Earlier this year when Erik Sorto, a quadriplegic man, used his thoughts to direct a robot arm to bring a beer to his lips, the media went wild. It was an impressive feat. The catch is that the technology behind it—an electrode-laden chip implanted in Sorto's brain—is expensive and invasive and often requires months of training. Worse, few paralyzed people have the psychological and physical profile the technology requires.
There could be a better way. Rather than creating a direct link between the brain's electrical activity and machines, Aldo Faisal, an associate professor of neurotechnology at Imperial College London, wants to use eye movements to control wheelchairs, computers and video games. With off-the-shelf video-game cameras, Faisal and his colleagues built goggles that record the user's eye movements and feed those data to a computer. Software then translates the data into machine commands. Almost anyone can use the technology, including amputees, quadriplegics and those suffering from Parkinson's disease, multiple sclerosis or muscular dystrophy. The system costs less than $50 to build. At a science exhibition, the vast majority of thousands of volunteers grasped the technology well enough after 15 seconds to play the game Pong, no instructions needed.
Scientists have long known that the eyes can reveal people's objectives—where they want to go, what they want to do, who they want to interact with. Drawing on 70 years of research into the neuroscience of eye movements, Faisal and his colleagues wrote algorithms that turn a glance into a command for a wheelchair, a wink into a mouse click or the dart of a pupil into the swing of a game paddle. To predict intention, the algorithms depend on training from real-world data, acquired by recording volunteers' eyes as they drove a wheelchair with a joystick or operated a robotic arm. Gradually the software learned to tell the difference between, for example, the way people look at a cup when they are evaluating its contents and when they want to pick it up and take a drink.
On supporting science journalism
If you're enjoying this article, consider supporting our award-winning journalism by subscribing. By purchasing a subscription you are helping to ensure the future of impactful stories about the discoveries and ideas shaping our world today.
Before Faisal can commercialize any medical devices based on the invention, he must secure funding for clinical trials. In the meantime, a €4-million grant from the European Union will support his group as it develops robotic exoskeletons that paralyzed people could control using the eye-tracking software it developed. “I want to see what I can do to help people move again,” Faisal says. “That's my focus.”
