ARLINGTON, Va., July 9, 2004 -- Four adults quickly learned to play a simple video game---and win---by using only their thoughts to control the computer.

"It took six minutes of training and they all achieved control in less than 24 minutes," said Eric Leuthardt, M.D., a neurosurgeon at Washington University in St. Louis. One player hit the target on every try.

"That's what was most impressive," said Whitaker investigator Daniel Moran, Ph.D., who collaborated with Leuthardt in the study. "The patients were getting good at it very quickly."

The experiment, reported last month in the Journal of Neuroengineering, demonstrated how it may be possible to someday give disabled people a measure of self-sufficiency using though-controlled computer interfaces to accomplish such tasks as moving artificial limbs.

Previous human experiments along these lines have used electroencephalographic (EEG) signals taken from electrodes placed on the scalp. Animal studies have used single electrodes implanted in the brain. But EEG signals can suffer from noise, and electrode implants can move about and become walled off from the nerve cells they are trying to monitor.

Moran and his colleagues used a new approach to collect brain signals for motor control. They recorded neural activity from the surface of the brain, obtaining strong signals without penetrating the cortex. Surface recordings such as these are typically made prior to surgery for epilepsy.

Moran and his colleagues obtained permission to couple their study with the presurgical evaluations of four epilepsy patients. Each patient had a subdural electrode array put in place for evaluation. These arrays are thin sheets of electrodes that are surgically placed on the surface of the brain and left there for several days to detect the locations where seizures originate. Tissue at these locations is removed to stop the seizures.

While hospitalized with the electrode array in place, the three men and one woman each watched a computer screen and performed one of six tasks: to open or close the right or left hand, stick out the tongue, say the word "move," or imagine performing one of these three actions. While doing so, the researchers recorded brain signals corresponding with each task.

Then the volunteers received online feedback in which a cursor would move across the computer screen, controlled by the matching brain signals that had been recorded previously.

In the game, the volunteers were asked to direct the moving cursor to a particular spot on the computer screen by thinking about it. For example, one volunteer imagined moving his right hand in order to move the cursor up and relaxing his hand to move the cursor down. They trained in this way for up to 24 minutes, controlling the cursor with their thoughts with 74 to 100 percent accuracy. The expected accuracy rate without any mental control would be 50 percent.

The volunteers also used a joystick to move a cursor from the center of a video screen to a target at various locations around the edge of the screen. Brain activity during these exercises was recorded and used to predict the vertical and horizontal directions of joystick movement. Further research is planned to explore thought-controlled two-dimensional cursor movements that simulate joystick control.

Any medical applications from this type of work lie far in the future and a great deal of research remains to be done.

"There will have to be a rigorous study on monkeys for an indeterminate number of years, but we're really excited about this advance," Moran said. "Brain-computer interface research is one of the hottest things going in biomedical engineering today."

Other members of the research team include Gerwin Schalk and Jonathan Wolpaw of the New York State Department of Health and Jeffrey Ojemann of the University of Washington School of Medicine in Seattle.

Moran received a 2002 Biomedical Engineering Research Grant from the foundation for similar studies of motor control in arm movements.

Contact:
Daniel Moran, Washington University
Frank Blanchard, The Whitaker Foundation