Researchers at the University of Washington have created a kind of “telepathic” brain network, allowing three people to send simple signals directly into each other’s brains. It’s called “BrainNet,” and, according to its designers, it is the world’s first non-invasive, multi-person “brain-to-brain interface for collaborative problem solving.”
The experiment involved three people collectively playing a game of Tetris (kind of like a scaled-down, collaborative version of “Twitch Plays Pokémon”). Subjects were able to correctly transmit and interpret the brain signals from the other players 81% of the time, and they were even able to distinguish between good and bad “telepaths,” learning to ignore the partners who usually signaled incorrect moves.
How It Works
Fifteen players were grouped into teams of three and hooked up to a simple, non-invasive form of “mindreading” called an electroencephalograph or EEG. Electrodes are stuck to the head to measure brain waves, and sometimes those wave patterns can be correlated with mental activity.
For instance, looking at an LED pulsing at 17 times per second (17 Hz) produces a brain pattern that is distinct from when you're looking at one running at 15 Hz. By focusing on one LED or the other, people can choose to send a simple “yes/no” signal using only their thoughts.
The second component was transcranial magnetic stimulation (TMS), a kind of neurostimulation using powerful electromagnets. It’s also a safe, non-invasive tool, and it can be used to stimulate flashes of light, called phosphenes, in a user’s field of vision—without any light actually entering the eye.
You can see these types of spots if you rub your eyes too hard, or if you stare at a bright light too long and then look away. With a less powerful stimulation, TMS creates only a peripheral tingling and a noise.
This kind of direct neural stimulation allows people to receive messages (in the form of a flash, noise, or slight tingling) directly into the brain.
Researchers combined the EEG and TMS systems to let players send and receive yes/no messages using only their minds. Players were split up into three different rooms and hooked up to the machines.
This Is What All Those Hours of Tetris Were For…
Their task was a simple Tetris-like game. Only one player (“the receiver”) could control the falling shape, but she wasn’t allowed to see the hole it needed to fill at the bottom of the screen.
The other two players (“the senders”) could see both the shape and the hole, so they had to tell the controller whether to rotate it or not—using only their thoughts.
To send a “rotate” signal, the senders focused on a 17 Hz LED; to signal “don’t rotate,” they focused on a 15 Hz LED. The EEG recorded their brain waves, and then a computer transmitted those signals through TMS to the receiver, one after the other.
If the signal was “rotate,” the electromagnet gave her a phosphene flash; if the signal was “don’t rotate,” she wouldn’t see anything, but would only feel a peripheral sensation or noise, letting her know she had still received a message.
Finally, to actually rotate the block, the receiver would use her EEG to move a dot on the computer screen between “yes” and “no” buttons. If the receiver made the wrong move, the senders could see it and would have a chance to send a follow-up signal to correct her. When the shape hit the bottom, the receiver would get points added to her screen if she had rotated it correctly.
On average, the five teams got it right about 81% of the time, significantly better than chance.
Messing With Their Minds
For a final twist, researchers threw a curve ball at them to test what would happen if the players sent conflicting signals to the receiver. The experimenters randomly chose one player to be a “bad sender,” and the computer would sometimes flip that person’s signal to be the opposite of what they intended.
Because the receiver got their messages in the same order every time, she might learn to distrust sender #1 or #2, based on whether she scored when she followed to their advice. The results showed that the receivers were able to identify and ignore bad senders, and eventually they only listened to the signals from the “good sender.”
“Telepathic Team Tetris” is a fun concept, but is there a more important upshot to this study? The paper (which is only in preprint at this point) is sketchy on the bigger picture, only suggesting that experiments like this can help “open new frontiers in human communication” and may “provide us with a deeper understanding of the human brain.”
That’s pretty vague, but the authors also say their experiment is “bringing us a step closer to a ‘social network of brains.’” It's not clear what that would look like, but they are charging ahead to wherever that leads. The paper hints at plans to upgrade BrainNet with fMRI to “increase the bandwidth of human [brain-to-brain interfaces],” allowing it to carry more information. They also suggest “using TMS to stimulate higher-order cortical areas to deliver more complex information such as semantic concepts.”
If that’s not cool enough, they also hope to bring the BrainNet system online, making it “globally operable through the Internet, thereby allowing cloud-based interactions between brains on a global scale.”