where are we where are we going

Like the GPS in our phones and cars, the system in our brain estimates where we are and where we intend to go, by combining many signals relating to our location and the passage of time. The brain makes these calculations as a matter of routine with minimal effort, so we are hardly aware of them at all. Only when we get lost or when our navigational skill is impaired due to injury or degenerative disease of the nervous system, do we implicitly perceive how essential this mapping and navigation system is to our very existence.

The ability to decipher where we are and where we should go is essential to our survival. Without it we would not be able to find food or reproduce, like us and all other animals. In its absence we would all be extinct, not just individuals but the entire biological species.

The sophistication of the mammals' system is especially evident when comparing it to that of other animals. The simple ringworm, Caenorhabditis elegans, which has a total of 302 nerve cells, navigates almost exclusively in response to odor signals, and moves along a path dictated by an increase or decrease in the concentration of some odor.

Animals endowed with more complex nervous systems, such as desert ants or the honey bee, find their way with the help of additional strategies. One of these methods is called path integration, a GPS-like mechanism in which nerve cells calculate a position based on constant monitoring of the direction and speed of the animal's movement relative to the starting point - a task that is done without reference to external signs such as physical landmarks. In vertebrates, and especially in mammals, the collection of behaviors that allows the animal to place itself in its environment is even wider.

Mammals, more than any other class of animals, rely on the ability to create neural maps of the environment: patterns of electrical activity in the brain during which groups of nerve cells send electrical signals in a way that reflects the outline of the environment and the animal's position within it. The conventional wisdom is that the creation of such mental maps occurs in the cerebral cortex, the wrinkled upper layer of the brain that developed quite late in evolution.

Over the past few decades, researchers have deeply understood the way the brain creates these maps and changes them while the animal is in motion. The latest studies, most of which were done in rodents, revealed that the navigation systems consist of several types of specialized cells that continuously calculate the animal's position, the distance traveled, the direction of its movement and its speed of movement. By combining the activity of these different cells, a dynamic map of the local space is obtained, which functions not only at the time it was created, but can be saved as a memory for future use.

The neuroscience of space

The research on spatial maps of the brain began with the work of Edward K. Tolman, professor of psychology at the University of California at Berkeley, in the years 1918-1954. Laboratory studies conducted in rats before Tolman's research began showed that the animals found their way around the environment, and remembered it, in response to successive stimuli received along their path of movement. When learning to run a maze, for example, the researchers thought they were retrieving from memory the sequences of turns in the maze from beginning to end. This idea did not take into account the possibility that the animals might have in memory an overall picture of the entire maze that would allow them to plan the best path.

Tolman disputed the views that were accepted at the time. He saw rats take shortcuts or detours, moves that would not have been expected if they had only learned one long sequence of behaviors. Based on his observations, he hypothesized that the animals create mental maps of the environment that reflect the spatial geometry of the external world. These cognitive maps not only helped them find their way; They seemed to record information about events that the animals experienced at specific sites.

Tolman's ideas, first proposed around 1930, remained controversial for decades. Their acceptance was slow, in part due to the fact that they were based only on observations of the behavior of experimental animals, which could be interpreted in many ways. Tolman had neither the concepts nor the tools needed to examine the question of whether an internal map of the environment actually exists in an animal's brain.

Forty years passed until direct evidence for the existence of such a map was obtained from studies of neural activity. The development of microelectrodes in the 50s made it possible to monitor the electrical activity of single cells in the brain of non-anesthetized animals. These incredibly thin electrodes allowed the researchers to detect the sending of signals, "firing" in the researchers' parlance, from single neurons while the animals were engaged in their normal activities. We say of a nerve cell that it "fires" when it generates an action potential: a short change in the voltage drop on both sides of its cell membrane. Action potentials cause the nerve cells to release neurotransmitter molecules, which transmit the signals from one nerve cell to another.

John O'Keefe of University College London used microelectrodes to monitor action potentials in the hippocampus of rats, an area of ​​the brain known many years earlier to be important in memory activity. In 1971 he reported that neurons in this area fired when a rat that was in a box was found for a period of time in a certain location, and as a result he called these neurons "place cells". O'Keefe noticed that different place cells fired at different places in the box, and also saw that the firing pattern of all the cells together created a map of places in the box. Based on the combination of the activity of many place cells, it was possible to read and deduce from the electrodes the specific position of the animal in the box at any given moment. In 1978, O'Keefe and his colleague Lynn Neidel, then working at the University of Arizona, proposed the possibility that place cells are actually an integral part of the cognitive map that Tolman envisioned in his theory.

cortical map

The discovery of the place cells opened a window to the deepest parts of the cerebral cortex (the cortex), in the areas furthest from the sensory cortex (the areas of the cortex that receive input from the senses) and the motor cortex (the areas that produce signals to perform movements or control movement). In the late 60s when O'Keefe began his work, very little was known about the timing of the start and stop of a nerve cell's activity, and all knowledge was limited only to areas called the primary sensory cortex, where neural activity is directly dependent on sensory input such as light, sound and touch.

