New research shows that when neurons get information about the changing world around them (task-relevant sensory input), it changes the way they behave, putting them in a limbo state so that tiny inputs can trigger an "avalanche" of brain activity, supporting a theory called the critical brain hypothesis.

Microscope image of a neural cell with fluorescent markers showing different types of cells. Green marks neurons and axons, purple marks neurons, red marks dendrites, and blue marks all cells. When multiple markers are present, the colors merge and often appear yellow or pink depending on the proportion of markers. Image credit: Cortical Labs

Researchers from CorticalLabs and the University of Melbourne used DishBrain, a collection of 800,000 human nerve cells, to learn to play table tennis. The research was recently published in the journal Nature Communications.

This is the strongest evidence yet supporting a controversial theory about how the human brain processes information. According to the critical brain hypothesis, large and complex behaviors are only possible when neurons are in a limbic state and small inputs can trigger an "avalanche" of brain activity. This finely balanced state is known as "neurocritical" and lies between two extremes: the uncontrolled excitement seen in diseases such as epilepsy, and the coma where signaling is stalled.

Dr. Brett Kagan, chief scientific officer of biotech startup CorticalLabs, said: "Not only does it show that the network reorganizes into a near-critical state when receiving structured information, but that reaching that state also leads to better task performance." The results were surprising and far exceeded our expectations. ”

This study adds an important piece to the puzzle of the critical brain hypothesis.

TAGP H19ForoughHabibollahi, first author of the study

Key findings and implications.

So far, there is little experimental evidence as to whether criticality is a general feature of biological neuronal networks or whether it is related to information load TAGPH27

Dr. Kagan said: “Our results show that near-critical network behavior occurs when a neural network performs a task that does not occur when it is not stimulated. ”

However, Dr. Kagan’s research shows that criticality alone is not enough to drive neural network learning. “Learning requires a feedback loop that provides the network with additional information about the consequences of an action,” he said. ”

Latest research highlights DishBrain’s potential to help unlock the secrets of the human brain and how it works, which is not possible in animal models.

First author Dr. Forough Habibollahi said: "Usually, in order to study the brain, especially at the neuronal scale, researchers have to use animal models, but there are many difficulties in doing so and the number of study subjects is limited, so when I saw that DishBrain was able to answer different types of questions in a way that others could not

Applications and Future Possibilities

Doctors also see great potential for this research to help discover treatments for serious brain diseases.

“The DishBrain Pivotal Project has been an amazing collaborative experience between cortical laboratories, biomedical engineering and neurology,” said paper author Dr. Chris French, head of the Neurodynamics Laboratory in the Department of Medicine at the University of Melbourne. “The key dynamics of DishBrain neurons could provide key biomarkers for diagnosing and treating a range of neurological diseases from epilepsy to dementia. "

By constructing living brain models, scientists will be able to conduct experiments using real brain function rather than flawed similar models such as computers, not only to explore brain function but also to test how drugs affect it.

Professor Anthony Burkitt, the author of the paper and head of the Department of Biosignals and Biological Systems at the University of Melbourne, said that this research may also solve the challenges faced by brain-computer interfaces and can restore functions lost due to nerve damage.

"A key feature of the next generation of neuroprostheses and brain-computer interfaces we are currently investigating involves utilizing real-time closed-loop strategies, so the results of this study may have important implications for understanding how these control and stimulation strategies interact with neural circuits in the brain. The field of biological brain modeling is in its infancy, but opens the way to an entirely new field of science," said Dr. Kagan.