Evaluating the Stability of Muscle Synergies during Circle Tracking

Previous studies mostly applied the nonnegative matrix factorization algorithm (NNMF) upon the reaching tasks or some relatively stable activities to extract the muscle synergies aiming for observing the coordination of muscles in humans. However, few studies have used this algorithm in the tracking task since its high complexity and unpredictability, which means that the muscle synergies become more unstable and are hard to embody the neural mechanisms behind the human motion. Therefore, we implemented NNMF on tracking tasks and calculated two synergy indices, including the synergy stability index (SSI) and synergy coordination index (SCI). $\mathrm {S}\mathrm {S}\mathrm {I}_{\mathrm {w}}$ and $\mathrm {S}\mathrm {S}\mathrm {I}_{\mathrm {C}}$ respectively quantify the similarity between synergies and activation coefficients, and SCI indicates the size of the synergy space. In our results, $\mathrm {S}\mathrm {S}\mathrm {I}_{\mathrm {w}}$ was about 0.8, $\mathrm {S}\mathrm {S}\mathrm {I}_{\mathrm {C}}$ was nearly 0.3 and SCI was 0.6. The value of these indices indicated that humans tended to keep synergies stable and control them more flexibly by changing the activation time and amplitudes when undertaking changeable and complex tasks.