Memory weaves itself into our language through countless idioms. You can clear it, jog it, or have it etched in your mind. You might take a stroll down memory lane or lose your train of thought, capture a moment with a mental picture or simply let something slip your mind. Some of us have a memory like a sieve, while others are said to remember like an elephant, or have a mind like a steel trap.
We have so much language for describing memory—much of it contradictory—because what and how we remember is so central to our identity and sense of self. Our recollections connect us to our past and influence how we move forward. Yet, our experience of memory is anything but consistent; sometimes it feels sharp and vivid, and at others it’s murky or unreliable.
Now a team of researchers led by Flavio Donato, a neurobiologist at the University of Basel, has discovered an intricate system of parallel memory storage in the mouse brain that could help explain how our recollections remain both stable and adaptable over time. Their findings, published in Science, suggest that memories are left in multiple traces—or patterns of neuronal activity—each on a different population of neurons, and each with its own trajectory and purpose.
These distinct groups of neurons form at different stages of the brain’s development and are found in the hippocampus—the seahorse-shaped structure embedded deep in the temporal lobe that plays a critical role in memory processing. One set emerges early in embryonic development and is responsible for the stability of long-term memories. The researchers call these “early-born” neurons, and they are necessary for retrieval of a memory at longer delays. Neurons that emerge later in embryonic development, which they call “late-born,” are primarily used to retrieve a memory during a short window of time right after the event. A third set, which arises during an intermediate stage of embryonic development, is relied upon for recollection of a memory at both recent and remote time points, and may help provide a kind of continuity to memories. Whether these populations of neurons are just three of many, or exist on a kind of gradient, is not yet clear.
“I’ve always been fascinated by memories,” says Donato. “How this type of knowledge allows us to build our expectations about the future, to define our behavior.” Donato says he became hooked on memory research after reading the work of 17th-century English philosopher and physician John Locke and coming away with the sense that memories form the essence of who we are.
A single memory is left in multiple traces, each on a different population of neurons.
In their study, Donato and his colleagues find that which copy is accessed when one is remembering determines, in part, how easily that memory can be modified or used to form new associations. Recalling an event soon after it occurs generally draws on those neurons that emerge late in development, which store the more malleable trace of the memory—which means that as we remember it, we can layer onto that memory trace associations with related events and ideas, and other new information. For example, you may learn to associate your first memory of a room with more recent experiences in that room, such as of a bad smell or painful accident. The different neuron populations allow us to preserve fundamental aspects of a memory over the long term, while also enabling us to adapt and incorporate new or related information we have learned about the world.
“On one hand, memories are these things that we believe are very crystallized—we refer to the idea of taking a mental picture,” explains Donato. “But actually, research has shown that more than a picture it’s like a painting that changes with time. Every time you look at it, it is slightly different.”
Using fear memories in mice as their model, the researchers tagged neurons to observe their activation in memory formation. They found that the early-born and late-born neurons had opposing dynamics. The late-born neurons were highly activated when recent memories were recalled, but that activity soon faded while activity in the early-born neurons ramped up over time. The authors measured the activity level of the neurons primarily by tracking calcium transients, or brief increases in the concentration of calcium inside cells that are indicative of neuron firing. Tracking activity in the different neuron populations revealed that the early-born neurons were relatively rigid in their behavior, showing stable and consistent activity patterns that did not change significantly in response to external events. The late-born neurons were more plastic: their activity patterns and connections to one another fluctuated when the mice learned something new that was associated with a particular memory.
The researchers also performed a series of behavioral experiments with the mice, creating associations between a lab-controlled environment and a mildly unpleasant stimulus and testing how well the mice remembered the unpleasant experience through their later reactions to the environment. When they artificially stimulated the late-born neurons, the researchers were able to extend the period of memory plasticity, allowing the memory to be updated or altered by new experiences. Conversely, when they inhibited the activity of these late-born neurons shortly after the experience, the memory became less flexible and more resistant to change.
“This led us to the conclusion that not only was the activation of this subpopulation of plastic neurons indeed correlated to the plasticity of the memory, but also we could extend this plasticity if we could force this neuron to be recruited into the memory trace,” says Donato. They also found that activation of the late-born neurons shortly after the event was necessary for the long-term permanence of the memory.
Future research could lead to potential therapeutic applications—for example, in treating trauma or PTSD by restoring plasticity in memories when these memories become fixed and pervasive. Scientists might also attempt to make a memory more rigid and fixed if it were too plastic, says Donato, as in the case of memory disorders like dementia.
The findings help to explain how our recollections balance the old and new and how our stories of ourselves are woven into our neurobiology. 
Lead image: N Universe / Shutterstock
Get the Nautilus newsletter
Cutting-edge science, unraveled by the very brightest living thinkers.
