Among the vast store of memories we carry around in our heads, there is a large and crucial collection of maps. Most of these have little to do with geography in the usual sense; they’re more like road maps to our everyday surroundings, enabling us to recognize the places we know and navigate between them. Typically, it’s a sense of place we take for granted, until we become disoriented and confused. Then the result isn’t just unsettling; it’s similar to what occurs when someone experiences the early symptoms of Alzheimer’s disease.
This color-coded illustration shows the firing pattern of a rat’s grid cell as the animal moves around in a one-meter square box. Red indicates high grid-cell activity and blue indicates no activity. The neuronal firing in this cell is spaced at regular distances, with the active areas forming a triangular or hexagonal grid pattern within the box. (Jonathan Whitlock)
This is one reason why the brain’s navigation system is of great interest to neuroscientists. By studying how specialized cells interact to create the sense of place, researchers in the Kavli Institute for Systems Neuroscience at the Norwegian University of Science and Technology (NTNU) in Trondheim hope that they will shed some light on Alzheimer’s. However, the main focus of the Institute is on the mechanisms of spatial computation in the normal brain, explains Kavli Institute director Edvard Moser. The interest in spatial computation is motivated by the possibility for achieving something they have not seen before: the complete start-to-finish mechanics of a simple cognitive function. “This may tell us how the brain works generally, and it may tell us what goes on in other diseases,” Moser says.
Moser and his colleagues are focusing on the role of neurons known as grid cells. Located in the entorhinal cortex, these cells seem essential to the memory of place. As a rat moves around on a flat surface, the grid cells fire in patterns that show up as triangular grids when plotted on a map of the surface. Tens of thousands of these cells, each with its own grid, send data to the hippocampus, an area of the brain crucial to long-term memory and navigation.
“In some way the system is able to calculate position, so that when you get back in the same position, the cell fires,” says Moser. Significantly, the brain carries grid patterns with it into new surroundings, enabling it to map new objects and locations as it comes upon them. “The peaks [in cell firing patterns] exist even when a rat enters an unfamiliar area. In every unique environment, the system is anchored to the environment in a unique way.” These patterns may work like overlays, converting raw sense data to points and paths defined by coordinates on a neural map.
“This Changes Everything”
Interest in grid cells runs high at the NTNU Kavli Institute not just because they are intriguing in their own right, but because they are an NTNU discovery. They first came to light through research at the university’s Centre for the Biology of Memory (CBM), where Edvard and May-Britt Moser are co-directors. The Mosers, along with Torkel Hafting, Marianne Fyhn and Sturla Molden, first described grid cells in a 2005 paper in the journal Nature.
A zinc stain in this cross-section of a hippocampus shows the zone of connections (dark brown) between two regions, the dentate gyrus and CA3. This zone is important for making memories as different as possible before they are stored in the hippocampal network. The role of these connections was described by Jill Leutgeb and other researchers affiliated with the Kavli Institute. (Jill and Stefan Leutgeb)
That article was a major event in cognitive science. Writing in Scientific American’s “Mind Matters” blog, University of Texas neuroscientist James A. Knierim recalls reading the paper for the first time and saying to himself, “This changes everything.” Knierem says he realized “that I was reading a work of historic importance in neuroscience. No one had ever reported a neural response property that was so geometrically regular, so crystalline, so perfect. How could this even be possible? Yet the data were convincing.” The grid-cell model is striking not just for its elegance, but also because (as Edvard Moser notes) it forms a key link between perception and memory. The geometric grid patterns in the entorhinal cortex may be the mechanism by which perception of place and direction is encoded in the hippocampus.
Since the discovery of so-called “place cells” in the 1970s and head-direction cells in the 1980s, scientists have known that the hippocampus contains neurons that discharge when a rat comes to a certain place in its environment or points its head in a certain direction. These cells contain memory of where it has been and the pathways it has taken. How the memories are created in the first place (that is, how the sense of place is computed in new surroundings) was largely a mystery until the discovery of grid cells. Now, the grid cell system provides a possible explanation. The cells produce patterns as an animal moves, whether or not it recognizes any landmarks. In this way, the brain may constantly be creating computational grids that enable it to remember new locations in unfamiliar territory, as long as it can perceive the speed and direction of movement.
Still Plenty to Learn
Researchers have plenty yet to learn about this process, such as how grid cells create their repetitive patterns. Another question, says Edvard Moser, is how grid cells interact with other neurons (such as head-direction cells) that also play a role in navigation: “How is that information put together? How does the integration take place?” Also, scientists are not sure they have found all the navigation-related systems in the brain. Moser suggests that grid cells in the entorhinal cortex might work in concert with as-yet-undiscovered cells in other regions of the brain, such as the brain stem, the striatum, or the parietal cortex.
Research at Trondheim into the mechanisms of memory goes back at least to 1996, when the Mosers arrived there from the University of Oslo. The two, trained as clinical psychologists, had met in Oslo and wanted to study the links between the brain and behavior. Considering opportunities abroad in the 1990s, instead they landed two posts at NTNU in Trondheim, and in 1996 set up a laboratory there. In 2002, as the CBM, their lab was named a Center of Excellence by the Norwegian government and received a 10-year grant for its research.
The CBM now forms the core of the new Kavli Institute at Trondheim, which was inaugurated in August 2007 and where research continues on grid cells, as well as areas such as understanding the role of the hippocampus in shaping memory. Looking ahead, the Mosers see a wider, multi-disciplinary effort on the horizon. In particular, they are now looking for expertise in molecular biology, especially in virus-based techniques for introducing new genes into specific cell types in the grid-cell circuit. By expressing foreign genes in target cell types, they hope to be able to specifically silence or activate cells of interest. This is crucial for understanding how the grid network operates. ”The endeavor also requires computational modeling and the Institute aims to recruit researchers in this field too,” May-Britt Moser says. “When we found grid cells, computer science people went crazy because they had never before seen these geometrical patterns in the brain.” Now, she says, “we need good modelers” to take a closer look at those elegant designs.