Decomposition of EMG patterns as combinations of time-varying muscle synergies

A key issue in the study of the neural control of movement is understanding how the CNS coordinates the large number degrees of freedom of the musculoskeletal system to achieve a variety of behavioral goals. The organization of muscle synergies, i.e. groups of muscles controlled as units, might simplify this problem by reducing the dimensionality of the control space. We propose a model for the generation of muscle patterns as linear combinations of synergies with a specific spatiotemporal structure. We introduce an algorithm to extract multiple instances of time-varying muscle synergies from EMG patterns of arbitrary length. Simulation shows that the algorithm is capable of recovering a set of synergies from the data constructed by their combinations. We use this algorithm to decompose the EMG patterns recorded during swimming in an intact, unrestrained frog as combinations of three synergies. The recruitment of different synergies over time and episodes appears to capture the characteristic of the corresponding behavior. These results suggest that this approach can provide new insight into the mechanism of biological motor control.