Researchers in California have created a way to place a call on a cell phone using just your thoughts. Their new brain-computer interface is almost 100 percent accurate for most people after only a brief training period.
The system was developed by Tzyy-Ping Jung, a researcher at the Swartz Center for Computational Neuroscience at the University of California, San Diego, and colleagues. Besides acting as an ultraportable aid for severely disabled people, the system might one day have broader uses, he says. For example, it could create the ultimate hands-free experience for cell-phone users, or be used to detect when drivers or air-traffic controllers are getting drowsy by sensing lapses in concentration.
Like many other such interfaces, Jung’s system relies on electroencephalogram (EEG) electrodes on the scalp to analyze electrical activity in the brain. An EEG headband is hooked up to a Bluetooth module that wirelessly sends the signals to a Nokia N73 cell phone, which uses algorithms to process the signals.
Participants were trained on the system via a novel visual feedback system. They were shown images on a computer screen that flashed on and off almost imperceptibly at different speeds. These oscillations can be detected in a part of the brain called the midline occipital. Jung and his colleagues exploited this by displaying a keypad on a large screen with each number flashing at a slightly different frequency. For instance, “1” flashed at nine hertz, and “2” at 9.25 hertz, and so on. Jung says this frequency can be detected through the EEG, thus making it possible to tell which number the subject is looking at.
“From our experience, anyone can do it. Some people have a higher accuracy than others,” says Jung, who himself can only reach around 85 percent accuracy. But in an experiment published in the Journal of Neural Engineering, 10 subjects were asked to input a 10-digit phone number, and seven of them achieved 100 percent accuracy.
In theory, the approach could be used to help severely disabled people communicate, says Jung. But he believes the technology doesn’t have to be limited to such applications. “I want to target larger populations,” he says.
“It’s interesting work,” says Rajeev Raizada, a cognitive neuroscientist at Dartmouth College who published work last year on a similar concept called the Neurophone. “People have used this sort of visually evoked response before, but the notion of making it small, cheap, and portable for a cell phone is attractive.”
The Neurophone used a brain signal known as the P300. This signal is triggered by a range of different stimuli and is used by other brain-control interfaces to gauge when something has caught a person’s attention. But this typically involves a longer training period.
However, Eric Leuthardt, director of the Center for Innovation and Neuroscience Technology at Washington University, is not convinced. “Reducing the size of the processors to a cell phone is a natural step,” he says. He says the kind of visually evoked response used in Jung’s research has been around for years, but it usually requires a large visual stimulus, which small cell phone displays are unlikely to elicit.
How a Russian cyberwar in Ukraine could ripple out globally
Soldiers and tanks may care about national borders. Cyber doesn't.
Meet Altos Labs, Silicon Valley’s latest wild bet on living forever
Funders of a deep-pocketed new "rejuvenation" startup are said to include Jeff Bezos and Yuri Milner.
A horrifying new AI app swaps women into porn videos with a click
Deepfake researchers have long feared the day this would arrive.
Meta’s new learning algorithm can teach AI to multi-task
The single technique for teaching neural networks multiple skills is a step towards general-purpose AI.
Get the latest updates from
MIT Technology Review
Discover special offers, top stories, upcoming events, and more.