Sleep Enhances Memory by Reorganizing Neuronal Activity Patterns

A team of neuroscientists from the Institute of Science and Technology Austria (ISTA) has uncovered how brain activity during sleep helps solidify spatial memories. Using extended wireless recordings of rat hippocampal neurons for up to 20 hours following a spatial learning task, the researchers identified a dynamic reorganization of neuronal activity patterns that mirror those seen during waking memory recall. The findings were published in Neuron.

Memory representations shift during sleep

The hippocampus is essential for spatial learning and navigation, forming internal maps by activating specific neurons at given locations. In rats trained to locate food rewards in a maze, neuronal patterns encoding these rewards were reactivated during subsequent sleep. Over the course of extended non-rapid eye movement (non-REM) sleep, these patterns changed: some neurons stopped firing while new ones began to activate, forming a new configuration of the same memory trace.

These reorganized patterns closely resembled the neuronal activity seen when rats later successfully recalled the reward locations. This evolution of neural representations suggests a memory optimization process during sleep, which may reduce the number of neurons needed to store a specific memory.

“We showed that the neuronal assemblies in the early stages of sleep reflect recently learned spatial memories. However, as sleep progresses, neuronal activity patterns gradually transform into those seen later, when the rats awaken and remember the locations of their food rewards.”

Dr. Jozsef Csicsvari.

Reward-related neurons become more efficient

Earlier studies by the Csicsvari group had shown that the reactivation of specific reward locations during sleep improved memory performance. In the current research, wireless neural recordings enabled a much longer observation window than previous work, capturing reactivation patterns across full sleep cycles.

The data revealed that a “stable subgroup” of neurons continued to represent the reward locations throughout sleep, while a different set of neurons gradually replaced others. This transformation was not random. Instead, it aligned with memory recall activity observed after the rats woke up.

Non-REM sleep drives representational drift

The reorganization occurred primarily during non-REM sleep. In contrast, rapid eye movement (REM) sleep appeared to counteract the changes, maintaining previous patterns. The gradual shift in neuronal representation, known as representational drift, may play a role in freeing up neuronal capacity for future memories.

“It is possible that memory representations must be formed quickly during learning but that such representations are not optimal for long-term storage. Therefore, a process may take place in sleep that optimizes these representations in sleep to reduce brain resources to store a specific memory.”

Dr. Jozsef Csicsvari.

The reduction in neurons assigned to a given memory after sleep supports the idea that these representations are streamlined for more efficient storage. This adaptation may also help new memories integrate into existing networks by partially updating the underlying neural code.

Reference: Bollmann L, Baracskay P, Stella F, Csicsvari J. Sleep stages antagonistically modulate reactivation drift. Neuron. doi: 10.1016/j.neuron.2025.02.025

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.