Year: 2010
Type: article
Abstract: Explaining motor control in humans and animals is one of the greatest challenges facing cognitive science, and has significant implications for engineering.Understanding the mechanisms underlying the control of a complex, nonlinear, high degree-of-freedom system like the body could serve as the basis for more sophisticated limbed robots with very advanced capabilities.One widely held theory about motor control is that it happens through the selective combination of pre-existing motor primitives called synergies, encoded as patterns of activity in the spinal cord and higher brain regions.This mechanism shows promise in addressing the problem of excess degrees-of-freedom that is the most serious difficulty faced by any motor controller.Recent experiments have provided significant support for this hypothesis, and invariant motor synergies have been extracted from electromyogram data in many animals, including humans.Scaled and time-shifted combinations of these synergies can replicate a variety of movements in these animals.However, the neurobiological basis of these synergies remains obscure.In this thesis, we propose a neural model for the synergy-based motor control of a 2 jointed arm consisting of 4 single jointed and 2 double jointed muscles.The neural model proposes that synergies are encoded as metastable attractors in modular recurrent networks in the motor system.These attractors are triggered with different amplitudes and delays by neurally plausible mechanisms, and integrated through neural networks in the spinal cord.As a result, the simulated arm can produce a large repertoire of movements using only a few synergies.The synergies generated by the model are compared with those observed experimentally, and richness of this system is explored through several simulations.
Author: Mithun C. Perdoor
Institution: University of Cincinnati
Language: English
Topic: Motor Control and Adaptation
Subfield: Cognitive Neuroscience
Field: Neuroscience
Domain: Life Sciences
Open Access status: green
OpenAlex ID: W2272528730