Researchers have been cultivating neurons on microelectrode arrays (MEAs) and studying how these living neural cells form networks and respond to electrical stimuli since the late 1990s. At the time, it was purely basic science, lacking any practical computational application.
Throughout the 2000s, cells on MEA plates began responding to stimuli and forming activity patterns, leading some researchers to “condition” these cultivated neural networks to perform certain behaviors through repeated stimulation.
A significant leap forward came in 2021 when Australian company Cortical Labs conducted the DishBrain experiment. The goal was for the neurons to not only respond to a fixed stimulus, but also to receive feedback on the outcome of their actions. The platform chosen for this was the 1972 video game “Pong.”
More recently, Cortical Labs created the CL-1 — a hybrid biological computer — in a demonstration that substantially increased the field’s credibility. Using approximately 200,000 living human brain cells, grown on a microchip, the system successfully played the classic first-person shooter (FPS) game, “Doom,” a cultural phenomenon from 1993.
From “Pong” to “Doom”: Firing Up Neurons
The evolution from “Pong,” one of the simplest games computationally, to “Doom” represents a genuinely complex cognitive challenge. The game’s three-dimensional environments and constant, unpredictable encounters with enemies pose a significant hurdle for 200,000 neurons grown on a plate, without eyes, a nervous system, or a body.
To make this confrontation viable, despite the lack of any evolutionary context for the brain cells, researchers had to translate the digital world of the video game into the natural language of biology: electricity. The cell culture was positioned on a plate with multiple electrodes, allowing for two-way communication between the machine and the living tissue.
The dynamics are impressive. When an Imp creature appears on the left side of the screen, the electrodes stimulate that specific region of the neural culture, causing the neurons to react and fire electrical signals back. If the system recognizes this firing pattern, the character shoots or moves quickly. This research demonstrates the potential of combining biological and artificial systems for advanced computing.
What’s most fascinating about the entire operation is that no one explicitly programmed the neurons to associate a specific pattern with shooting or moving the character. The cells themselves learned to make this association through feedback, spontaneously developing functional patterns useful for interacting with the game environment.
Even while performing as a functional newbie (an inexperienced player who can still perform their role) and losing matches, the system reached its current level of performance much faster than various traditional silicon-based artificial learning systems. AI systems trained to play games, for example, typically require millions of simulated matches to achieve comparable performance.
Learning: An Exit from Hell?
Having real neurons play “Doom” is an experience that goes beyond mere entertainment or scientific curiosity. It’s a real proof that the hybrid organic technology present in the CL-1 is viable, uniting the power of the human brain — energy efficiency, plasticity, rapid learning — with the best of silicon: processing speed and precision.
Just as the protagonist of Doom needs to survive and locate an exit, traversing the chaos of hell, Cortical Labs’ research seeks an exit from the limitations of silicon. If conventional computing is experiencing its own “hell” of energy efficiency and learning speed, wetware (the fusion of brain tissue with hardware) could be an escape portal.
Progressing through levels here doesn’t mean combat skill, but demonstrating the evolution of the cognitive process, which, in this case, is adaptive learning in real time directed towards concrete objectives. In other words, it’s an elegant and effective way to understand how biological systems react and reorganize under stimulus.
The system functions as a continuous and accelerated conversation between the game and the cells — the game speaks, the cells respond, the system interprets this response as action, the action changes the game, and the cycle restarts. With each round, the neurons reorganize progressively to respond more efficiently — what we call learning.
