Researchers have developed a novel hybrid system that combines living human neurons with conventional computing hardware, enabling the biological component to control video games in real time. This breakthrough, achieved by the Australian start-up Cortical Labs, offers a new platform for observing neuronal activity and studying brain function outside the human body.

The device, known as the CL1, is a shoebox-sized unit containing hundreds of thousands of neurons grown from reprogrammed human skin cells. These cells undergo a process that transforms them into induced pluripotent stem cells, which are then differentiated into electrically active neurons. The neurons are cultured on a grid of microscopic electrodes that both stimulate and record their electrical signals, creating a closed-loop system where input signals influence neuronal responses and vice versa. This feedback loop allows the neurons to adapt their activity and effectively “learn” behaviors through trial and error.

Cortical Labs has demonstrated the system's capabilities by teaching the cultured neurons to play classic 1990s video games such as Doom and Pong. Gameplay actions—like moving and shooting in Doom—are translated into electrical pulses sent to the neurons, while their electrical outputs are decoded into game commands. Although the neurons’ performance in these games is rudimentary, the primary objective is not game mastery but rather providing a window into real-time neuronal processing and learning.

The CL1 unit maintains the neurons in a nutrient-rich environment, supporting their viability for up to six months before replacement is required. The device costs approximately $35,000 but is also accessible via internet rental for researchers worldwide.

Neuroscientist Sven Truckenbrodt of the MRC Laboratory of Molecular Biology in Cambridge, who is not affiliated with Cortical Labs, praised the demonstration as “super impressive,” noting the potential for this technology to illuminate the underlying mechanisms of neurological conditions such as schizophrenia, depression, ADHD, and autism. These disorders are thought to involve dysfunctional neural connections, and the ability to monitor how neuronal networks form and change outside the body may lead to important insights.

Beyond medical research, the biological-computing hybrid raises questions about future applications, including the development of computational systems that leverage the brain’s energy efficiency. Whereas current artificial intelligence requires substantial electrical power, the human brain operates on roughly 20 watts, suggesting potential for more efficient bio-computers.

The technology also prompts ethical considerations. The closer these neural clusters resemble actual brains, the more challenging it becomes to define their moral status. While the neuronal networks in the CL1 remain significantly simpler than even an insect brain, and no evidence suggests any form of consciousness, Cortical Labs is collaborating with ethicists to establish guidelines regarding the limits of such experimentation and to identify any potential ethical boundaries.

As this emerging field develops, it promises to reshape neuroscience research, offering new avenues to explore brain function, disease, and the interface between biology and computing.