Skip to content
DraconiticResearch
Search Sources API
203,234 studies31 sourcesUpdated 46 min ago
Peer-reviewedCC BY-NC-NDNature Communications

Assistive algorithms influence neural representations in motor brain-computer interfaces

  • Pavithra Rajeswaran1
  • Alexandre Payeur2,3
  • Guillaume Lajoie4,5
  • Amy L. Orsborn6,7,8
8 affiliations
  1. 1Department of Bioengineering, University of Washington, Seattle, WA, USA
  2. 2Department of Mathematics and Statistics, Université de Montréal, Montréal, QC, Canada
  3. 3Mila - Québec Artificial Intelligence Institute, Montréal, QC, Canada
  4. 4Department of Mathematics and Statistics, Université de Montréal, Montréal, QC, Canada. [email protected]
  5. 5Mila - Québec Artificial Intelligence Institute, Montréal, QC, Canada. [email protected]
  6. 6Department of Bioengineering, University of Washington, Seattle, WA, USA. [email protected]
  7. 7Department of Electrical and Computer Engineering, University of Washington, Seattle, WA, USA. [email protected]
  8. 8Washington National Biomedical Research Center, Seattle, WA, USA. [email protected]
Open at sourcePDFDOICite

Abstract

Task errors are used to learn and refine motor skills. We investigated how task assistance influences learned neural representations using Brain-Computer Interfaces (BCIs), which map neural activity into movement via a decoder. We analyzed motor cortex activity as monkeys practiced BCI with a decoder that adapted to improve or maintain performance over days. Over time, task-relevant information became concentrated in fewer neurons, unlike with fixed decoders. At the population level, task information also became largely confined to a few neural modes that accounted for an unexpectedly small fraction of the population variance. A neural network model suggests the adaptive decoders directly contribute to forming these more compact neural representations. Our findings show that assistive decoders manipulate error information used for long-term learning computations like credit assignment, which informs our understanding of motor learning and has implications for designing real-world BCIs.

Related

  1. Learning Dynamic Neural Evidence Representations for Time-Adaptive Brain-Computer Interfaces
    Beining Cao, Ziyi Zhao, Xiaowei Jiang +5
    arXivPreprintCC BY 4.0Full textNeurosciencearXiv
  2. Human learning of noninvasive brain–computer interfaces via manifold geometry
    Erica L. Busch, E. Chandra Fincke, Guillaume Lajoie +2
    Nature NeurosciencePeer-reviewedLicence not stated7 citedNeuroscienceOpenAlex
  3. Real-time noise cancellation for EEG monitoring and brain-computer interfaces
    M. Makkonen, O.-P. Kahilakoski, J. Metsomaa +3
    bioRxivPreprintAll rights reservedFull textNeurosciencebioRxiv
  4. Large language models integrated into brain–computer interfaces for communication and control: a systematic review
    Alon Gorenshtein, Mahmud Omar, Yiftach Barash +2
    Biomedical Physics & Engineering ExpressPeer-reviewedLicence not statedFull text5 citedNeuroscienceOpenAlex
  5. Regulating Next-Generation Implantable Brain-Computer Interfaces: Recommendations for Ethical Development and Implementation
    Renée A. Sirbu, Jessica Morley, T. Schröder +5
    NeuroethicsPeer-reviewedLicence not stated10 citedNeuroscienceOpenAlex

Cite

BibTeXRIS
BibTeX
@article{rajeswaran2026assistive,
  title = {{Assistive algorithms influence neural representations in motor brain-computer interfaces}},
  author = {Pavithra Rajeswaran and Alexandre Payeur and Guillaume Lajoie and Amy L. Orsborn},
  journal = {Nature Communications},
  year = {2026},
  month = sep,
  doi = {10.1038/s41467-026-76109-y},
  url = {https://doi.org/10.1038/s41467-026-76109-y},
  note = {CC BY-NC-ND},
}
RIS
TY  - JOUR
TI  - Assistive algorithms influence neural representations in motor brain-computer interfaces
AU  - Pavithra Rajeswaran
AU  - Alexandre Payeur
AU  - Guillaume Lajoie
AU  - Amy L. Orsborn
PY  - 2026
DA  - 2026/09/15
JO  - Nature Communications
DO  - 10.1038/s41467-026-76109-y
UR  - https://doi.org/10.1038/s41467-026-76109-y
AB  - Task errors are used to learn and refine motor skills. We investigated how task assistance influences learned neural representations using Brain-Computer Interfaces (BCIs), which map neural activity into movement via a decoder. We analyzed motor cortex activity as monkeys practiced BCI with a decoder that adapted to improve or maintain performance over days. Over time, task-relevant information became concentrated in fewer neurons, unlike with fixed decoders. At the population level, task information also became largely confined to a few neural modes that accounted for an unexpectedly small fraction of the population variance. A neural network model suggests the adaptive decoders directly contribute to forming these more compact neural representations. Our findings show that assistive decoders manipulate error information used for long-term learning computations like credit assignment, which informs our understanding of motor learning and has implications for designing real-world BCIs.
SN  - 2041-1723
ER  -