Tracking Forearm Motor Units Across Constrained and Unconstrained Finger Force Control Tasks

Reliable motor unit (MU) tracking across varied finger force conditions is crucial for neural interfaces that must function beyond controlled laboratory settings. We investigated whether individual MUs can be tracked between constrained force tasks and naturalistic finger force patterns using high- density surface EMG. Twenty-five participants performed iso- metric flexion of thumb, index, and middle fingers at 10 % and 20 % of maximum voluntary force, plus unconstrained multi-finger tasks, while recording multichannel EMG from the anterior forearm muscles. MU decomposition identified 302 ± 90 MUs per participant, with recruitment increasing 45- 70% when force doubled. Spatial activation patterns matched anatomical expectations: thumb MUs were localised radially; index/middle finger MUs, ulnarly. Notably, 96.7 % of MUs could be tracked across different trials of the same or different conditions at correlation threshold 0.7, with 62.8 % remaining trackable at 0.9 threshold. More MUs were shared between index and middle finger tasks than with thumb tasks, reflecting anatomical organization. Cross-condition tracking was highly dependent on the correlation threshold, with 0.9 required for reliable individual MU identification. These findings show that subject-specific MU patterns can be tracked across controlled and naturalistic conditions, which is crucial for future MU-based neural interfaces in real-world applications.

Recommended citation: J. Bodenschlägel, R. Mio and A. A. Faisal, “Tracking Forearm Motor Units Across Constrained and Unconstrained Finger Force Control Tasks,” 2025 International Conference on Neural Engineering (NER), San Diego, CA, USA, 2025, pp. 634-639.
Download Paper