Brain signals let monkeys regain wrist use

Aided by external wires that re-routed signals from their brains, two monkeys were able to regain control of their paralyzed wrists and played a simple video game, scientists said last week.

The study published in the journal Nature could lead to devices that allow people to get back some limb control after suffering spinal cord injuries and other forms of paralysis, scientists said.

The research is part of a growing field in which scientists are harnessing the power of the brain to overcome paralysis. In a previous experiment, for example, scientists demonstrated that people who have lost all use of their limbs can control a cursor on a computer screen using their thoughts.

The latest result is "an important step forward," said Dawn Taylor of Case Western Reserve University in Cleveland, who studies the concept of using brain signals to overcome paralysis. She was not involved in the new work, which she said illustrates the potential of the approach and the flexibility of brain cells.

For the experiment published last week, scientists separately trained two pigtail macaques to make a cursor reach targets on a computer screen by moving their wrists up or down.

Beneath the monkeys' skulls, scientists had implanted electrodes into an area of the motor cortex that is responsible for hand and wrist movements. As the monkeys played, the electrodes recorded signals from individual brain cells so scientists could determine a firing rate.

Once the monkeys had learned the game, scientists implanted wires that ran from the electrodes into the muscles of each animal's forearm. And they immobilized the animals' wrists temporarily by using an anesthetic.

A small battery-operated device used the firing rate to convert brain signals into electrical stimulation to the monkeys' muscles.

The animals continued to play the target practice game using their otherwise paralyzed wrists to move the cursors.

The monkeys' gaming skills improved with practice, researchers said, peaking at an average of nearly 15 correct hits per minute compared with an average of fewer than five hits per minute at the outset. Game playing sessions were limited by the duration of the anesthetic.

Researchers said they learned that only a single neuron was needed to control a pair of muscles, such as wrist flexors and extensors, and that cells anywhere in the motor cortex were capable of stimulating activity.

"Remarkably, every neuron we tested in the brain could be used," said Chet T. Moritz, a researcher at the University of Washington and an author of the study.

The discovery that neurons could be trained to perform new tasks greatly expands the number of cells that the experimental devices might tap into, he said.

Clinical applications for the technology are at least a decade away, said Eberhard E. Fetz, a professor of physiology and biophysics at the University of Washington, another author of the study.

Methods for controlling multiple muscles must be developed, along with a reliable and safe implantable device, he said. The device used in the study pokes through the skull, which increases the risk of infection.

Andrew Schwartz, a professor of neurobiology at the University of Pittsburgh who works on developing devices for paralyzed patients, said the challenges were huge.

"What they have done is a very simplistic muscle activation with a single cell going to one or two muscles," he said. "It is a nice sort of glimmer of something, but in the real world you need many muscles acting simultaneously."

"The problem of translating this into a neural prosthesis is quite daunting," said Gerald Loeb, a professor of biomedical engineering at the University of Southern California whose research is focused on reanimating paralyzed limbs.

Besides the technical hurdles, he said, cost looms as an obstacle.

"These systems are going to be very expensive to build and implant, and they have to be fitted and adjusted to the patient, so the threshold of performance you have to achieve to justify the cost and risk is quite high," he said. "This is a difficult problem in terms of making a commercial device insurance companies would pay for."