Deep in the mouse brain, a cellular switchboard is quietly managing one of neuroscience's greatest mysteries: how we learn new things without erasing what we already know. Researchers at NYU Langone Health discovered that roughly 1 in 4 memory cells in the hippocampus—the brain region that organizes fresh experiences into lasting memories—acts as a shared hub, cleverly reusing the same neurons to store multiple memories while keeping them perfectly separate. The findings, published in Nature, reveal a neural architecture as elegant as it is efficient.
Scientists have long puzzled over how the brain can be flexible enough to absorb new information while remaining stable enough to preserve old knowledge. This study offers a compelling answer. The team, led by postdoctoral fellows Joaquín Gonzalez and Mihály Vöröslakos, along with professor Zhe S. Chen, focused on three connected brain regions: the cornu ammonis 3 (CA3), which receives incoming signals; the cornu ammonis 1 (CA1), which serves as the central hub; and the retrosplenial cortex, which stores long-term information and helps with navigation and scene reconstruction.
The breakthrough came through meticulous observation. When the researchers trained six mice to run back and forth on a straight track for water rewards, they used high-density electrodes to record activity from hundreds of individual neurons simultaneously—a technical feat that hadn't been possible before. They discovered something remarkable: the minority of CA1 neurons that carry most incoming messages from CA3 fire in one pattern when receiving signals, then switch to a completely different firing pattern when sending signals outward to the retrosplenial cortex. It's as if the same telephone lines carry incoming and outgoing calls on separate frequencies, preventing any crossed wires.
"By changing how the same cells fire together instead of turning on new cells, the brain can keep information organized and protect older memories," Gonzalez explained. This elegant solution solves a fundamental challenge: the brain doesn't need to constantly recruit new neurons to store new information. Instead, it simply reorganizes how existing cells communicate, much like an electronic switchboard routing multiple calls without tangling the lines.
The discovery extends beyond waking hours. During sleep, those same CA1 neurons remain active during sharp-wave ripples—brain events known to consolidate memories. Because these core cells handle both daytime processing and nighttime replay, the pathway between hippocampus and cortex stays open, allowing memories to solidify and integrate into long-term storage. This means learning and memory consolidation don't compete for resources; they coexist in the same network.
Vöröslakos highlighted the significance of their methodology: "Our discovery was made possible because, for the first time, we were able to record hundreds of individual neurons across all the key regions simultaneously in animals that were moving around naturally." This naturalistic recording—capturing brain activity while mice behaved freely rather than in constrained conditions—made all the difference.
The implications ripple outward. As Chen noted, understanding how memory circuits function at this level may illuminate why these circuits fail in Alzheimer's disease and other neurodegenerative conditions. The research provides a biological blueprint that could even inform artificial intelligence design, showing how systems can be both flexible and stable. In explaining how the brain threads the needle between change and continuity, researchers have illuminated a fundamental principle of learning itself.