Theoretical prediction of conditions for “sleep learning”

In the cerebral cortex, numerous neurons exchange information through junctions known as synapses. The strength of each synaptic connection changes depending on the activity levels of the neurons involved, and these changes are thought to form the basis of learning and memory. There are several established principles governing the relationship between neuronal activity patterns and changes in synaptic strength, referred to as "synaptic learning rules ⁽¹⁾ ". Although it is well known that sleep plays a crucial role in learning and memory, how synaptic connections are altered during sleep has remained unclear.

A research group led by Professor Hiroki Ueda of the Graduate School of Medicine, The University of Tokyo, has demonstrated that the strength of synaptic connections in the cerebral cortex during sleep changes depending on synaptic learning rules and the level of neuronal activity during sleep. They revealed that it was possible to theoretically predict the conditions under which "sleep learning ⁽²⁾ " may occur.

The researchers used computational simulations to reproduce the activity of neural networks composed of various types of interconnected neurons, and they investigated changes in synaptic connections during the neural activity observed in the sleep-wake states. These results showed that synaptic connections in the cerebral cortex are strengthened during sleep when specific levels of neural activity were accompanied by typical synaptic learning rules. This finding clarified the conditions under which synaptic strengthening can occur even during sleep, thereby enabling theoretical predictions of when "sleep learning" is possible.

Based on these predictions, these insights are expected to lead to a deeper understanding of the relationship between sleep, learning and memory. Moreover, they may contribute to elucidating the mechanisms of brain disorders associated with sleep disturbances, such as neuropsychiatric conditions.

These findings were published in the online version of the American scientific journal PLOS Biology on June 12, 2025.

This result was achieved in the Ueda Biological Timing Project, a research area of the Exploratory Research for Advanced Technology (ERATO) by the Japan Science and Technology Agency (JST). Under this project, JST pursues "systems biology for understanding humans" using the sleep-wake rhythm as a model system and aims to understand information on "biological time," which transcends from molecules to individual humans livingin society.

<Notes>

(*1) Synaptic learning rules

Rules that describe how the strength of synaptic connections between neurons changes depending on the timing and frequency of neural activity. Examples include Hebbian rule and spike-timing-dependent plasticity (STDP).

(*2) Sleep learning

The enhancement of memory and learning performance through the organization and integration of new information by the brain during sleep.

About the Japan Science and Technology Agency

Our world faces unprecedented global challenges — such as climate change, energy crises, and emerging infectious diseases — that demand innovative solutions. JST will rise to these challenges through “ Science and Technology, ” as a national research and development agency that plays a central role in implementing Japan’s science, technology, and innovation policy. We support fundamental research and startups to create new value, develop R&D strategies, foster the next generation of talent, disseminate vital information, and manage the Japan University Fund. Like a compass guiding ships through turbulent waters, JST will chart the way towards a vibrant and secure future by empowering science through a multifaceted approach.