Neuroscientists at the time believed that the hippocampus was too far away from the sense organs to be able to understand how it processes the sensory input with the help of the microelectrode recordings. The discovery of cells in the hippocampus that created a map of the immediate environment of the animal put an end to this speculation.

However, as impressive as the find was and indicated an important role that place cells might play in navigation, for decades after the discovery no one knew what that role might be. The place cells were in an area of ​​the hippocampus called CA1, which is the end point of a chain of signals that started elsewhere in the hippocampus. It has been hypothesized that the place cells receive many of the calculations related to navigation from other areas of the hippocampus. In the early 2000s, we both decided to deepen the research of this idea in the new laboratory we established at the Norwegian University of Science and Technology in the city of Trondheim. This research work eventually led to a great discovery.

In collaboration with Menno Witter, who currently works at our institute, and with the help of a group of particularly creative students, we used microelectrodes to monitor the activity of place cells in the hippocampus of rats, after damaging part of a neural circuit known to transmit information to these cells. We expected that our work would confirm the importance of this neural circuit for the normal function of place cells. To our surprise, the neurons at the edge of the circuit, in the CA1 region, still continued to fire even when the animals reached specific locations. The inescapable conclusion our team reached was that place cells did not depend on this circuit in the hippocampus to estimate the animal's location. Following this, our attention was directed to the only neural pathway that was not affected by our intervention: the direct connections of CA1 to the entorhinal cortex, a nearby area that provides an interface to the rest of the cerebral cortex.

Humans and other mammals produce internal maps of the environment - patterns of neural activity that are created when brain cells fire and thereby reflect where the animal is and how it is positioned in relation to its environment.

In 2002, we inserted microelectrodes into the entorhinal cortex, still in collaboration with Witter, and began recording the activity while the animals performed tasks that were similar to the tasks we imposed on them as part of our work on place cells. We directed the electrodes to a region of the entorhinal cortex that has direct connections with parts of the hippocampus and where place cells have been recorded in almost all studies prior to our study. It turned out that many cells in the entorhinal cortex fired when the animal was in a certain location in the enclosure, just like the place cells in the hippocampus behave. But unlike a place cell, a single cell in the entorhinal cortex fired not only in a single location that the rodent reached, but in many places.

The most striking feature of these cells, however, was the way they transmitted the electrical signal. The pattern of their activity became clear to us only when in 2005 we increased the area of ​​the compound within which we made the records. After expanding it to a certain size we found that the many locations where the entorhinal cells fired formed the vertices of a hexagon. At each vertex, the cell, which we called a grid cell, fired when the animal passed over it.

It seems that the hexagons, which covered the entire area of ​​the complex, were the units of a grid of coordinates, similar to the squares formed by the lines of longitude and latitude on road maps. The firing pattern raised the possibility that grid cells, unlike place cells, provide information about distances and directions, and help the animal identify its movement path while relying on internal cues arising from body movements without relying on information coming from the environment.

Different aspects of the network structure changed when we examined the activity of cells in different parts of the entorhinal cortex. In the dorsal part, near the top of this brain structure, the cells created a network map of the complex that was made up of dense hexagons. The spaces between the hexagons grew in a series of steps, or modules, as we moved towards the lower or ventral part of the entorhinal shell. In each module, the hexagonal units in the coordinate grid were spaced at a different spacing.

The degree of spacing between the grid cells in successive modules from the dorsal to the ventral direction, could be determined by multiplying the distance between the cells in the previous module by a constant factor of 1.4, which is approximately the square root of 2. In the module at the upper end of the entorhinal cortex, a rat activated a grid cell at one vertex of a hexagon, had to pass between 30 and 35 centimeters to the adjacent vertex. In the next module below it, the animal had to move between 42 and 49 centimeters, and so on. In the lowest module, the distance between the vertices increases to several meters.

The grid cells and their neat organization caused us great excitement. In most parts of the cerebral cortex, the nerve cells have firing patterns that seem completely disordered and difficult to decipher, but here, deep in the cortex, a system of cells was found that fired signals in an orderly and predictable pattern. We were eager to explore them. But these cells and place cells were not the only ones involved in mapping in the mammalian world. More surprises awaited us.

