THE sense of place and the ability to navigate relate to two of the most important functions of the brain. The ability of the brain to know where we are in space has been one of the mysteries in human physiology and psychology that has engaged philosophers and scientists since the late 18th century. How are we able to find the way from one place to another? How is this information stored in the brain so that we recall the route when we revisit a path? This year’s Nobel Prize in Physiology or Medicine is shared by three scientists who unravelled this mystery. Together they found the underlying cellular basis for this higher cognitive function, a kind of positioning and navigation system, an “inner GPS” as it were, that enables us to orient ourselves and find our way in space.
In 1971, John O’Keefe, an American-British neuroscientist of Irish origin, found a type of neuron in the rear part of an area of the brain called the hippocampus being activated when a freely moving rat explored an enclosed space. Certain neurons fired when the rat was at a certain place and certain others fired when it was at another place. This collection of neurons was in effect a two-dimensional representative map of the room. O’Keefe called these specialised cells “place cells”.
Navigation involves a sense of distance and direction that is based on the integration of motion and knowledge of previous locations. More than three decades later, in 2005, a Norwegian scientist couple, May-Britt Moser and Edvard Moser, discovered another set of neurons in the medial entorhinal cortex, a region of the brain next to the hippocampus, which created a coordinate system and enabled precise positioning and path finding. They called this set of neurons “grid cells”. These two groups of nerve cells together form interconnected nerve cell networks that are critical to the mental computation of spatial maps and navigational tasks which enable mammalian systems to determine their position and move about.
The work by O’Keefe and the Mosers, noted the Nobel Foundation’s document providing background scientific information on the award-winning research, “has dramatically changed our understanding of how fundamental cognitive functions are performed by neural circuits in the brain and sheds new light onto how spatial memory might be created”.
Seventy-five-year-old O’Keefe, who did his doctoral degree in physiological psychology from McGill University in Canada, is currently Director of the Sainsbury Wellcome Centre for Neural Circuits and Behaviour at University College London. Fifty-one-year-old May-Britt Moser, who obtained her PhD in neurophysiology from the University of Oslo, where she met her future husband, is currently Director of the Centre for Neural Computation at the Norwegian University of Science and Technology in Trondheim.
Fifty-two-year-old Edvard Moser’s doctoral degree is also in neurophysiology and he is currently Director of the Kavli Institute for Systems Neuroscience in Trondheim. The Mosers were postdoctoral fellows together, first in Richard Morris’ laboratory at the University of Edinburgh and later in 1996 as visiting scientists in O’Keefe’s laboratory at University College London. Edvard worked in O’Keefe’s laboratory for three months and May-Britt for one month. “He trained us to do the type of single cell recordings that we have been doing since,” Edvard told the Nobel Prize website’s interviewer in his post-award announcement interview. “The three months that I spent in his lab are the most efficient learning period I ever had. I learnt so much…. He has been a fantastic mentor and it’s extremely nice that we can now share the prize together,” he said. O’Keefe gets one half of the eight million Swedish kronor ($1.11 million) award money, while the other half is shared equally by the Mosers.
In his Critique of Pure Reason (1781-87), the German philosopher Immanuel Kant said that some mental abilities existed as a priori knowledge independent of experience. As Aiden Arnold pointed out in a blog on the website of Scientific American , Kant argued that space as we know it was a preconscious organising feature of the mind, a scaffold upon which we were able to understand the physical world of objects, extension and motion. “In a sense,” Arnold explained, “space to Kant was a window into the world, rather than a thing to be perceived in it.”
This Kantian theory actually provided 20th century scientists the framework for empirical investigation into how the space that we perceive is constructed in the mind. The work of the American psychologist Edward Tolman in 1948 was a key step towards this final understanding, which the research of the Nobel laureates has now made possible.
