Thought-Controlled Prosthetics?

In experiments published in the June 7 issue of the journal Science, monkeys were able to move balls around in 3D space on a computer screen just by thinking about it. With a little practice, they got even better at it.

"They achieved nearly the same accuracy and speed as normal arm movements," said senior author Andrew Schwartz, Ph.D., of the Department of Bioengineering at Arizona State University and the Neurosciences Institute in San Diego.

To begin with, two monkeys learned a computer game that required moving balls around in 3D space on the screen. The balls moved in response to the monkeys' arm movements. The monkeys were rewarded for playing and were allowed to stop when they tired of it.

Then tiny electrodes were painlessly implanted in the monkeys' brains to record the motor control signals emitted when they moved their arms to play the game. The recorded signals were matched against specific arm/ball movements.

Finally, the monkeys were encouraged to play the game without using their arms. They had to move the balls on the screen using brain signals that corresponded with the appropriate arm movements.

At first they tried to use their arms. But as they realized they could move the balls without moving their arms, they relaxed their arms and continued the game using thought control. As they played, their game skills improved.

Controlling the movement of a prosthetic limb, however, would require learning a wider range of movements, so the two monkeys were given new tasks and 180-degree changes in directions.

"There was no significant difference between the novel and trained target hit rates in either animal, and both monkeys improved their performance with daily practice," the researchers reported.

Even when the monkeys were not allowed to first practice the movements with their arms, they still learned through visual feedback to control the computer images with their brain waves. This suggests that people who are paralyzed and have not had a chance to practice with their arms could still learn to move objects with thought control.

Schwartz's group cited recent case studies indicating that motor control centers are still active in the brain even after years of paralysis. This activity might be harnessed to control a prosthesis, but the technique and the electrodes used in the current experiments are not yet ready for human testing.

Schwartz's group is working in the relatively new area of neuroprosthetics. His earlier research in this area was supported by a Special Opportunity Award from The Whitaker Foundation. The lead author on the current paper is Whitaker graduate fellow Dawn Taylor of Arizona State's Department of Bioengineering.

Other groups have reported some early successes in neuroprosthetic research, mostly in animal experiments. Two years ago, researchers at Case Western Reserve University reported activating a prosthetic implant using human brain waves. The experiments did not require implanting electrodes in the brain.

A quadriplegic wearing a hat dotted with electrodes gained mental control of an arm prosthesis after a series of training sessions in which he learned to regulate his beta-rhythm through biofeedback.

He learned to move a cursor up or down on a computer screen just by thinking about it. Then he was connected to the neuroprosthesis. By thinking about moving the cursor up, he opened his hand, and by thinking "down," his hand closed. He demonstrated using the device to pick up and hold objects like a drinking glass and a fork.