HD-EMG to Assess Motor Learning in Myoelectric Control

Sigrid S.G. Dupan*, Ivan Vujaklija, Giulia De Vitis, Strahinja S. Dosen, Dario Farina, Dick F. Stegeman

*Corresponding author for this work

Research output: Chapter in Book/Report/Conference proceedingChapterScientificpeer-review


Online myoelectric control involves two types of adaptation: computational adaptation, in which the controller learns to associate muscle patterns with performed forces; and behavioural adaptation, where the users learn the new interface, and adapt their motor control strategies based on the errors they observe. In order to study the behavioural motor learning during online myoelectric control, twelve able-bodied participants performed single and 2-finger presses through force and myoelectric control. Myoelectric control was obtained with linear ridge regression, and was based on a training set only containing single finger presses. The distance between muscle patterns of force and EMG control trials indicated that motor learning leads to changes in neural drive, even on the trained presses. This suggests that motor learning is an integral part of myoelectric control, where the ability of the user to learn the EMG-to-force mapping impacts the overall performance of the myoelectric controller.

Original languageEnglish
Title of host publicationConverging Clinical and Engineering Research on Neurorehabilitation III
Subtitle of host publicationProceedings of the 4th International Conference on NeuroRehabilitation (ICNR2018), October 16-20, 2018, Pisa, Italy
Number of pages5
ISBN (Electronic)978-3-030-01845-0
Publication statusPublished - 1 Jan 2019
MoE publication typeA3 Part of a book or another research book

Publication series

NameBiosystems and Biorobotics
ISSN (Print)2195-3562
ISSN (Electronic)2195-3570


Dive into the research topics of 'HD-EMG to Assess Motor Learning in Myoelectric Control'. Together they form a unique fingerprint.

Cite this