Interfaz cerebro a cerebro transmite información de una rata a otra.

Brain-to-brain interface transmits information from one rat to another

Electronically linked brains could facilitate rehabilitation and revolutionize computing

Image: Katie Zhuang/ Miguel Nicolelis/ Duke University In Star Trek, the Borg is a menacing race of cybernetically-enhanced beings who conquer other races and assimilate them. They do not act as individuals, but rather as an interconnected group that makes decisions collectively. Assimilation involves integrating other life forms into the Collective, using brain implants that connect them to the "hive mind," such that their biology and technology can help the Borg to become the perfect race. This is a popular concept that can be found elsewhere in science fiction, but scientists have now moved a step closer to making it a reality.
Earlier this month, Miguel Nicolelis of Duke University Medical Center and his colleagues reported the development of a brain-machine interface that enables rats to detect infrared light via their sense of touch. Now, the same group of researchers has taken this technology in an entirely new direction – they have developed a brain-to-brain interface that can transmit information from one rat directly to another, enabling the animal on the receiving end to perform behavioural tasks without training.

In one experiment, Nicolelis and his colleagues placed rats into a box containing two levers, and trained the animals to press one of them whenever it lit up, or to poke their noses into one of two different-sized holes in order to get a drink. They then trained another group of rats to perform both tasks while their brains were stimulated with electrodes implanted into the motor cortex, which controls movement, or the somatosensory cortex, which processes touch information, mostly from the whiskers. In this way, the second group of animals learned the gist of both tasks and became accustomed to pressing one of the levers and poking their noses into one of the holes, depending on the frequency of the electrical stimulation.
The rats were then paired up, placed in separate boxes, and their brains electronically linked – animals from the first group (the "encoders") had recording electrodes implanted into their motor cortex, and the implants were connected, via a computer, to the stimulating electrodes implanted into animals from the second group (the "decoders"). Next, each encoder rat was made to perform the lever-pressing task again, and while they did so, the pattern of brain activity encoding their behavioural responses was transmitted to the decoder.
The researchers found that the decoder rats could learn to perform the same movements, and successfully complete the task, guided solely by the information they received from the brains of the encoder rats. Likewise, when the implants were embedded into the somatosensory cortex, the decoders could use the sensory information they received to mimic the encoders' actions and poke their nose into the right hole to get a drink. They could also transmit the information over the internet in real time, so that the brain activity of an encoder rat in the lab at North Carolina could guide the behaviour of a decoder animal in Brazil.
Even more remarkably, this direct brain-to-brain communication created a feedback loop between the two animals. "The encoder would get a reward if the decoder performed the task correctly," says Nicolelis. "But when the decoder got it wrong, the encoder would move more accurately the next time, so that its brain activity pattern became clearer." It's not clear exactly how the decoder rats integrated natural stimuli with the virtual information received via their implants, and this is something the researchers would like to investigate.
In the future, implants such as these could be used to facilitate the rehabilitation of stroke patients and people who suffer from motor neuron disease, Parkinson's and other movement disorders. Nicolelis also believes that brain-to-brain interfaces could eventually be used to transmit more complex patterns of brain activity. "We are now working with monkeys and training them in pairs to control [computer-generated] body avatars," he says. "They will meet in a virtual space and learn to play a game. They'll have to share the rules by direct brain-to-brain interaction, and combine their brain activity to complete the game."
In his recent book Beyond Boundaries, Nicolelis proposes the idea of a brain net – multiple, interconnected brains that work collectively to solve problems. "We can test the emergent behaviour that would come out when many rats or monkeys exchange information via brain-to-brain interfaces," he says. "This could lead to organic computers that perform heuristically instead of using algorithms. I have no doubt that human brain nets will be possible in the future, but I certainly won't see this in my lifetime."
Reference: Pais-Vieira, M., et al. (2013). A Brain-to-Brain Interface for Real-Time Sharing of Sensorimotor Information. Scientific Reports, DOI: 10.1038/srep01319