A man who is unable to move or speak can now generate words and sentences on a computer using only his thoughts.
The ability comes from an experimental implanted device that decodes signals in the man’s brain that once controlled his vocal tract, as researchers reported Wednesday in The New England Journal of Medicine.
The man is currently limited to a vocabulary of just 50 words and communicates at a rate of about 15 words per minute, which is much slower than natural speech.
“This tells us that it’s possible,” says Edward Chang, a neurosurgeon at the University of California, San Francisco. “I think there’s a huge runway to make this better over time.”
A device that allowed people who can’t speak to communicate using brain circuits previously used for speech would be “more natural, and hopefully effortless compared to current assistive devices,” says Chethan Pandarinath, an assistant professor in the Department of Biomedical Engineering at Emory University and Georgia Tech.
Currently, people with paralysis who have lost the ability to speak usually rely on devices that use eye or head movements to spell out words one letter at a time. Some use a device that allows them to control a computer cursor with thoughts.
Chang’s team wanted to find a better solution for the man, identified only as BRAVO1 to protect his privacy. The name refers to his status as the first patient in a study called BRAVO, or Brain-Computer Interface Restoration of Arm and Voice.
BRAVO1, who is in his late 30s, has been paralyzed and unable to speak since he had a stroke 15 years ago, Chang says.
“The stroke left him nearly completely paralyzed in his arms and legs but also in the muscles of his vocal tract,” Chang says. But the areas of the brain that once issued speech commands are intact.
Dormant brain signals tapped for communication
Previously, Chang’s team had developed a system designed to recognize the brain signals associated with the intention to speak specific words. Tests showed that the system worked in people who were still able to move and speak.
But success was far from certain in someone such as BRAVO1, Chang says.
“We didn’t know if the speech commands in the brain would still work after 15 years,” he says. “And even if we could revive those dormant brain signals for speech, could we actually translate those into full words?”
To find out, the team implanted sensors on the surface of the man’s brain. Then it had a computer study the patterns of electrical activity produced when he attempted to speak 50 different words. The process took months.
Once BRAVO1 could reliably generate words on a computer screen, the team began having him form sentences. To help improve accuracy, the team added a program that analyzed the context of each word as it was added.
The system is a bit like the texting software on most smartphones. “So, for example, if one word is just not decoded correctly, this autocorrect function can correct it,” Chang says.
After months of adjustments to the system, the man was able to generate a word reliably every four seconds, or roughly 15 words per minute.
“Normal speech is on the order of 120, 150 words per minute, so there’s a lot of room to improve,” Chang says.
A device able to decode words in the brain could eventually help thousands of people who’ve had a stroke or a traumatic brain injury, says Krishna Shenoy, a professor in the School of Engineering at Stanford University.
Such a device might also help people with amyotrophic lateral sclerosis, or ALS, a paralyzing disease that eventually makes it impossible to speak, Shenoy says.
The ability to recognize even 50 words in the brain is a huge achievement, Shenoy says. “But I think that’s just the tip of the iceberg. I think that could easily become 500 or 5,000 words.”
Until that happens though, devices that recognize entire words will be limited to basic communication.
Keeping some thoughts private and others public
In the meantime, brain implants that allow people to spell out words keep getting better, Shenoy says. Earlier this year, he and a team published results of a study showing that people could spell quickly and accurately using a device that decoded the brain signals usually used for handwriting.
“We know far more now than 10 years ago or 20 years about how to eavesdrop on those conversations in the brain,” Shenoy says.
That sort of eavesdropping could compromise a person’s privacy though, Pandarinath says. That’s because devices wired directly to the brain could make it hard for people to separate private thoughts from those they choose to make public, he says.
“We want to make sure the devices we create allow that separation, allow people to be able to think their private thoughts without anything just being broadcast to the world,” he says.
That may be easier with devices that rely on brain signals that control muscles, he says. That’s because these signals generally aren’t sent unless a person makes a conscious attempt at movement.