From studies on animals scientists know that specific nerve cells become active just before the animal makes a certain movement. By processing the electric signals from these cells it is possible to predict the movement before it is actually made. The same information could also be used to make an artificial limb perform the same movement.

Moran and colleagues measure the neuronal signals by placing a grid of electrodes directly on the surface of the brain (figure 1). This method, which is known as electrocorticography (ECoG), is routinely used to monitor and control seizures in patients with epilepsy. It provides much better signals than electroencephalography, in which the electrodes are placed on the scalp, and is much less traumatic than procedures in which the electrodes actually penetrate the brain.

Leuthardt, Moran and co-workers asked four volunteers -- who were already wearing the electrodes -- to perform certain movements that included opening and closing their hands and sticking out their tongue. During this time, the scientists identified which brain signals were associated with each of the different movements.

Next, the patients were asked to attempt to control a cursor on a computer screen by imagining to make these movements. This part of the research relied on a computer program called BCI2000 that had been developed by team member Gerwin Schalk at the Wadsworth Center in New York. After just a few minutes of training, the patients were able to control the cursor using their thoughts with an accuracy of over 70%.

"The ability to operate a prosthetic device using the electric signal of the brain is rapidly leaving the realm of science fiction and becoming a realistic goal of the scientific community," Leuthardt told PhysicsWeb. "Our research could ultimately be used as a tool for people with severe motor disability. It could allow them to better interact with the world through devices that are controlled by thought alone and do not require the use of peripheral nerves or muscles."