Skill-informed Data-driven Haptic Nudges for High-dimensional Human Motor Learning
This work addresses the challenge of enhancing motor learning efficiency in high-dimensional tasks, such as those involving hand exoskeletons, with potential applications in rehabilitation and training, though it is incremental in applying existing POMDP methods to this domain.
The authors tackled the problem of designing optimal haptic feedback for high-dimensional motor learning by proposing a data-driven skill-informed framework, which accelerated task performance significantly in a human-subject study with 30 participants compared to heuristic-based or no feedback groups.
In this work, we propose a data-driven skill-informed framework to design optimal haptic nudge feedback for high-dimensional novel motor learning tasks. We first model the stochastic dynamics of human motor learning using an Input-Output Hidden Markov Model (IOHMM), which explicitly decouples latent skill evolution from observable kinematic emissions. Leveraging this predictive model, we formulate the haptic nudge feedback design problem as a Partially Observable Markov Decision Process (POMDP). This allows us to derive an optimal nudging policy that minimizes long-term performance cost, implicitly guiding the learner toward robust regions of the skill space. We validated our approach through a human-subject study ($N=30$) using a high-dimensional hand-exoskeleton task. Results demonstrate that participants trained with the POMDP-derived policy exhibited significantly accelerated task performance compared to groups receiving heuristic-based feedback or no feedback. Furthermore, synergy analysis revealed that the POMDP group discovered efficient low-dimensional motor representations more rapidly.