Motor
A high-performance brain–computer interface for finger decoding and quadcopter game control in an individual with paralysis
- Authors
- Willsey MS … Henderson JM9 authors
- Institution
- Stanford University; University of Michigan
- Publication
- Nature MedicineNat Med 2025;31(1):96–104 · 20 Jan 2025
- Status
- Primary source
The 60-second view
- What problem?
- Cursor control offers two dimensions; many activities people value, including games, need more.
- What did they do?
- Decoded three independent finger groups, with two-dimensional thumb movement, and mapped them to quadcopter controls.
- What changed?
- Doubled the decoded degrees of freedom compared with earlier work while keeping fast target acquisition.
- Why does it matter?
- The target was chosen by the participant: recreation and social connection, a need that conventional assistive technology rarely addresses.
Method
- Participants
- One person with tetraplegia from spinal cord injury
- Interface
- Intracortical microelectrode arrays in motor cortex
- Signal
- Intracortical spiking activity
- Task
- Finger reach-and-hold targets; virtual quadcopter obstacle courses
- Decoder
- Continuous finger-position decoding
Results
- 76 targets/min
- Target acquisition
- 1.58 ± 0.06 s
- Completion time
- 4
- Degrees of freedom
Strength of evidencePeer reviewed, one participant.
Figures are quoted from the published abstract and were checked on 3 Oct 2026.
Limitations
- One participant.
- Virtual effector only.
What this could enable
Dexterous multi-finger control for typing, instruments and games.
Primary source
- 01
Willsey MS et al. Nat Med 2025;31(1):96–104 (opens in a new tab)
Peer reviewedNature Medicine20 Jan 2025Checked against source 3 Oct 2026
DOI 10.1038/s41591-024-03341-8