The rats were in different cages with no way to communicate other than through the electrodes implanted in their brains. The transfer of information from brain to brain even worked with two rats separated by thousands of kilometers, one in a lab in North Carolina and another in a lab in Brazil.
“We basically created a computational unit out of two brains,” says neuroscientist Miguel Nicolelis of Duke University, who led the study.
Nicolelis is a leading figure in brain-machine interface research and the man behind a bold plan to develop a brain-controlled exoskeleton that would allow a paralyzed person to walk onto the field and kick a soccer ball at the opening ceremony of next year’s World Cup in Brazil.
He says the new findings could point the way to future therapies aimed at restoring movement or language after a stroke or other brain injury by using signals from a healthy part of the brain to retrain the injured area. Other researchers say it’s an interesting idea, but it’s a long way off.
But Nicolelis’s group is known for pushing the envelope. Previously, they have given monkeys an artificial sense of touch they can use to distinguish the “texture” of virtual objects. More recently, they gave rats the ability to detect normally invisible infrared light by wiring an infrared detector to a part of the brain that processes touch. All this work, Nicolelis says, is relevant to developing neural prostheses to restore sensory feedback to people with brain injuries.
In the new study, the researchers implanted small electrode arrays in two regions of the rats’ brains, one involved in planning movements, and one involved in the sense of touch.
Then they trained several rats to poke their noses and whiskers through a small opening in the wall of their enclosure to determine its width. The scientists randomly changed the width of the opening to be either narrow or wide for each trial, and the rats had to learn to touch one of two spots depending on its width. They touched a spot to the right of the opening when it was wide and the spot on the left when it was narrow. When they got it correct, they received a drink. Eventually they got it right 95 percent of the time.
Next, the team wanted to see if signals from the brain of a rat trained to do this task could help another rat in a different cage choose the correct spot to poke with its nose — even if it had no other information to go on.
They tested this idea with another group of rats that hadn’t learned the task. In this experiment, one of these new rats sat in an enclosure with two potential spots to receive a reward but without an opening in the wall. On their own, they could only guess which of the two spots would produce a rewarding drink. As expected, they got it right 50 percent of the time.
Then the researchers recorded signals from one of the trained rats as it did the nose-poke task and used those signals to stimulate the second, untrained rat’s brain in a similar pattern. When it received this stimulation, the second rat’s performance climbed to 60 or 70 percent. That’s not nearly as good as the rats who could actually use their sense of touch to solve the problem, but it’s impressive given that the only information they had about which spot to chose came from another animal’s brain, Nicolelis says.
Both rats had to make the correct choice, otherwise neither one got a reward. When that happened, the first rat tended to make its decision more quickly on the next trial, and its brain activity seemed to send a clearer signal to the second rat, the team reports today in Scientific Reports. That suggests to Nicolelis that the rats were learning to cooperate.
The brain-to-brain communication link enables the rats to collaborate in a novel way, he says. ”The animals compute by mutual experience,” he said. ”It’s a computer that evolves, that’s not set by instructions or an algorithm.”
From an engineering perspective, the work is a remarkable demonstration that animals can use brain-to-brain communication to solve a problem, said Mitra Hartmann, a biomedical engineer who studies rats’ sense of touch at Northwestern University. “This is a first, to my knowledge, although the enabling technology has been around for a while.”
“From a scientific point of view, the study is noteworthy for the large number of important questions it raises, for example, what allows neurons to be so ‘plastic’ that the animal can learn to interpret the meaning of a particular stimulation pattern,” Hartmann said.
“It’s a pretty cool idea that they’re in tune with each other and working together,” said neuroscientist Bijan Pesaran of New York University. But Pesaran says he could use some more convincing that this is what’s actually going on. For example, he’d like to see the researchers extend the experiment to see if the rats on the receiving end of the brain-to-brain communication link could improve their performance even more. ”If you could see them learning to do it better and faster, then I’d really be impressed.”
Pesaran says he’s open to the idea that brain-to-brain communication could one day be used to rehabilitate brain injury patients, but he thinks it might be possible to accomplish the same thing by stimulating the injured brain with computer-generated patterns of activity. ”I don’t get why you’d need another brain to do that,” he said.