Neurons Use Multiple Plasticity Rules for Learning Behavior

Neurobiologists at the University of California San Diego have identified new mechanisms by which individual neurons in the brain adjust their connections during learning. Using high-resolution imaging techniques in mice, the researchers found that different parts of a single neuron can follow separate sets of rules to modify their synapses, the points where neurons communicate with one another.

This finding challenges the prevailing view that neurons follow a uniform rule set when adapting to new information. The results, published April 17 in Science, suggest that the brain’s learning process is more complex and flexible than previously understood.

Zooming in on learning

The researchers used two-photon microscopy, a high-resolution brain imaging tool, to observe how neurons in mice adapted as the animals learned a new behavior. They focused on synaptic plasticity – the ability of synapses to strengthen or weaken in response to activity – which is critical for encoding new information.

They discovered that different branches of a neuron’s dendritic tree, the part of the cell that receives input, could follow different plasticity rules at the same time. This was unexpected, as synaptic plasticity has typically been considered a process governed by a consistent rule across each neuron.

“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.”

Jake Wright

Addressing the credit assignment problem

One major question in neuroscience is how local changes in synapses contribute to overall learning – a puzzle referred to as the “credit assignment problem.” Much like individual ants contributing to the success of an entire colony without understanding the colony’s goal, synapses act based on local signals, yet their changes lead to coordinated behavior.

This study provides evidence that neurons may solve this problem by dividing computational labor across different compartments, each applying its own set of learning rules. This parallel processing allows neurons to respond more precisely to diverse input patterns.

Implications for artificial intelligence and brain health

The findings have potential implications for artificial intelligence (AI), where most models rely on consistent plasticity rules throughout the network. These results suggest that incorporating multiple plasticity rules within single units of a neural network might improve AI performance.

From a health perspective, the research could inform studies into brain disorders such as addiction, Alzheimer’s disease, post-traumatic stress disorder and autism, which involve disruptions to synaptic plasticity. Understanding how neurons normally assign “credit” during learning may help uncover how this process goes awry in disease.

Reference: Wright WJ, Hedrick NG, Komiyama T. Distinct synaptic plasticity rules operate across dendritic compartments in vivo during learning. Science. 2025;388(6744):322-328. doi: 10.1126/science.ads4706

This article has been republished from the following materials. Note: material may have been edited for length and content. For further information, please contact the cited source. Our press release publishing policy can be accessed here.

This content includes text that has been generated with the assistance of AI. Technology Networks' AI policy can be found here.