Quantifying the diverse contributions of hierarchical muscle interactions to motor function

Abstract

The muscle synergy concept suggests that the human motor system is organised into functional modules comprised of muscles ‘working together’towards common task-goals. However, recent innovative work has added further nuance to this idea, showing how muscles may also work together towards functionally different and independent task-goals, representing crucial attributes of flexible motor behaviour. Here we probed this proposed functional neural architecture by building upon an established theoretical framework to characterise distinct types of muscle interactions, i.e. functionally similar, complementary and independent, across scales. Through a novel application of the Partial Information Decomposition to large-scale muscle activations, we unveiled complex networks of inter- and intra-muscular interactions with distinct functional roles as well as independent muscle contributions to task performance. We showcased the effectiveness of this approach by extracting hierarchical and functionally diverse motor components that were a) generalisable across participants and tasks and b) predictive of balance performance across trials and of differences in motor variability between young and older adults. In aligning muscle synergy analysis with the forefront of understanding on human movement modularity, our findings suggest the proposed methodology can offer novel biological insights into the neural control of movement and research opportunities towards health and engineering applications.