Increased spinal excitation causes decreased locomotor complexity

De Graaf, Myriam Lauren; Mochizuki, Luis; Wagner, Heiko; Le Mouel, Charlotte

Abstract

Animals display rich and coordinated motor patterns during walking and running, that are generated and controlled within the central nervous system. Previous computational and experimental results suggest that the balance between excitation and inhibition in neural circuits may be critical for generating such structured motor patterns. In this paper, we explore the influence of this balance on the ability of a reservoir computing artificial neural network to learn human locomotor patterns, using mean-field theory analysis and simulations. We varied the numbers of neurons, connection percentages and connection strengths of excitatory and inhibitory neuron populations, and introduced the anatomical imbalance that quantifies their combined overall effect. Our results indicate that network dynamics and performance depend critically on the anatomical imbalance in the network. Inhibition-dominated networks work well, displaying balanced input to the neurons and good firing rate variability. Excitation-dominated networks, however, lead to saturated firing rates, thereby reducing the effective dimensionality of the networks and leading to simplified motor output patterns. This suggests that motor pattern generation may be robust to increased inhibition but not increased excitation in neural networks.