Linking Multiple Minds Could Help Damaged Brains Heal

Monkeys and rats hooked up as “brainets” may lead to innovative treatments for Parkinson’s, paralysis and more

Brain-to-brain interfaces may soon be a therapeutic technique. Nixxphotography/iStock

If someone has a damaged kidney, a donor can give them a spare organ so both people can live. Now, what if you could do something similar for a friend with a failing brain?

In recent lab experiments, Miguel Nicolelis and his colleagues melded together the brains of multiple monkeys and rats to function as “brainets”—shared networks able to cooperatively manipulate a virtual arm and make calculations and decisions. Nicolelis hopes that linking human brains this way could unleash a powerful new suite of neurological tools that might help heal people with ailments from Parkinson's to paralysis.

“We are trying a completely new field of brain rehabilitation,” explains Nicolelis, director of Duke University's Center for Neuroengineering. “We are going to try to act on the circuits of the brain and actually improve the functional activity of the brain.”

“We tend to forget that the brain is one of the most computationally powerful devices ever developed,” adds Andrea Stocco of the University of Washington, who wasn't involved in the research. “For the things that it's been developed to do, like making sense of a scene we see for the first time or controlling complex movements of our limbs, it's just unbeatable. Now Nicolelis shows we can combine the computational power that the brain has to possibly solve the kinds of ill-defined problems that are really hard for our software but easy for our biological hardware, our neurons, to solve.”

Brain-machine interfaces have been around for about two decades, and the technology is finding uses in a variety of medical treatments. For instance, some devices use the brain's electrical output, translated by a computer, to allow people to control prosthetics or manipulate a wheelchair. But previous work involved just a single operator. Nicolelis hoped to learn if more than one subject could work together as a shared brain-machine interface to enhance neural activity.

His team fitted three monkeys with implanted electrodes that monitored and recorded neural activity, which could then be combined by a computer. The monkeys were stationed in separate rooms, each with a digital display on which the monkey could use its brain-machine interface to manipulate a virtual arm toward a reward. In some tests, the monkeys shared control of the arm, while in others each one controlled movement in a particular direction. None of the animals knew they were collaborating to move the arm. Amazingly, they not only accomplished the task, they improved with practice. 

“It turns out, apparently, that just by having visual feedback and getting a reward for performing an action, these animals can actually synchronize their brains, and they can learn to meet the demands of a particular task,” says Nicolelis, whose team described the results last week in Scientific Reports

“The monkeys, working together, were slower to complete the task than what any of them could do manipulating a joystick—but they were learning exactly as fast,” Stocco says. “That is amazing, and it seems to mean that for the brain, this problem becomes as easy to interpret as any problem of sensory motor coordination. From this great beginning you can see how you might create more complex tasks that the monkeys will be able to do better working together than any of them could do by themselves.”

In a separate experiment also described in Scientific Reports, four rats were physically linked with a microwire to explore how their brains worked together as a networked unit to tackle a suite of problems. The rats were fed electrical pulses of information and rewarded when they synchronized their brains. They also received data, like temperature and barometric pressure. The rats stored, retrieved and shared this information—enabling their brainet to perform better at analyses like weather prediction than a single, wireless rat.

“What they did was really pushing the envelope, and I found it fascinating,” says Stocco, who made history two years ago, with his colleague Rajesh Rao, with the first remote human-to-human brain interface. Rao sent a brain signal via the internet that moved Stocco's hand even as he sat in a room across campus.

Nicolelis suggests that humans may already engage in a natural form of brain sharing when subjected to common feedback—without realizing it's happening. “The interesting thing is that this probably happens all the time with us. When we're watching a movie in a theater, this type of feedback is probably synchronizing brains in the audience so that we have those group reactions, laughing or crying at the same moments," he says.

"It can also explain why groups of people can collaborate to achieve a common goal. Like a sports team for example, where we often watch and say that a team is finally playing like a team and not like a bunch of individuals. Nobody has actually put their hands on what exactly is that chemistry that makes a soccer team play better. It may be that we've found the mechanism—synchronization of brain action.”

His team is now working on attempts to translate the monkey study to noninvasive clinical practices to possibly help rehabilitate paralyzed humans through group thought and action.

“For example, in my paraplegic patients I noticed it was very difficult to start training using a brain signal, because the brain can literally forget that you have legs,” he explains. “So part of this neurological rehab is to use another brain to reintroduce the concept to the brains of the patients. We could possibly have a physical therapist or even the patient's relatives help during the training phase by basically combining their brain activities with the patient's brain activities.” 

Get the latest Science stories in your inbox.