Motor synergy generalization framework for new targets in multi-planar and multi-directional reaching task

Abstract

Humans can rapidly adapt to new situations, even though they have redundant degrees of freedom (d.f.). Previous studies in neuroscience revealed that human movements could be accounted for by low-dimensional control signals, known as motor synergies . Many studies have suggested that humans use the same repertories of motor synergies among similar tasks. However, it has not yet been confirmed whether the combinations of motor synergy repertories can be re-used for new targets in a systematic way. Here we show that the combination of motor synergies can be generalized to new targets that each repertory cannot handle. We use the multi-directional reaching task as an example. We first trained multiple policies with limited ranges of targets by reinforcement learning and extracted sets of motor synergies. Finally, we optimized the activation patterns of sets of motor synergies and demonstrated that combined motor synergy repertories were able to reach new targets that were not achieved with either original policies or single repertories of motor synergies. We believe this is the first study that has succeeded in motor synergy generalization for new targets in new planes, using a full 7-d.f. arm model, which is a realistic mechanical environment for general reaching tasks.

Synergy quality assessment of muscle modules for determining learning performance using a realistic musculoskeletal model

“…It suggests that minor difference in biomechanical contexts can easily make modularity using context-specific synergies less effective. Consistent with the results of the previous study (Kutsuzawa and Hayashibe, 2022), the modularity performance largely depends on the biomechanical context from which the synergies are extracted.…”

“…In line with this, researchers started to study what effects the central nervous system (CNS) would have in adopting modularity in motor control rather than full-rank control. The typical aspects regarded as advantages of modularity are its facilitating effects of motor learning or planning (Rückert and d'Avella, 2013 ; Diamond and Holland, 2014 ; Hagio and Kouzaki, 2018 ; Al Borno et al, 2020 ; Chen and Qiao, 2020 ; Berg et al, 2023 ) and partial generalization capability (Rückert and d'Avella, 2013 ; Taïx et al, 2013 ; Al Borno et al, 2020 ; Chen and Qiao, 2020 ; Kutsuzawa and Hayashibe, 2022 ; Berg et al, 2023 ). These aspects will provide evolutionary benefits to organisms because combining a limited number of learned repertoires for realizing a new motor skill is more efficient than an exhaustive search in full-rank motor space for an optimal solution if the generated motor skill is enough (Loeb, 2021 ).…”

Decomposition into dynamic features reveals a conserved temporal structure in hand kinematics

“… 10 , 38 Interestingly, motor cortex dynamics appear to differ for reaching and grasping movements, with variable use of static and spatiotemporal synergies depending on movement type. 39 , 40 Interestingly, the ability to use generalized motor synergies to generate new movements has recently been demonstrated in a computational model of reach-and-grasp movements 41 ; our results demonstrate this phenomenon in kinematic data of functional hand movements.…”

Imitation Learning With Time-Varying Synergy for Compact Representation of Spatiotemporal Structures

“…The same synergies are often observed in different movements [32], [33], suggesting that synergies can be reused in new situations. Furthermore, studies reported that even a few synergies can represent diverse movements [30], [33], [34]. Therefore, synergies are expected to improve the generalization ability of NNs for generating new movements.…”