Numerical detection of patterns in CPGs: Gait patterns in insect movement

摘要

The study of the synchronization patterns of small neuron networks that control different biological processes has become a growing discipline. This paper is focused on numerical techniques to detect patterns in Central Pattern Generators (CPGs). We develop two techniques that can be used directly in general CPG models: a lateral phase lag analysis based on a graphic representation of some Poincare maps, and a quasi-Monte Carlo sweeping with an optimized classification of the different patterns.As test example we consider a CPG of insect movement consisting of six coupled neurons following the model developed by Ghigliazza and Holmes (2004) for motoneurons in cockroaches. Previous studies in literature analyzed reduced models of dimension two obtained using phase resetting curves and averaging theory. This approach introduces a lot of simplifications that do not cover numerous non-symmetric patterns. We present an analysis of the complete model developed by combining the two proposed techniques, showing symmetric and non-symmetric patterns coexisting for different parameter values, and how the dominant patterns evolve to the tripod movement. (C) 2019 Elsevier B.V. All rights reserved.