Neurotechnology
High-Performance Brain-to-Text Communication via Handwriting
Willett et al. · 2021 · Nature
An intracortical BCI decoded imagined handwriting into text at 90 characters per minute.
Research objective
Restore communication for paralyzed individuals by decoding neural activity associated with imagined writing.
Methodology
Recorded from intracortical microelectrode arrays in the motor cortex of a paralyzed participant. Trained recurrent neural networks to map neural activity to letters as the participant imagined handwriting.
Key findings
- Achieved 90 characters/minute, comparable to smartphone typing in age-matched controls.
- 94.1% raw accuracy, >99% with autocorrect.
- Demonstrated that fine motor imagery produces decodable neural signatures.
Strengths
- Largest leap in BCI throughput in over a decade.
- Proof that high-dimensional motor cortex signals can encode rich symbolic information.
Limitations
- Invasive - requires surgical implantation.
- Single-participant study; long-term electrode stability remains a challenge.
Practical implications
- Validates the broader BrainGate program and intracortical BCIs.
- Influences Neuralink, Synchron, and Precision Neuroscience product strategy.
Related entities
Related research
Speech Synthesis from Neural Decoding of Spoken Sentences
Synthesized intelligible speech directly from cortical activity in the vocal motor area.
Read summary
Millisecond-Timescale, Genetically Targeted Optical Control of Neural Activity
Founding paper of optogenetics - using light to control genetically targeted neurons with millisecond precision.
Read summary