Neurons Break Rules During Learning Process

Brain Learns Through Multiple Rules Simultaneously

Scientists have upended conventional wisdom about how we learn, revealing that individual neurons in our brains don’t follow one uniform set of rules as previously thought. This breakthrough discovery shows that different parts of the same neuron simultaneously follow distinct rules when forming memories, potentially transforming our understanding of learning disorders and artificial intelligence.

Using sophisticated brain imaging techniques, University of California San Diego researchers visualized individual synapses—the connection points between neurons—as learning took place in real time. Their findings, published April 17 in the journal Science, reveal a far more complex and nuanced learning process than previously understood.

“When people talk about synaptic plasticity, it’s typically regarded as uniform within the brain,” said William “Jake” Wright, a postdoctoral scholar and lead author of the study. “Our research provides a clearer understanding of how synapses are being modified during learning, with potentially important health implications since many diseases in the brain involve some form of synaptic dysfunction.”

Learning occurs as our brain’s neural networks adapt, with some connections strengthening while others weaken—a process known as synaptic plasticity. While scientists have long studied the molecular mechanisms behind these changes, they’ve struggled to understand why specific synapses are selected for modification while others remain unchanged.

The researchers employed cutting-edge two-photon imaging to observe mice brains during learning tasks, allowing them to track changes at the level of individual synapses. What they discovered challenges fundamental assumptions about neural processing.

“This discovery fundamentally changes the way we understand how the brain solves the credit assignment problem, with the concept that individual neurons perform distinct computations in parallel in different subcellular compartments,” explained senior author Takaki Komiyama, a professor with appointments across multiple departments at UC San Diego.

The “credit assignment problem” refers to how individual synapses—which only have access to local information—collectively produce complex learned behaviors. It’s similar to how individual ants perform specific tasks without understanding the colony’s broader goals.

This discovery could transform artificial intelligence development. Current AI neural networks typically operate on uniform plasticity rules, but implementing multiple rules within single units might enable more sophisticated systems that better mimic human learning.

The findings also offer promising pathways for treating neurological and psychiatric conditions.

“This work is laying a potential foundation of trying to understand how the brain normally works to allow us to better understand what’s going wrong in these different diseases,” Wright noted, mentioning potential applications for addiction, PTSD, Alzheimer’s disease, and autism.

The research team, funded primarily by the National Institutes of Health, is now investigating how neurons manage to utilize different rules simultaneously and what advantages this multi-rule approach provides.