A recent study published in Science has identified a previously unknown mechanism in the brain that occurs during sleep, helping to reset memory pathways. Researchers found that a burst of neural silence in a specific part of the brain, the hippocampus, allows neurons involved in memory to reset and prepare for new learning the following day. This phenomenon, termed a “barrage of action potentials” or “BARR,” allows neurons to reset, ensuring that our brains can continue storing new information without overwhelming the neural networks responsible for memory.
Despite years of research into sleep and memory, we still have much to learn about how the brain manages memory consolidation during sleep. Scientists have long known that sleep plays a key role in maintaining physical and mental health. For instance, sleep helps restore energy levels, supports immune function, and is vital for emotional regulation.
But it also plays a crucial role in memory formation. When we experience something new, neurons in the hippocampus—an area of the brain responsible for learning and memory—become active and encode that information. During sleep, those same neurons repeat the patterns of activity observed during the day. This process, called memory replay, strengthens the neural connections involved in those experiences, helping to solidify the memory.
However, a major unanswered question has been: how does the brain continue learning and storing new information every day without running out of available neurons? If the same neurons are used repeatedly for new memories, wouldn’t they eventually become “full” or overloaded? This study was designed to explore this very issue, investigating how the brain prevents memory circuits from being overwhelmed by constant learning.
“We expend about a third of our live sleeping, yet, we don’t know which processes are implemented during this time, in our body but also in our brain, that makes this function so fundamental for our health and in extreme cases, for our survival,” said study author Azahara Oliva, an assistant professor of neurobiology and behavior at Cornell University
The research team studied the brain activity of mice to better understand how the hippocampus balances memory storage during sleep. Mice are frequently used in memory studies because their brain structure, particularly the hippocampus, is similar to humans.
The researchers used a combination of advanced techniques, including electrode implants and optogenetics, to monitor and manipulate neuronal activity in the hippocampus. The electrodes allowed the team to track specific brain regions, while optogenetics—a method that uses light to control genetically modified neurons—enabled them to selectively activate or silence particular groups of neurons during sleep.
The hippocampus is divided into three main regions: CA1, CA2, and CA3. CA1 and CA3 are well-known for their roles in processing memories, but the role of CA2 has been less understood. To investigate this, the researchers implanted tiny electrodes in the mice’s hippocampi, specifically targeting the CA1, CA2, and CA3 regions. These electrodes recorded the firing patterns of neurons during both awake and sleep states.
The mice were first exposed to various memory tasks, including an object displacement task, where they had to remember the location of objects, and a social memory task, which involved recognizing other mice. After these tasks, the researchers monitored the brain activity during sleep to see how memories were being consolidated.
A key feature of their experiment was the ability to observe and record sharp-wave ripples (SWRs)—brief bursts of brain activity that occur in the hippocampus during deep sleep. SWRs are thought to be crucial for replaying memories and strengthening neural connections. However, the team also noticed a new type of brain event during sleep in the CA2 region: the BARR. Unlike SWRs, which involve a high level of synchronized firing across many neurons, BARR events were characterized by the silencing of certain neurons, particularly in the CA2 region, which temporarily halted their activity.
By using optogenetic tools, the researchers were able to disrupt these BARR events in real time, allowing them to study how this silencing affected memory consolidation. They could precisely control when to silence or activate neurons and measure the resulting impact on the mice’s memory performance.
The most significant discovery was the identification of the BARR event, which acts as a kind of “reset button” for neurons in the hippocampus. During sleep, when the brain is replaying memories through SWRs, certain neurons in the CA2 region switch off. This period of silence allows the neurons that were heavily used during learning to reset, preparing them for new learning the next day.
The researchers believe that without this reset mechanism, the hippocampus would quickly become overwhelmed by repeated use of the same neurons for storing memories. The BARR events give the brain a way to reuse neurons for new tasks without compromising their ability to encode new information.
“The main message in our study is that, while it is known that sleep is necessary for our recent experiences to get incorporated into long term memories, we just discovered that our brain also reset our memories, it prepares the system for new memories to be incorporated the next day, so our neural circuits do not saturate and keep working optimally,” Oliva told PsyPost.
In the study, when the BARR events were disrupted using optogenetics, the mice showed significant impairment in their ability to recall memories from the earlier tasks. This was particularly evident in the object displacement task, where the mice struggled to remember which objects had been moved.
Interestingly, the disruption of BARR events did not seem to affect the overall rate of sharp-wave ripples, suggesting that while SWRs are important for replaying and strengthening memories, BARRs play a separate but equally critical role in preventing memory overload.
Further analysis showed that the neurons most involved in encoding new experiences—those that fired the most during learning—were the same neurons that were silenced during BARR events. This indicates that the brain selectively turns off the most active neurons during sleep to avoid overloading them. The balance between SWR-driven reactivation and BARR-driven silencing appears to be essential for maintaining healthy memory circuits.
“We were very surprised to find that this phenomenon of memory resetting is an active process: if we block memory resetting by silencing the neurons responsible for it, the memory doesn’t get consolidated properly,” Oliva said. “This means that memory is a two-fold process, with neural circuits that enhance the consolidation of a given experience and neural circuits that control that this consolidation doesn’t go over a healthy limit.”
While this study offers new insights into how the brain manages memory during sleep, there are still several unanswered questions. One limitation is that the study was conducted in mice, which have brain structures similar to humans but not identical. Further research will be needed to confirm whether these findings apply directly to human memory consolidation.
Another limitation is that the researchers focused primarily on the hippocampus. Although the hippocampus is crucial for memory, other brain regions, such as the cortex, also play a significant role in storing long-term memories. It remains unclear how the resetting mechanisms in the hippocampus interact with these other brain areas.
Looking ahead, the researchers plan to investigate several key questions: “How the brain knows which neurons are to be consolidated and reset? What is the marker? How much can we boost memory without saturating the brain? And can we use the reset mechanisms to erase unwanted memories? We don’t know this yet,” Oliva explained.
“Our long term goal is to understand what are the physiological markers for the different cognitive processes that benefit from a good sleep and how are they regulated from the neural circuit to the molecular level.”
The study, “A hippocampal circuit mechanism to balance memory reactivation during sleep,” was authored by Lindsay A. Karaba, Heath L. Robinson, Ryan E. Harvey, Weiwei Chen, Antonio Fernandez-Ruiz, and Azahara Oliva.
