Main menu


Living Brain Cells Playing Pong in a Dish Can Illuminate the Mechanics of the Mind

featured image

Scientists have created a player from cells in a laboratory.

An Australian-led research team placed 800,000 living human and mouse brain cells on a plate and connected them to electrodes and a simulation of the classic game of Pong. The scientists then watched as the mini-mind quickly taught itself the game and developed more as they practiced. They were able to track the cellular responses by turning it into a visual depiction of the game very similar to the original.

They call their system DishBrain and say it proves that neurons in a dish can learn and show basic signs of intelligence. The team details the new setup, called synthetic biological intelligence, or SBI, in a study published Wednesday in the journal Neuron.

Finally, the authors say SBI can help unlock longstanding mysteries of brain mechanics and leads to better treatments for certain neurological conditions. “DishBrain offers a simpler approach to testing how the brain works and learning about debilitating conditions like epilepsy and dementia,” says Hon Weng Chong, CEO of biotechnology startup Cortical Labs.

SBI could also offer an alternative to animal testing, which scientists often turn to to study the viability of new drugs and therapies.

“Now, in principle, we have the ultimate biomimetic ‘sandbox’ for testing the effects of drugs and genetic variants – a sandbox made up of exactly the same computing (neuronal) elements found in your brain and mine,” he adds. The author is Professor Karl Friston, a theoretical neuroscientist at University College London.

biological intelligence versus artificial intelligence

The team found that biological intelligence, or living brain cells, behaves quite differently in terms of artificial intelligence than a computer can.

“In the past, brain models were developed based on how computer scientists thought the brain might work,” says Brett Kagan, Cortical Labs’ chief science officer and co-author of the study. “This is often based on our current understanding of information technology, such as silicon computing… But we don’t really understand how the brain actually works.”

Interestingly, DishBrain learned to play Pong naturally due to its tendency to act on its surroundings in a way that would make it more predictable and less random. In other words, this system behaves more like a real living brain than an artificial intelligence.

For example, when DishBrain successfully returned the “ball” in Pong, this resulted in the system being able to better predict where to go next. If DishBrain failed, he would lose points and a new point would start when the computer randomly dropped a ball from a starting point, and so on. Since DishBrain uses a feedback loop, it seems to get better and better as you play.

“This is remarkable because you can’t teach this kind of self-organization because — unlike a pet — these mini-brains have no sense of reward and punishment,” adds Friston.

Now Cortical Labs, an Australian biotech startup, is working on next-generation biological computer chips to create a generalized form of SBI. advantage of biological systems.”

“We know that our brains have the evolutionary advantage of adjusting for hundreds of millions of years to survive,” explains co-author Adeel Razi of Monash University. “Now, it looks like we have this incredibly powerful and inexpensive biological intelligence in our hands.”

The researchers also tried the system in other simple games.

“You know when the Google Chrome browser crashes and you get that dinosaur that you can jump over obstacles (Project Bolan),” says Kagan. “We’ve done that and have seen some nice preliminary results, but we still have a lot of work to do to create new environments for specific purposes.”

Next up, the team has plans to show DishBrain how to have a good time.

“We’re trying to build a dose response curve with ethanol – basically get them ‘drunk’ and see if they play the game worse, just like when people drink,” says Kagan.

While we’ll be eagerly awaiting the results of the drunken DishBrain study, let’s keep intoxicated neurons out of autonomous car codes.