neurons is used in simulation of a recurrent ladder network. Starting from a generalized approach, it is shown that, in the steady state, the result converges to a constant-phase behavior. The outcome of this paper indicates that the proposed model is a suitable tool for specific neural models in various neuroscience applications, being able to capture their fractal structure and the corresponding fractal dynamic behavior. A link to the dynamics of EEG activity is suggested. By studying specific neural populations by means of the ladder network model presented in this paper, one might be able to understand the changes observed in the EEG with normal aging or with neurodegenerative disorders.