A monkey controls his robotic arm with a brain-machine interface.
If this monkey can eat marshmallows with his robotic arm, mind-controlled prosthetics for humans can’t be far off, right? Well, that’s true if all you ever wanted to do with your prosthetic was sit strapped in a chair reaching for marshmallows. But as Michael Chorost explains in a recent feature for Wired, challenges abound when building an arm that works in everyday life.
Over the course of a day, you might use your arm to pick up a chair, unzip your jacket, or scratch your neck—each one of these actions are unique. But statistical algorithms used now can translate the firing of neurons into only a few stereotyped motions. And it’s not just about writing better algorithms; it’s an input problem too. Getting electrodes to pick up signals from the same neurons over time is a continuous battle against the body’s natural defenses:
Electrodes are made of metal. The body is loaded with water, salt, and a dizzying array of other chemicals. Putting them together is like trying to bond a fork and a steak. And the steak fights back by trying to dissolve the fork.
The steak treats the fork as a threat—which, of course, it is. Confronted with foreign bodies, the brain mounts an inflammatory response called gliosis, wrapping cells like astrocytes and microglia around the electrodes to wall them off. Over time, the electrodes become encapsulated in a sheath of scar tissue that acts as an insulator.
Prosthetics that avoid electrodes and attach sensors to the skin are also bedeviled by a lack of specificity. Muscles grow and shrink, the body changes, and sensors end up in a slightly different place each time the arm is strapped on, so movements have to be repeatedly recalibrated. Chorost is ultimately optimistic about the future as scientists develop techniques to record from hundreds of thousands of neurons at once and prevent scar tissue build-up around electrodes. It’s just that no one getting a prosthetic today should expect it to behave like a real limb.