BrainGate participants type at up to 110 characters a minute by attempting finger movements
In Nature Neuroscience, a BCI that maps attempted finger movements to a QWERTY keyboard reached able-bodied-level accuracy for one user at home.
Researchers at Mass General Brigham and Brown University report in Nature Neuroscience a brain-computer interface that lets people type by attempting finger movements, which it maps onto a standard QWERTY keyboard. Of two BrainGate participants, one with ALS and one with cervical spinal cord injury, the top performer reached 110 characters (22 words) per minute with a 1.6% word error rate, comparable to able-bodied typing accuracy. The other reached up to 47 characters per minute. Calibration took as few as 30 sentences, and both used the system at home.
A familiar keyboard layout may suit users who prefer typing to speech decoding.
Learn more: Motor cortex · Neural decoding · BrainGate