In a groundbreaking experiment, scientists have taught human brain cells on a microchip to play the iconic video game, Doom. But wait, how is this possible? Well, it's all thanks to the innovative minds at Cortical Labs, an Australian biotech company, who have pushed the boundaries of computing and neuroscience.
The team developed the CL1 biological computer, a remarkable system housing a staggering 200,00—yes, you read that right—living human neurons on a microchip. This microchip, known as a multi-electrode array, is the key to this fascinating experiment. It enables the neurons to receive electrical signals from a computer and, in turn, send responses as control inputs for the game.
But here's where it gets truly mind-boggling: these human neurons were trained to control the gameplay of Doom. Building on their previous success with lab-grown neurons playing Pong, the researchers tackled the more intricate 3D environments and enemy interactions of Doom. And this is the part most people miss—the researchers had to translate the digital world of Doom into the language of neurons, which is electricity.
The system works by converting gameplay into electrical stimulation patterns, which are then delivered to various parts of the neural culture. When enemies approach, specific electrodes stimulate the corresponding neuron regions, triggering electrical signals that are interpreted as movement or shooting commands. It's like teaching a beginner gamer how to play, but with brain cells!
While the cells demonstrate basic learning and adaptation, their performance is still limited. The researchers humorously noted, 'It's definitely not an esports champion.'
Doom, released in 1993, has been ported to numerous devices, and its versatility is well-known. Interestingly, researchers have even simulated Doom using gut bacteria, showcasing the game's adaptability. But the question remains: what does this experiment mean for the future of computing and our understanding of the human brain? Could this be a step towards creating more advanced AI, or is it a fascinating yet isolated achievement? Share your thoughts in the comments below!
