Nicolelis called the experiments an initial step towards the creation of a biological computer or “brain-net” linking multiple brains. “We cannot even predict what kinds of emergent properties would appear when animals begin interacting as part of a brain-net.
“In theory, you could imagine that a combination of brains could provide solutions that individual brains cannot achieve by themselves,” he said. “One animal might even incorporate another’s sense of self.”
Previous research has shown that visual and tactile information can be fed into and out of an individual rat’s brain, using microscopic electrodes implanted in the cortex – most recently when the animals learned to “feel” invisible infrared radiation.
The new study involved training pairs of animals in distant cages to solve a simple problem in return for a drink. In the first experiment, they had to press the correct lever corresponding to a particular indicator light; in the second they had to distinguish between wide and narrow openings.
The electrical brain activity picked up by electrodes in the brain of the first rat, the “encoder,” was fed into the brain of the second animal, the “decoder,” which had the same levers in its cage but received no visual cue about which one to press.
It took up to a month for the rats to tune into one another’s brainwaves, but eventually the best decoder animals achieved a 70 percent success rate – not far from the theoretical maximum of 78 percent, which the scientists calculated was the maximum achievable with their experimental set-up.
The encoder got a better reward if the decoder also made the right choice, leading to two-way neural collaboration between the two rats.
“We saw that when the decoder rat committed an error, the encoder basically changed both its brain function and behavior to make it easier for its partner to get it right,” said Nicolelis. “The encoder improved the signal-to-noise ratio of its brain activity that represented the decision, so the signal became cleaner and easier to detect.”
A scan of the decoder’s brain showed that it began to represent the encoder’s whiskers as well as its own whiskers in the tactile cortex, “which means that the rat created a representation of a second body on top of its own,” he said.
The collaboration worked not only when both rats were in the same lab but also when one was at Duke and the other in Brazil. “Even though the animals were on different continents, with the resulting noisy transmission and signal delays, they could still communicate,” said Miguel Pais-Vieira, another member of the team. “This tells us that we could create a workable network of animal brains distributed in many different locations.”
Professor Christopher James, an expert on brain-computer interfacing at Warwick University in Britain, said the Duke team’s plans for multi-brain networks might work in animals but, for ethical and practical reasons, would not be appropriate for humans in the foreseeable future.
“The system would require placing invasive electrodes in participants and the visual and tactile brain signals involved are quite crude,” James said. “You could not exchange abstract thoughts.”
— Financial Times