Brain Controlled Computer Input Devices Advance a StepApr 23rd, 2011 | By Roy Rasmussen | Category: Technology and Science
Cell phone users can now dial a number just by thinking it, thanks to a brain computer interface device invented by researchers at the Swartz Center for Computational Neuroscience at the University of California, San Diego. The device brings brain-controlled computer input devices a step closer to practical application, though experts say reducing the size of the supporting display screen remains a challenge.
Developed by a team led by Swartz Center Associate Director Dr. Tzyy-Ping Jung, the device works by using electroencephalogram (EEG) electrodes on the user’s scalp to measure electrical activity in their brain. An EEG headband attaches to a Bluetooth module, which wirelessly sends signals to a Nokia N73 cell phone for processing. The EEG input can detect the frequency of visual input being processed by the brain.
To put this to practical use, Jung created a large computer screen keypad display where each number flashes at a slightly different frequency. This frequency difference enables the input device to tell which number the user is looking at.
Jung found that after a period of training, users could control their visual responses enough to use the input device to virtually dial numbers by looking at the display screen. Input accuracy varied from one user to the next. Many users can use the device to dial with 100 percent accuracy. However Jung himself can only achieve 85 percent accuracy.
Clearly, improved accuracy will be needed to move the technology forward. One solution is to improve accuracy by training users over a longer period. A Dartmouth research team has developed a device called a NeuroPhone which uses a different method of analyzing input requiring longer training time.
Another step forward would be reducing the size of the input screen, which is currently about the size on an iPad. Eric Leuthardt of the Center for Innovation and Neuroscience Technology at Washington University, who recently published related research in the Journal of Engineering, says an input screen closer to the size of a cell phone would be desirable, but the problem is that the technology requires relatively large visual input, which is difficult to achieve with a cell phone.
If this can be achieved, however, Jung and his colleagues soon hope to see practical applications of brain computer interface devices. Applications are currently geared towards people with disabilities who have difficulty using regular input devices. But Jung looks forward to mass-market applications as well. “I want to target larger populations,” Jung says.