A Multifractal Dynamical Model of Human Gait

TL;DR

This paper introduces a multifractal dynamical model of human gait that captures the complex, rhythmic, and multifractal properties of stride intervals influenced by biological and stress factors.

Contribution

The paper proposes the super central pattern generator model, a novel approach to simulate and understand the multifractal properties of human walking behavior.

Findings

Gait exhibits fractal and multifractal properties influenced by pacing and health conditions.

The model reproduces known properties of stride interval fluctuations.

Multifractal characteristics become more pronounced under certain conditions.

Abstract

Summary: Walking is regulated through the motorcontrol system (MCS). The MCS consists of a network of neurons from the central nervous system (CNS) and the intraspinal nervous system (INS), which is capable of producing a syncopated output. The coupling of the latter two systems produces a complex stride interval time series that is characterized by fractal and multifractal properties that depend upon several biological and stress constraints. It has been shown that: (i) the gait phenomenon is essentially a rhythmic cycle that obeys particular phase symmetries in the synchronized movement of the limbs; (ii) the fractal and multifractal nature of the stride interval fluctuations become slightly more pronounced under faster or slower paced frequencies relative to the normal paced frequency of a subject; (iii) the randomness of the fluctuations increases if subjects are asked to synchronize their gait with the frequency of a metronome or if the subjects are elderly or suffering from neurodegenerative disease. Here we present a new model, called the super central pattern generator, able to reproduce these known properties of walking and discuss the physiological and psychological interpretations of the model parameters.