0
0
A groundbreaking study by researchers at Cornell University has unveiled how the size of the pupil during sleep can offer crucial insights into how the brain processes and consolidates memories. Published in Nature, the study reveals that the pupil plays a pivotal role in memory formation, shedding light on the brain’s sophisticated mechanism for organizing memories and preventing “catastrophic forgetting.”
The research team, led by Assistant Professors Azahara Oliva and Antonio Fernandez-Ruiz, conducted an experiment with mice equipped with brain electrodes and eye-tracking cameras. Over the course of a month, the mice were taught various tasks, such as collecting water or cookie rewards in a maze. Once they learned a task, their brain activity and pupil changes were recorded during sleep.
The findings showed that when the mice entered a substage of non-REM sleep, their pupils contracted, signaling the consolidation of newly learned tasks. On the other hand, during periods when the pupils were dilated, older memories were replayed and integrated. This distinction is key to understanding how the brain prevents one memory from interfering with another—a phenomenon that can result in “catastrophic forgetting” when one memory wipes out another.
“This discovery suggests that the brain has a micro-structure that separates the consolidation of new memories from the replay of older ones,” said Oliva. “These processes occur in very short, undetectable bursts—on the scale of 100 milliseconds—but they allow the brain to organize memories in a way that prevents interference.”
The study also uncovered that the sleep stages of mice are more complex and human-like than previously believed. By interrupting the sleep at various points and assessing the mice’s ability to recall their learned tasks, the researchers were able to map the exact moments when memories were being consolidated.
“This research highlights a previously unknown mechanism that allows the brain to safeguard the integrity of memories by distinguishing between new and old information during sleep,” said Fernandez-Ruiz.
These findings could have significant implications for memory enhancement techniques in humans and may offer insights for computer scientists working on artificial neural networks, helping them design more efficient systems for information processing.
The study, titled Sleep Micro-Structure Organizes Memory Replay, can be accessed in Nature (2024), DOI: 10.1038/s41586-024-08340-w.
