New Study Explores How the Brain Maps Out Space

A recent investigation by neuroscientists from the Massachusetts Institute of Technology has revealed insights into how our brains create internal maps of space. Their research involving mice shows that while some brain cells quickly identify specific locations, multiple neurons working together and repeated experiences—along with sleep—are essential to forming a clear mental representation of one’s surroundings.

Published in Cell Reports, this study supports the concept that cognitive mapping gradually evolves, relying not only on specialized “place cells” found in the hippocampus but also on a collective of neurons that initially respond weakly to individual locations. Over time, with exploration and sufficient sleep, these neurons begin to synchronize with place cells, developing cohesive patterns reflecting the environment’s layout.

The researchers were keen to address a long-standing question in neuroscience: how does the brain transition from recognizing particular places to constructing a comprehensive internal map? Since the 1970s, it’s been known that certain neurons in the hippocampus activate when an animal occupies a particular space. However, a functional map isn’t just a collection of points; it requires a network connecting those points. Edward Tolman first introduced the idea of cognitive maps in 1948, and while the discovery of place cells supported his theory, the specifics of how the brain links distinct locations into a complete map remained unclear.

“I’m interested in this project because understanding how memory is formed in the brain is one of the most fundamental questions in neuroscience,” said research scientist Wei Guo from the Picower Institute of Learning and Memory, who worked under Professor Matthew Wilson.

To delve deeper, the research team observed mice that roamed unfamiliar mazes without receiving any rewards or punishments, which allowed them to study how the brain learns spatial arrangements without external reinforcement. Their focus was on a type of passive learning called latent learning, where knowledge is gained without immediate behavioral changes.

To monitor brain activity, the researchers employed advanced calcium imaging techniques. By genetically altering hippocampal neurons to emit a fluorescent protein indicating activity, they installed tiny lenses and microscopes in the mice’s brains. This setup enabled the researchers to capture activity from hundreds of neurons in the hippocampus while the mice explored the mazes or rested.

Utilizing a method called manifold learning, they visualized neural activity patterns over time. Initially, on the first day in the maze, each neuron exhibited its unique firing pattern, but the collective did not form a recognizable map. By the fifth day, however, the neural activity began to coalesce into a low-dimensional structure, termed a “neural manifold,” that mirrored the maze’s design. This demonstrated that the brain gradually transitioned to representing the entire environment instead of just isolated points.

One notable finding was the critical role of sleep in this transformation. In one experiment, mice explored the maze twice in a single day, with a three-hour interval in-between. Some were allowed to sleep, while others were gently kept awake. Only the mice that could sleep showed improvements in their neural activity aligning with the maze layout, indicating that sleep played a significant role in reorganizing hippocampal neural patterns into a more coherent mental map.

The researchers also concentrated on two categories of neurons. Some, labeled as strongly spatial cells, displayed clear firing patterns when the mouse visited specific maze areas even during their first exploration. These cells maintained stability over time and didn’t significantly modify their behavior. In contrast, weakly spatial cells initially had less defined firing patterns but gradually became more spatially tuned. Notably, these weakly tuned cells became increasingly coordinated with the broader neural network, particularly during sleep.

They assessed each neuron’s “mental field” to understand its activity in relation to the broader neural network. They found that even if weakly spatial neurons never developed into strong place cells, they significantly influenced the overall structure of the cognitive map. These neurons established correlations with others, forming a network that could represent not just isolated locations but also their interconnections.

When the researchers tried reconstructing the neural map using exclusively strong place cells, the resulting patterns showed minimal change over time. It was only when they included weakly spatial neurons that the comprehensive map-like structure emerged. This implies that while place cells provide the basic elements, it’s the subtle shifts in these less specialized neurons that contribute to forming a complete mental picture.

The study suggests a broader significance for weakly tuned neurons in the learning process. Instead of being mere background noise, these cells assist the brain in creating flexible, interconnected representations. They seem to react not just to specific locations but also to the overall activity combinations within the network. Over time, their activity became more synchronized, helping to weave together a spatial map.

“I was surprised to find that a subset of previously overlooked neurons with weak activity turned out to be crucial for memory formation,” Guo remarked.

Interestingly, sleep seemed to enhance this process. After maze exploration, neural activity patterns during rest mirrored those observed during navigation—a phenomenon known as replay. This replay likely aids the brain in solidifying connections between various locations. Post-sleep, the neural states during rest appeared more like those recorded during maze navigation, suggesting that sleep plays a role in reinforcing and refining the cognitive map.

This research underlines that memory formation often involves slower, more distributed changes tied to experiences and sleep, rather than just quick, isolated events. Guo noted, “Sleep is very important in transforming your experience into memory.”

However, as with any study, there are limitations. The reliance on calcium imaging provides slower and less precise readings than electrical recordings. Additionally, their recordings were limited to one section of the hippocampus, and other brain areas that contribute to spatial memory were not included in their observations.

Looking ahead, Guo intends to further explore the local circuits within the hippocampus and their interactions with other brain regions during memory formation.

The study, titled “Latent learning drives sleep-dependent plasticity in distinct CA1 subpopulations,” was conducted by Wei Guo, Jie J. Zhang, Jonathan P. Newman, and Matthew A. Wilson.