Education

Duke scientists link brains of living rats

A graphic representation of the “brainet” created by neuroscientists at Duke University, in which hundreds of hair-like microfilaments were implanted in the brains of four rats. Each filament sensed the electrical activity of nearby neurons and transmitted that activity – in the form of a mild current – to the filaments implanted in another rat’s brain.
A graphic representation of the “brainet” created by neuroscientists at Duke University, in which hundreds of hair-like microfilaments were implanted in the brains of four rats. Each filament sensed the electrical activity of nearby neurons and transmitted that activity – in the form of a mild current – to the filaments implanted in another rat’s brain. Katie Zhuang, Nicolelis Lab, Duke University

A scientist links the brains of three lab rats and has them work together, using only electric pulses produced by their brain waves, to predict the weather.

It may sound like science fiction, but a neuroscience lab at Duke University claims to have done just that. The work is described in a pair of papers published Thursday in the journal Scientific Reports.

Miguel Nicolelis and colleagues are experimenting with the possibility of brain-to-brain communication by establishing “brainets,” or networks of living animal brains.

In one of the new studies, researches delivered stimuli representing weather data on temperature and barometric pressure to the brains of up to four rats. Using the brainets, the rats were able to combine the information to successfully predict the possibility of rain.

Nicolelis and his colleagues employed a technique called intracortical microstimulation after implanting hundreds of microfilaments in the brain of each rat. Each filament sensed the electrical activity of nearby neurons and transmitted that activity – in the form of a mild current – to the filaments implanted in another rat’s brain, which in turn stimulated its neurons.

That stimulation created a sensation that the rats were trained to interpret as a signal to perform a particular behavior, such as pressing a lever.

The scientists also used the technology on adult Rhesus monkeys in a set of experiments they describe in the second paper. Each monkey in the brainet – consisting of up to three monkeys – had partial control over a virtual arm. The brain activity of each monkey was sent to a computer, which generated an output that moved the arm. Working together, the monkeys were able to move the virtual arm to point to a target on the screen.

“We were surprised that when we combined these monkeys together, they really learned to do this,” said Nicolelis, a professor in the neurobiology department at Duke and co-director of the Center for Neuroengineering. “They really could acquire the ability to work in groups.”

Other scientists are skeptical of the practical applications of brainets.

The ideas presented in the two papers are “interesting and provocative,” said Amy Orsborn, a postdoctoral researcher at New York University, “But what exactly the applications are, I think, is something that is not immediately obvious to me.”

When the team first unveiled brain-to-brain communication in rats in 2013, the news received significant attention and criticism.

Other neuroscientists told The Scientist magazine in 2013 that Nicolelis’ team was not presenting brand new technology but was instead combining two existing technologies: One to interpret signals from neurons and the other to use those signals to stimulate machines or muscles.

‘Hollywood science fiction’

Scientists also criticized that study’s experimental design and said the meaning of its results was overstated. Some were particularly critical of Nicolelis’ goal of connecting brains to serve as an “organic computer.”

That concept reads “like a poor Hollywood science fiction script,” Lee Miller, a neuroscientist at Northwestern University, told The Scientist and Nature News in 2013.

Nicolelis argues that linking multiple brains could harness a computational power inaccessible to man-made machines. Multiple brains working together could solve “the kinds of problems that our brains are good at,” such as recognizing and generating patterns.

Linking brains to move an arm together could be useful for training patients to use new prosthetic limbs, Nicolelis said.

Last year, scientists at the University of Washington published a study claiming to establish brain-to-brain communication in humans. “Senders,” who could see but not control a video game, were outfitted in an electron-studded cap that recorded the electrical activity in a certain portion of their brain. These signals were sent over the Internet to a magnetic coil placed on the scalps of “receivers,” who could control but not see the game. The coil magnetically stimulated the area of the receiver’s brain that controls the wrist muscle.

Together, the pairs of volunteers were able to time and operate the firing of a virtual cannon.

“Such devices (which have been long cherished by science fiction writers) have the potential to not only revolutionize how humans communicate and collaborate, but also open a new avenue for investigating brain function,” the researchers, led by Rajesh Rao, wrote, although they did call their results a “rudimentary” type of communication.

Not telepathy

Nicolelis is quick to point out that this technology is not telepathy.

“You cannot broadcast your emotions, inner thoughts, memories, feelings. That probably will never happen,” he said. “That’s because they are confined to the analog domain of your brain. And that I cannot describe or report or compress in the kind of digital transmission that I did for the brain-to-brain interface.”

The idea of brain-to-brain interface is an offshoot of the much more widely accepted idea of brain-machine interface, a technology that scientists are developing to help people operate prosthetic limbs.

To be successful in such applications, scientists must not only enable a brain to send signals to the machine but also route sensation from the machine back to the brain.

We are far from perfecting the brain-machine interface or truly understanding how the brain works, said Sliman Bensmaia, a neuroscientist at the University of Chicago.

“When we can read signals from the brain and feed them back into the brain from machines, effectively and robustly, then maybe we can start to think about how brains can interface with each other,” he said.

But Orsborn, the researcher at NYU, sees the value in exploring ideas that some consider far-fetched.

“You sort of need these provocative papers that promote people to dig in further,” she said. “I don’t think that with these data that we can really say that we understand these systems well – just that it’s possible to create these systems and they do sort of interesting things. But the nuts and bolts of how they’re working is a little less clear.”

Rimler: 919-829-4526

  Comments