Skip to main content

Thank you for visiting nature.com. You are using a browser version with limited support for CSS. To obtain the best experience, we recommend you use a more up to date browser (or turn off compatibility mode in Internet Explorer). In the meantime, to ensure continued support, we are displaying the site without styles and JavaScript.

BRAIN–COMPUTER INTERFACES

From unstable input to robust output

Nature Biomedical Engineering volume 4, pages 665–667 (2020)Cite this article

Subjects

In the presence of recording instabilities, the performance of brain–computer interfaces can be robustly maintained by exploiting ‘hidden’ structures underlying neural activity.

In people with paralysis, brain–computer interfaces (BCIs) can restore voluntary movements by interfacing directly with the brain to translate movement intention into action. The best-performing BCIs monitor movement-related neural signals via implanted microelectrodes. To translate the monitored signals into commands, a decoder is trained to find a mapping from recorded neural activity to a control signal. Progress in the development of BCIs has enabled their use in a range of applications, such as rapid typing, the control of anthropomorphic robotic arms, the production of synthetic speech and the stimulation of paralysed muscles to enable reaching and grasping1,2,3,4. However, neural recording instabilities incurred over time present challenges to maintaining robust closed-loop performance. For example, slight displacements of the implanted electrodes (relative to the surrounding brain tissue) can cause changes in recorded neuron identity and lead to intraday and inter-day instabilities, confounding the decoding of intent5,6. Reporting in Nature Biomedical Engineering, Byron Yu and colleagues now show that the decoding performance of BCIs can be stabilized by harnessing ‘hidden’ structures (known as low-dimensional neural manifolds) underlying the activity of large numbers of neurons7.

This is a preview of subscription content, access via your institution

Access options

Rent or buy this article

Get just this article for as long as you need it

$39.95

Learn more

Prices may be subject to local taxes which are calculated during checkout

Fig. 1: Patterns in the underlying activity of simultaneously recorded neurons can be leveraged to improve the robustness of decoders for BCIs.

References

  1. Pandarinath, C. et al. eLife 6, 18554 (2017).

    Article  Google Scholar 

  2. Collinger, J. L. et al. Lancet 381, 557–564 (2013).

    Article  Google Scholar 

  3. Anumanchipalli, G. K., Chartier, J. & Chang, E. F. Nature 568, 493–498 (2019).

    Article  CAS  Google Scholar 

  4. Ajiboye, A. B. et al. Lancet 389, 1821–1830 (2017).

    Article  Google Scholar 

  5. Sussillo, D., Stavisky, S. D., Kao, J. C., Ryu, S. I. & Shenoy, K. V. Nat. Commun. 7, 13749 (2016).

    Article  CAS  Google Scholar 

  6. Downey, J. E., Schwed, N., Chase, S. M., Schwartz, A. B. & Collinger, J. L. J. Neural Eng. 15, 046016 (2018).

    Article  Google Scholar 

  7. Degenhart, A. D. et al. Nat. Biomed. Eng. https://doi.org/10.1038/s41551-020-0542-9 (2020).

  8. Gallego, J. A., Perich, M. G., Miller, L. E. & Solla, S. A. Neuron 94, 978–984 (2017).

    Article  CAS  Google Scholar 

  9. Sadtler, P. T. et al. Nature 512, 423–426 (2014).

    Article  CAS  Google Scholar 

  10. Kao, J. C., Ryu, S. I. & Shenoy, K. V. Sci. Rep. 7, 7395 (2017).

    Article  Google Scholar 

  11. Dyer, E. L. et al. Nat. Biomed. Eng. 1, 967–976 (2017).

    Article  Google Scholar 

  12. Pandarinath, C. et al. Nat. Meth. 15, 805–815 (2018).

    Article  CAS  Google Scholar 

  13. Farshchian, A. et al. In International Conference on Learning Representations (OpenReview.net, 2019).

  14. Gallego, J. A., Perich, M. G., Chowdhury, R. H., Solla, S. A. & Miller, L. E. Nat. Neurosci. 23, 260–270 (2020).

    Article  CAS  Google Scholar 

  15. Jarosiewicz, B. et al. Sci. Transl. Med. 7, 313ra179 (2015).

    Article  Google Scholar 

  16. Oby, E. R. et al. Proc. Natl Acad. Sci. USA 116, 15210–15215 (2019).

    Article  CAS  Google Scholar 

Download references

Author information

Authors and Affiliations

  1. Wallace H. Coulter Department of Biomedical Engineering, Emory University and Georgia Institute of Technology, Atlanta, GA, USA

    Lahiru N. Wimalasena & Chethan Pandarinath

  2. Physiology Department, Northwestern University, Chicago, IL, USA

    Lee E. Miller

  3. Department of Physical Medicine and Rehabilitation, Northwestern University, Chicago, IL, USA

    Lee E. Miller

  4. Department of Biomedical Engineering, Northwestern University, Evanston, IL, USA

    Lee E. Miller

  5. Emory Neuromodulation and Technology Innovation Center, Emory University, Atlanta, GA, USA

    Chethan Pandarinath

Authors
  1. Lahiru N. Wimalasena
    View author publications

    You can also search for this author in PubMed Google Scholar

  2. Lee E. Miller
    View author publications

    You can also search for this author in PubMed Google Scholar

  3. Chethan Pandarinath
    View author publications

    You can also search for this author in PubMed Google Scholar

Corresponding author

Correspondence to Chethan Pandarinath.

Rights and permissions

Reprints and Permissions

About this article

Verify currency and authenticity via CrossMark

Cite this article

Wimalasena, L.N., Miller, L.E. & Pandarinath, C. From unstable input to robust output. Nat Biomed Eng 4, 665–667 (2020). https://doi.org/10.1038/s41551-020-0587-9

Download citation

  • Published:

  • Issue Date:

  • DOI: https://doi.org/10.1038/s41551-020-0587-9

Search

Advanced search

Quick links

Nature Briefing

Sign up for the Nature Briefing newsletter — what matters in science, free to your inbox daily.

Get the most important science stories of the day, free in your inbox. Sign up for Nature Briefing