Home Innovation Science and Technology Multiple Learning Rules Found ...

Multiple Learning Rules Found in Single Neurons

Science and Technology

Business Honor
22 April, 2025

UC San Diego study reveals neurons follow multiple learning rules, unlocking new frontiers in AI and medicine.

Historic NIH-funded research by neurobiologists at the University of California, San Diego, has uncovered that neurons apply multiple learning rules at once—a breakthrough that has the potential to transform brain-inspired artificial intelligence (AI) systems and improve treatments for neurological diseases.

Published April 17 in Science, the research team of William "Jake" Wright, Nathan Hedrick, and senior author Takaki Komiyama employed cutting-edge two-photon imaging to image real-time dynamics in the mouse brain during learning. The advanced method enabled them to visualize isolated synaptic connections on neuron dendrites with unprecedented clarity.

Their observations challenge a long-standing assumption in neuroscience—that synaptic plasticity operates under uniform rules across the brain. Instead, the study shows that neurons apply different synaptic rules in different compartments, reshaping our understanding of how the brain learns and adapts.

This completely alters our understanding of how learning is structured in the brain," said Komiyama. "Neurons that are doing different computations in parallel can assist in solving the credit assignment problem, a long-standing enigma in neuroscience and AI.

The "credit assignment problem" is the difficulty of identifying what precise synaptic modifications are responsible for learned behavior since synapses have access only to local information. It is similar to how single ants are working for the colony without being aware of its final purpose.

The discovery could open the door to the next generation of AI systems, moving away from monolithic neural network structures toward more advanced models based on the brain's compartmentalized plasticity.

Aside from technology, the health implications are enormous. A clearer grasp of synaptic modification can result in revolutionary advances in the treatment of Alzheimer's, PTSD, addiction, and autism.

In continuing to investigate how neurons exploit various learning rules, this research represents a significant advance toward cracking the brain's learning machinery—providing insights with scientific and social relevance.