Scientists have puzzled over the brain’s memory storage system for decades. Many have wondered how our minds can file away new information without wiping out what we already know.
After rigorous investigations, researchers have found that place cells hold essential clues to this process.
Ila Fiete, a member of MIT’s McGovern Institute for Brain Research and the senior author of a recent study, believes these discoveries open fresh perspectives on how internal “maps” guide us to recall events with ease.
Place cells were first reported about half a century ago. Early evidence showed that they become active when a rodent occupies a particular spot in a maze.
Further work revealed these cells reside in the hippocampus, a core region for memory. They align with grid cells in the entorhinal cortex.
Together, these populations form a foundation. Both spatial details and personal experiences rest on this scaffold, ensuring that details remain anchored to distinct neural patterns.
The recent model, called vector hippocampal scaffolded heteroassociative memory (Vector-HaSH), proposes that the hippocampus functions a bit like an index.
It holds pointers, which direct the brain to the finer content stored in vast areas of sensory cortex. That index can then “fill in” previously stored information whenever we recall it.
“This model is a first-draft model of the entorhinal-hippocampal episodic memory circuit. That’s the thing I’m really excited about,” said Fiete.
In simpler words, the hippocampus does not permanently hold every detail. It directs us to the right place in the cortex, where the nuts and bolts of each stored memory lie.
The hippocampus is tucked deep within the medial temporal lobe. Because of it’s crucial role in converting short-term memories into long-term ones, without the hippocampus, experiences would simply fade away moments after they occurred.
This brain region also helps us navigate space, build mental maps, and orient ourselves in our environments.
It works closely with the amygdala and prefrontal cortex, forming a network that governs how we learn from experience and respond emotionally to events.
Damage to the hippocampus can lead to serious memory issues, such as anterograde amnesia – the inability to form new memories.
This kind of impairment has been observed in patients with Alzheimer’s disease and in cases of brain trauma or stroke.
Research also shows that the hippocampus is highly plastic – it can grow new neurons even in adulthood, a process known as neurogenesis.
Grid cells create repeated patterns that look like triangular lattices. Each triangular arrangement, referred to as a well, can point to a specific episode.
The content itself is not housed in these wells. Instead, each well is a label that refers the brain to the correct memory storage location in the sensory cortex.
As more events pour in, the circuit spreads them out, allowing older memories to fade gently. The model bypasses the so-called memory cliff, a snag that often affects older computational frameworks.
This design reflects how real neural networks accommodate both new items and older recollections without hitting a harsh limit.
Training for events that require memorizing large sets of information can benefit from these findings. The time-honored memory palace method aligns with the notion of using spatial backdrops to link to new data.
Experts often picture a familiar location and place items there in a structured sequence. Later, they mentally revisit each spot to retrieve those items.
This new model suggests that tapping an already entrenched memory storage map can free up cognitive space for large-scale memorization.
The principle is straightforward. By designating specific neural wells as placeholders, we can link a chain of events without jumbling them up.
Our internal index system calls up the correct cluster of cells. That cluster cues the finer content that is stashed in sensory regions.
Sometimes, we need to recall a story in the exact order it happened. This framework highlights that each well holds cues about what comes next.
That helps us replay those experiences in the correct sequence. The model draws on the same arrangement for both location-based and experience-based details.
Memories about where you ate lunch last week or how you maneuvered a tricky hiking trail can emerge from the same scaffolding. You tap the index that leads you to the right well, and the rest naturally falls into place.
One area of interest is how short-term recollections solidify into more general facts. These are sometimes called semantic memories.
The fresh approach provides a structure for investigating how personal events might become separated from contextual details after enough time.
Researchers also want to clarify how episodes are chunked or defined.
Another question is whether a model like Vector-HaSH can fuel more efficient machine learning. Understanding how the brain organizes sequences might help AI systems handle incremental learning.
This study supports the idea that hippocampal and entorhinal circuits form an integrated reference system for storing huge amounts of information without abrupt loss.
All of this underscores that the same circuit that once guided our ancestors through forests and deserts can also manage recollections of birthdays, daily errands, and major life milestones.
The study is published in Nature.
—–
Like what you read? Subscribe to our newsletter for engaging articles, exclusive content, and the latest updates.
Check us out on EarthSnap, a free app brought to you by Eric Ralls and Earth.com.
—–