Brain cells in a lab dish learn the pong game and provide a window into your intelligence.

A plate of living brain cells learned to play the 1970s arcade game Pong. About 800,000 cells connected to the computer gradually learned to sense the game’s electronic ball’s position and control the virtual paddle, the team reports in the journal Neuron. This new achievement is part of an effort to understand how the brain learns and how to make computers more intelligent. “Silicon computing has made tremendous strides, but it’s still rigid and inflexible,” said Brett Kagan, the study’s author and chief scientific officer at Cortical Labs in Melbourne, Australia. “It’s something we don’t see in biology.” For example, both computers and humans can learn to make a cup of tea, Kagan says. But people can generalize what they learn in ways that computers cannot. “You may have never been to someone else’s house, but with a little mess and searching, I’ll be able to make a decent cup of tea as long as I have the ingredients,” he says. However, even a very powerful computer will have a hard time doing it in an environment you are unfamiliar with. So Cortical Labs has been trying to understand how living brain cells acquire this kind of intelligence. And Kagan said the Pong experiment is how the company could answer a key question about how brain cell networks learn to change behavior. It’s kind of a goal-directed way to change,” says Kagan. To figure this out, scientists developed a system called DishBrain. This scanning electron microscope image shows a neural culture growing on a high-density multi-electrode array. The system allowed researchers to train neurons to play the video game Pong: Cortical Labs hide caption toggle caption Cortical Labs This scanning electron microscope image shows a neural culture growing on a high-density multi-electrode array. This allowed the researchers to train neurons to play the video game Pong.Cortical Labs layers of living neurons are grown on special silicon chips on the bottom of a thumb-sized dish filled with nutrients.Chips connected to computers are created from neurons. It can detect electrical signals and transmit electrical signals.To test a cell’s ability to learn, computers created the game Pong, a two-dimensional version of ping-pong that gained a cult following in one of the first and most basic video games. The black rectangle defines the table, the white cursor represents each player’s hand and can be moved up or down to intercept the white ball In the simplified version used in the experiment, a single paddle is placed to the left of the virtual table. and the ball caromed on the other side until it diddge the paddle, so that the brain cells could play, the computer sent a signal to the brain cells telling them where the bouncing ball was at the same time monitoring the information coming from the cells in the form of electric pulses. “We took that information and allowed it to affect this Pong game they were playing,” Kagan says. “So they could move the oars around.” They didn’t understand and didn’t know the signals they were sending in the other direction, and they had no reason to play the game either. So, scientists have tried to motivate cells using electrical stimulation. When properly organized, electrical activity is nicely organized. When they were wrong, the result was a chaotic stream of white noise. “When they hit the ball, they gave me something predictable,” says Kagan. “When they missed it, they got something completely unpredictable.” This strategy is based on the free energy principle that brain cells want to be able to predict what is happening in their environment. So they will choose a predictable stimulus over an unpredictable stimulus. The approach worked. The cells began to learn to generate patterns of electrical activity that moved the oar in front of the ball, and gradually the rallies became longer. Brain cells didn’t do so well in Pong. But interestingly, human brain cells appeared to achieve slightly higher levels of play than mouse brain cells, says Kagan. And given that each network contains fewer cells than a cockroach’s brain, the level of play was surprising. “If you could see a cockroach playing a game of pong and hit the ball twice as often as you missed it, you’d be pretty impressed with that cockroach,” he says. The results hint at a future where biology helps computers become more intelligent by changing the way they learn, Kagan says. But the future is still far away, says Steve M. Potter, an adjunct associate professor at Georgia Tech. “The idea of ​​a computer with living components is exciting and is starting to become a reality,” he says. “But the kind of learning these things can achieve is at a very rudimentary stage right now.” Nevertheless, Potter says a system in which cells can learn Pong can be a great tool for conducting research. “It’s a kind of semi-biological animal model that can be used to study all kinds of mechanisms in the nervous system, not just learning,” he says.
#Brain #cells #lab #dish #learn #pong #game #provide #window #intelligence

Leave a Comment

Your email address will not be published. Required fields are marked *