Neural Dissociation of Cognitive Maps and Graphs in the Human Brain
Using fMRI and orthogonal manipulation of spatial and experiential distances, this study demonstrates that the human brain dissociates cognitive maps and graphs by representing Euclidean spatial proximity in the intraparietal sulcus and shortest-path experiential contiguity in the hippocampal/parahippocampal region and scene-selective cortical areas.
Original paper licensed under CC BY 4.0 (https://creativecommons.org/licenses/by/4.0/). This is an AI-generated explanation of a preprint that has not been peer-reviewed. It is not medical advice. Do not make health decisions based on this content. Read full disclaimer
The human mind has a remarkable ability to organize information, turning a chaotic world into a structured mental landscape. Two of the most fundamental ways we do this are through maps and through graphs. A cognitive map is like a mental snapshot of a place, arranging items based on their physical location in space, where the distance between two objects is measured by a straight line. A cognitive graph, by contrast, is a mental web of connections, linking items based on how we experience them moving from one to another, where the distance is defined by the path taken to get there. For decades, scientists have debated how the brain handles these two distinct forms of knowledge. Because our daily experiences often blend the two—places that are close together in space are usually also close together in the paths we walk—researchers have struggled to untangle whether the brain uses a single system for both or separate systems for each. Understanding this distinction is crucial, as it reveals the basic architecture of human memory and how we navigate not just physical spaces, but also complex ideas and social relationships.
A team of researchers at the Max Planck Institute for Human Cognitive and Brain Sciences and the University of Trento has now provided a clear answer to this question by designing a clever experiment that separates these two types of distance. They created a virtual world containing eight unique monuments connected by a specific network of paths. In this environment, they engineered a situation where two locations could be physically close to each other in a straight line, yet far apart in terms of the travel required to get from one to the other, because the connecting paths did not exist. Conversely, two locations could be far apart in a straight line but easily reachable via a short, direct path. This setup allowed them to test whether the brain treats the straight-line distance and the travel distance as separate pieces of information.
Sixty-one volunteers spent a day learning this virtual environment. They navigated through it from a first-person perspective, walking along the paths and viewing the monuments, while also being able to see the open space around them. To ensure they had truly learned both the layout of the space and the connections of the paths, the participants took tests where they had to estimate distances and choose the shortest routes. Once they demonstrated they understood the environment, they entered an MRI scanner. Inside the machine, they were shown images of the monuments and asked to recall the location and connections of each one. While they did this, the researchers used a technique called representational similarity analysis to look at the patterns of brain activity. This method checks whether the brain's response to two different monuments is more similar when those monuments are close together in the map or when they are close together in the path network.
The results revealed a striking division of labor in the brain. The researchers found that the straight-line, map-like distances were represented in a specific area at the back of the brain known as the intraparietal sulcus. This region is known for processing spatial information and visualizing layouts. In contrast, the travel-based, graph-like distances were represented in a different set of areas, including the hippocampus and surrounding regions that are typically active when we view scenes or navigate routes. The hippocampus, a structure deep in the brain often associated with memory, showed activity patterns that tracked the shortest path between locations, but it did not seem to track the straight-line distance. This suggests that the hippocampus is specialized for organizing experiences based on how we move through them, rather than where things sit in a static space.
The study also looked at other parts of the brain involved in recognizing scenes, such as the parahippocampal place area. These areas, which usually help us recognize the layout of a room or a landscape, also showed activity patterns linked to the path-based connections between locations. This finding suggests that the brain's scene-processing system works closely with the memory system to build a graph of our experiences. Meanwhile, the area responsible for the map-like representation remained distinct, showing no signs of tracking the path connections. The researchers were careful to rule out the possibility that these results were simply due to the shapes of the monuments or other visual features, confirming that the brain was indeed processing the abstract concepts of spatial distance and travel distance separately.
These findings offer a new perspective on a long-standing debate in neuroscience regarding the function of the hippocampus. While earlier theories suggested the hippocampus acts as a universal map of space, this study suggests it is more accurately described as a system for organizing the sequence of our experiences. The brain appears to maintain two parallel ways of knowing: a map in the parietal cortex that holds the layout of the world as a whole, and a graph in the hippocampus and scene-processing areas that records the journey and the connections between events. This separation allows us to understand our environment both as a fixed place and as a series of stories we have lived through, providing a richer and more flexible foundation for human thought and memory.
Drowning in papers in your field?
Get daily digests of the most novel papers matching your research keywords — with technical summaries, in your language.