Already in the late 80s and early 90s, James B. Rank of SUNY Downstate Medical Center and Jeffrey S. Taub, now of Dartmouth College, described cells that fired when a rodent turned in a certain direction. Rank and Taub discovered such cells that indicate the direction of head turning in an area called the presubiculum, another area of ​​the cerebral cortex adjacent to the hippocampus.

Our studies revealed that these cells are also found in the entorhinal cortex, when they are scattered among the cells of the network. Many head direction cells found in the entorhinal cortex also function as grid cells: the places in the complex where the cells fired also formed a grid, but the cells became active in those places only when the rat's head was pointed in a certain direction. These cells seem to have provided the animal with something like a compass; By tracking the cells it was possible to read the direction the animal's head was facing in relation to the environment at any given moment.

A few years later, in 2008, we discovered another cell type in the entorhinal cortex. These border cells fired whenever the animal approached a wall or the edge of the enclosure or some other demarcation factor. These cells seemed to calculate how far the animal was from the boundary. This information served the grid cells that could use it to estimate the distance the animal had traveled from the wall or border, and it might also serve as a reference point that could later remind the rat of the general location of the wall.

And finally, in 2015, another type of cell joined the scene. It responded uniquely to running speed, regardless of the animal's location or direction of travel. The firing rate of these neurons increased in direct proportion to the speed of movement. In fact we could verify how fast the animal was moving at any given moment just by looking at the firing rate of a small number of speed cells. In addition to the information from the head direction cells, the speed cells can be used as providers of continuously updated information to the network cells, informing about the movements of the animal - its speed, the direction of its movement and the distance from the starting point.

grid for place cells

The discovery of grid cells was born from our ambition to reveal the types of input that allow the place cells to provide the animal with an internal image of its environment. Now we understand that the place cells add together signals received from different types of cells found in the entorhinal cortex, while the brain tries to follow the path the animal is taking and to know where it is moving in the environment in which it is found. But even these processes are still not enough to tell us how mammals navigate their way.

Our first studies focused on the medial, that is, the inner, entorhinal cortex. Place cells may also receive signals from the lateral entorhinal cortex, which transmits processed input from several sensory systems, including information about smells and the identity of objects in the environment. By combining input types from the medial and lateral entorhinal cortex, place cells interpret signals coming from all over the brain. The complex interaction that takes place between messages arriving at the hippocampus, and the creation of location-specific memories made possible thanks to it, all of these are still being studied in our laboratory and in other laboratories, and there is no doubt that this research will continue for many years to come.

Perhaps we can understand how the spatial maps of the medial entorhinal cortex and the hippocampus join each other and enable navigation if we ask how they differ from each other. John Cubby and the late Robert Hugh Mahler, both of SUNY Downstate Medical Center, showed in the 80s that hippocampal maps formed by place cells can change completely when an animal moves to a new environment, even if it moves to a different colored enclosure in the same place, in the same room .

Experiments conducted in our laboratory, with rats that occasionally wandered through 11 enclosures in a series of different rooms, revealed that each room very quickly led to the creation of its own map that was independent of the others, a finding that supported the idea that the hippocampus produces spatial maps adapted to specific environments.

In contrast, the maps formed by the medial entorhinal cortex are universal. Grid cells - as well as head orientation cells and border cells - that fire together at certain combinations of locations on the grid map of one environment, also fire at analogous locations on the map of another environment, as if latitudes and longitudes of the first map were imposed on the new environment. The sequence of cells firing when the animal moves in a northeast direction in one room of the cage changes when the rat moves in the same direction in another room. The signaling pattern between these cells in the entorhinal cortex is the pattern the brain uses to navigate its environment.

These codes are transferred from the entorhinal cortex to the hippocampus, where they are used to create unique maps to a certain place. From an evolutionary point of view, two series of maps from which the information coming from is unified and directs the animal along its path seem to be an efficient solution for a system used by animals for spatial navigation. The networks formed in the medial entorhinal cortex, which measure distances and directions, do not change when moving from room to room. In contrast, the place cells of the hippocampus produce unique maps for each room to itself.

local maps

Much more work is required to reach an understanding of the navigation system. Almost all of our knowledge about place cells and grid cells has been gained from experiments that focused on recording electrical activity picked up by nerve cells of rats or mice randomly wandering through distinctly artificial environments: flat-bottomed boxes with no internal structures that might serve as landmarks.

