Shared and Specific Synchronous Muscle Synergies Arisen from Optimal Feedback Control Theory

In this study the properties of muscle synergies, arising from optimal feedback control are investigated. Three different tasks namely reaching, via-point, and hitting are performed using optimization of corresponding cost functions. Then by applying non-negative matrix factorization method to a dataset of muscle tensions, synchronous muscle synergies are obtained. In this way, different muscle patterns can be generated by linear combination of these basis vectors with non-negative time-varying scaling coefficients. According to our simulations some of obtained muscle synergies are shared between tasks and some of them are specific for one task. This finding is also in agreement with the properties of extracted muscle synergies from experimental approach.