Robust muscle synergies for postural control

• Main Author: Torres-Oviedo, Gelsy
• Published: Georgia Institute of Technology 2008
• Subjects: EMG; Motor control; Balance; Human mechanics; Kinesiology; Nervous system; Biomechanics; Posture; Muscles; Synergetics
• Online Access: http://hdl.handle.net/1853/22691

The musculoskeletal structure of the human and animal body provides multiple solutions for performing any single motor behavior. The long-term goal of the work presented here is to determine the neuromechanical strategies used by the nervous system to appropriately coordinate muscles in order to achieve the performance of daily motor tasks. The overall hypothesis is that the nervous system simplifies muscle coordination by the flexible activation of muscle synergies, defined as a group of muscles activated as a unit, that perform task-level biomechanical functions. To test this hypothesis we investigated whether muscle synergies can be robustly used as building blocks for constructing the spatiotemporal muscle coordination patterns in human and feline postural control under a variety of biomechanical contexts. We demonstrated the generality and robustness of muscle synergies as a simplification strategy for both human and animal postural control. A few robust muscle synergies were able to reproduce the spatial and temporal variability in human and cat postural responses, regardless of stance configuration and perturbation type. In addition inter-trial variability in human postural responses was also accounted for by these muscle synergies. Finally, the activation of each muscle synergy in cat produced a specific stabilizing force vector, suggesting that muscle synergies control task-level variables. The identified muscle synergies may represent general modules of motor output underlying muscle coordination in posture that can be activated in different sensory contexts to achieve different postural goals. Therefore muscle synergies represents a simplifying mechanism for muscle coordination in natural behaviors not only because it is a strategy for reducing the number of variables to be controlled, but because it represents a mechanism for simply controlling multi-segmental task-level variables.