A laboratory is very different from natural environments, which are constantly changing and full of three-dimensional objects. The reductionism that characterizes the studies raises questions about how place cells and grid cells work, such as whether they fire the same way when the animals are outside the lab.

Experiments conducted in complex mazes that try to mimic natural habitats of animals provide some clues as to what might be happening in them. In 2009, we recorded grid cell activity as the animals moved through a complicated maze, encountering at the end of each straight passage very sharp turns that marked the beginning of the next corridor. The study showed, as expected, that the grid cells form hexagon-shaped patterns that map the distances for the rats moving through single passages in the maze. But every time an animal turns from one passage to the next, there is a sharp change in the pattern. A separate mesh pattern is worn on the new passage, almost as if the rat entered a completely different room.

Later research done in our lab showed that grid maps decompose in open environments into smaller maps, if those spaces are large enough. Now we check how the smaller maps come together to create an integrated map of a given area. But even these experiments are oversimplified because the complexes are flat and horizontal. Experiments carried out in other laboratories - which are observed with flying bats and rats climbing the sides of cages - are beginning to yield some clues: place cells and head direction cells apparently fire in certain places in every three-dimensional space, and most likely grid cells also behave this way.

Space and memory

The navigation system in the hippocampus is not just a means of helping animals get from point A to point B. Besides receiving information about location, distance and direction from the medial entorhinal cortex, the hippocampus records what is in a certain location - a car or a mast - as well as the events that take place there. The spatial map created by the place cells contains not only information about the general location of the animal, but also details concerning the animal's experiences, in a manner similar to the concept of the cognitive map coined by Tolman.

Some of this additional information apparently comes from nerve cells located in the lateral part of the entorhinal cortex. Details about objects and events merge with the animal's coordinates and are stored as memory. When the memory is retrieved at a later time, both the event and the location emerge from it.

This coupling of place and memory is reminiscent of a remembering strategy invented by the ancient Greeks. The method known as the "Method of Places" allows a person to memorize a list of items while imagining that he has put each item in a location along a well-known path that passes through some place such as a field or a building - an arrangement known as a memory palace. People participating in memory competitions use this technique to recall long lists of numbers, letters or cards. Unfortunately, the entorhinal cortex is one of the first areas of the brain to lose function in people with Alzheimer's. The disease results in the death of brain cells in this area, and the reduction of its size is considered a reliable measure for diagnosing people at risk of the disease. The tendency to wander randomly and get lost is one of the earliest signs of the onset of the disease. In the later stages of Alzheimer's disease, the death of cells in the hippocampus causes the loss of the ability to recall events from the past as well as to remember concepts such as the names of colors. In fact, a recent study produced evidence that young people who carry a gene that places them at increased risk for Alzheimer's may suffer from failures in the tissue function of their network cells - a finding that may lead to new ways of diagnosing the disease.

Rich repertoire

Today, more than eighty years since Tolman first spoke of the existence of a mental map of our environment, it is clear to us that place cells are only one component of the incredibly complex representation that the brain constructs of the space around it to calculate position, distance, speed, and directions. The many cell types found in the navigation system of the rodent brain are also found in bats, monkeys and humans. The existence of the cells among many taxonomic series of mammals indicates that grid cells and other cells involved in navigation appeared early in mammalian evolution, and that similar neural algorithms are used to calculate position among many biological species.

Many of the building blocks of Tolman's map have already been discovered, and we are beginning to understand how the brain produces and places them. The system for representing space is now one of the most understood neural networks in the mammalian cerebral cortex. We are beginning to identify the algorithms used by it, and these will help us decipher the neural codes used by the brain for navigation.

As is the case in many other areas of research, new findings raise new questions. We know that there is an internal map in the brain, but we still do not sufficiently understand how the various components of the map work together and yield a consistent representation of location, while information is read by other systems in the brain that decide where to go and how to get there.

There are many more questions. Is the spatial network in the hippocampus and the entorhinal cortex limited to navigation only in the immediate environment? In rodents, we tested it in limited areas with a radius of only a few meters. Are grid cells and place cells also used for long-range navigation, for example when bats migrate for distances of hundreds or thousands of kilometers?

And finally, we wonder how the grid cells appear, whether there is a critical period for their formation in the animal's development, and whether it is possible to find place cells and grid cells in other vertebrates and invertebrates as well. If invertebrates use such cells, the finding would indicate that evolution has been using spatial mapping systems for hundreds of millions of years. The brain's GPS will continue to be a rich source of ideas for new studies that will occupy generations of scientists for decades to come.