A plate of living brain cells has learned to play the 1970s arcade game Pong.
About 800,000 cells connected to a computer gradually learned to sense the position of an electronic game ball and control a virtual paddle, a team reports in the journal. neuron.
The new achievement is part of an effort to understand how the brain learns and how to make computers more intelligent.
“We’ve made great strides with silicon computing, but they’re still rigid and inflexible,” says Brett Kagan, a study author and chief scientific officer at Cortical Labs in Melbourne, Australia. “That’s something we don’t see with biology.”
For example, both computers and humans can learn to make a cup of tea, Kagan says. But humans are able to generalize what they have learned in a way that a computer cannot.
“You may never have been to someone else’s house, but with a bit of research and searching you can make a good cup of tea as long as I have the ingredients,” he says. But even a very powerful computer would struggle to perform that task in an unfamiliar environment.
So Cortical Labs has been trying to understand how living brain cells acquire this kind of intelligence. And Kagan says the Pong experiment was a way for the company to answer a key question about how a network of brain cells learns to change its behavior:
“If we let these cells know the outcome of their actions, will they actually be able to change in some goal-directed way,” Kagan says.
To find out, the scientists used a system they developed called DishBrain.
This scanning electron microscope image shows a neural culture growing on a high-density multi-electrode array. This system allowed the researchers to train neurons to play the video game Pong.
Cortical laboratories
hide title
change the subtitles
Cortical laboratories
This scanning electron microscope image shows a neural culture growing on a high-density multi-electrode array. This system allowed the researchers to train neurons to play the video game Pong.
Cortical laboratories
A layer of living neurons is grown on a special silicon chip at the bottom of a thumb-sized dish filled with nutrients. The chip, which is connected to a computer, can detect electrical signals produced by neurons and give them electrical signals.
To test the cells’ learning ability, the computer generated a game of Pong, a two-dimensional version of table tennis that gained a cult following as one of the first and most basic video games.
Pong is played on a video screen. A black rectangle defines the board and a white cursor represents each player’s paddle, which can be moved up or down to capture a white ball.
In the simplified version used in the experiment, there was a single paddle on the left side of the virtual table and the ball would bounce off the other sides until it avoided the paddle.
To allow the brain cells to play the game, the computer sent them signals telling them where the bouncing ball was. At the same time, she began to monitor the information coming from the cells in the form of electrical pulses.
“We took that information and allowed it to influence the game of Pong they were playing,” Kagan says. “So they can move the paddle around.”
At first, the cells didn’t understand the signals coming from the computer, or didn’t know what signals to send in the other direction. They also had no reason to play the game.
So the scientists tried to motivate the cells using electrical stimulation: a well-organized burst of electrical activity if they got it right. When they got it wrong, the result was a chaotic stream of white noise.
“If they hit the ball, we gave them something predictable,” Kagan says. “When they lost that, they got something that was totally unpredictable.”
The strategy was based on the Free Energy Principle, which states that brain cells want to be able to predict what is happening in their environment. So they would choose the predictable stimulus over the unpredictable stimulus.
The approach worked. The cells began to learn to generate patterns of electrical activity that would move the paddle in front of the ball, and gradually the bouts lengthened.
Brain cells were never so good at Pong. But interestingly, human brain cells seemed to reach a slightly higher level of play than mouse brain cells, Kagan says.
And the level of play was remarkable, given that each network contained fewer cells than a cockroach’s brain, Kagan says.
“If you could see a cockroach playing a game of Pong and it was able to hit the ball twice as often as it missed it, you’d be very impressed with that cockroach,” he says.
The results suggest a future in which biology helps computers become more intelligent by changing the way they learn, Kagan says.
But that future is probably still a long way off, says Steve M. Potter, an associate professor at Georgia Tech.
“The idea of a computer that has some living components is exciting and is starting to become a reality,” he says. “However, the kinds of learning these things can achieve are pretty rudimentary right now.”
Even so, Potter says the system that allows cells to learn Pong could be a great tool for doing research.
“This is a kind of semi-living animal model that can be used to study all kinds of mechanisms in the nervous system, not just learning,” he says.
