Nyu scientists identify memory switchboard that preserves older memories

Thursday 04 - 07:57

By: Dakir Madiha

Nyu scientists identify memory switchboard that preserves older memories

Researchers at NYU Langone Health have identified a small group of neurons in the hippocampus that acts as a biological switchboard, enabling the brain to create new memories without disrupting those already stored. The findings, published in Nature on May 13, 2026, offer a new explanation for a long-standing neuroscience question: how the brain remains flexible enough to learn while maintaining stable access to existing memories.

The study found that roughly 25% of neurons in the hippocampus's CA1 region function as shared communication hubs. These cells receive rapid streams of information from the neighboring CA3 region and relay them to the retrosplenial cortex, an area linked to spatial navigation and memory retrieval. Researchers observed that the hub neurons use different activation patterns for incoming and outgoing signals, effectively creating separate communication channels within the same neural structure. This arrangement allows multiple streams of information to pass through simultaneously without interference.

According to the research team, the brain does not need to recruit entirely new neurons for every experience. Instead, it adjusts activation patterns within a stable core network of cells. This mechanism enables efficient organization of information while protecting memories that have already been encoded. The discovery suggests that memory storage depends not only on individual neurons but also on how networks of cells coordinate and route information.

The same CA1 hub neurons were also found to remain active during sleep. They participated in sharp-wave ripple events, bursts of neural activity long associated with memory consolidation. Because the same group of cells supports both daytime information processing and nighttime memory replay, the pathway linking the hippocampus and cortex can remain continuously engaged, strengthening long-term memory formation.

The research involved six mice trained to navigate a linear track while high-density electrode arrays recorded activity from hundreds of neurons across interconnected brain regions. The ability to monitor multiple areas simultaneously allowed scientists to trace how information moved through the memory network and identify the specialized role of the CA1 hub neurons.

The findings could have important implications for understanding neurodegenerative diseases. Researchers believe the newly identified memory switchboard may provide clues about the earliest stages of memory failure in conditions such as Alzheimer's disease. Disruptions within this communication network could help explain why memory problems emerge before widespread brain damage becomes apparent.

The study may also influence the development of artificial intelligence systems. Many AI models experience "catastrophic forgetting," a problem in which learning new tasks causes the loss of previously acquired knowledge. By revealing how the brain preserves older memories while continuously incorporating new information, the research could inspire AI architectures that learn more efficiently without sacrificing past knowledge.