Researchers have recently taught human brain cells in a petri dish to play the retro videogame Pong faster than artificial intelligence (AI).
Scientists created what they dubbed the “cyborg brain” in order to measure biological neurons against AI. This “brain” consists of just brain cells in a petri dish grown on microelectronic arrays.
These arrays give researchers the ability to stimulate the cells, causing them to grow and reconfigure themselves in order to solve new problems. Current tests indicate that the “cyborg brain” can learn to prevent goals in Pong in as fast as five minutes.
Cortical Labs, the Australia-based team that trains the cyborg brain, says it typically takes AI ninety minutes to learn the same thing.
Brett Kagan, chief scientific officer of Cortical Labs, explains that the brain cells experience themselves as being inside the virtual world of Pong as they play the video game. The cells begin to restructure themselves as if they are a living thing inside Pong’s interface. As the cells play Pong, they start to believe they are the paddle.
“We often refer to them as living in the Matrix,” Kagan says. “When they are in the game, they believe they are the paddle.”
Vasileios Maroulas advances AI by studying human brain
The fascinating complexity of how the human brain recognizes itself in relationship to its environment—and the edge this understanding has over AI—is the central focus of Greek mathematician Vasileios Maroulas’ research.
Maroulas, who studied applied mathematics at the University of Athens in Greece, recently received recognition from the US Department of Defense (DOD) for his work in advancing artificial intelligence.
His current research explores the process by which the brain experiences space and time, closely analyzing electroencephalogram (EEG) that records the electrical activity inside the brain. Maroulas uses this data to recognize patterns in the brain connected to spatiotemporal processing.
“Let’s imagine a quarterback who is processing the information he needs to make a decision about where to throw the ball,” he states. “If you really look at the brain when there is high focus, you can see the activity increase in the information processing centers of the brain. I am trying to understand these patterns, to develop ways to embed them into an algorithm that can be programmed into a machine.”
Maroulas is attempting to give machines a more nuanced relationship to spatial awareness which mirrors our own.
“Humans understand space very well, based on a lifetime of experiences that allow us to create context,” said Maroulas, adding that “machines don’t have those experiences and have to start from scratch, scanning an area to gather information. That information can then be used to compare the machine’s previous experiences with other structures and other spaces.”