Tolman was interested in the behaviour of rats in labyrinths. He wanted to know whether rats came to understand the spatial layout of the environment by behavioural mechanisms alone or whether there was a cognitive process in the brain that governed this navigational ability. Tolman found that rats were able to navigate through a maze efficiently even though they had not been exposed to the locations previously. This suggested to him that rats spontaneously formed a mental map of the maze that allowed them to identify locations and navigate their way through. Tolman hypothesised that this ability to generate a “cognitive map” was the primary mechanism in the brain through which mammals learned about and navigated through complex environments. While this idea of a cognitive map in the brain began to be accepted widely in the 1960s, it still remained a hypothesis. The discovery of the neural basis for the creation of this cognitive map and, as Arnold points out, how this differed from other potential strategies of navigation and spatial learning achieved through chains of sensory-response relationships had to wait for the evolution of techniques using chronically implanted wires to record signals from the nerve cells of animals moving freely in an environment. This happened in the late 1950s.
O’Keefe became fascinated by the question and began in the late 1960s to attack the problem with the new methods of recording signals from neurons. While other researchers used these new techniques to merely record signals of restricted behavioural aspects or stimulus-response relationships, O’Keefe recorded cellular activity during natural behaviour. In 1971, with his colleague Jonathan Dostrovsky, he came to discover the place cells in the dorsal, or rear, partition of the hippocampus called CA1 when recording signals from neurons of rats moving freely in an enclosed space (Figure 1). The firing pattern in the cells that he found was completely unexpected. These cells got activated in a manner that had not been seen before in the other cells of the brain. Certain cells fired only when the rat was in a particular place in the space. By systematically changing the environment and testing different theoretical possibilities, O’Keefe showed that the firing not merely registered the visual input in the neurons but was effectively generating the “place field”, a positional map, of the environment in the brain.
In subsequent experiments, O’Keefe demonstrated that these neurons had memory functions as well. In different environments, the place cells formed a different pattern or map and thus generated numerous maps. The memory of an environment can thus be stored as a specific combination of place cell activities in the hippocampus.
At first, O’Keefe’s conclusions were met with scepticism, but with meticulous demonstration, the role of place cells as a substrate for generating spatial maps and storing the information about a given environment came to be accepted and today it is central to our understanding of how the brain performs this cognitive function. The series of works by O’Keefe and his co-workers culminated in a book titled The Hippocampus as a Cognitive Map (John O’Keefe and Lynn Nadel, 1976).
O’Keefe’s work began to trigger a large number of experimental and theoretical studies on both mapping and memory functions. These studies revealed that place cells were involved in the function of distance measurement—necessary for navigation—only under certain circumstances.
When the Mosers came onto the scene after their exposure to the field in O’Keefe’s laboratory, the prevalent theory was that the formation of place fields occurred within the hippocampus itself. But at the turn of the century, the Mosers were interested in looking at what happened in the exterior connections to the hippocampus when rats were allowed to move freely in an environment. They discovered that there was also a pattern of activity in the entorhinal cortex, which is in the hippocampus’ neighbourhood and is located in the dorsal edge of the rat’s brain.
The neuroscientist couple found that a major part of the output from the cortex was linked to a part of the hippocampus called the dentate gyrus, which in turn connected to the region CA3 in the hippocampus and further to CA1, the very place where O’Keefe had identified his place cells. Interestingly, in 2002, they found that disconnecting the inputs from the cortex through CA3 did not disturb or affect the place fields in CA1. They inferred that the cells in the cortex were simultaneously and reciprocally encoding different sets of information relating to the place fields. They then began to look for the cells which were coding the place cells that had been triggered in the hippocampus.
It was known that the cortex too contained cells that were similar to the place cells in the hippocampus, which the Mosers established. But more detailed studies soon revealed that grid cells, which had properties distinct from the place cells, were involved in the process. The firing in these grid cells showed an unexpected pattern. They were active in multiple places which together formed the nodes of an extended hexagonal grid similar to a honeycomb. Each of these cells was activated in a unique spatial pattern, and collectively these grid cells constituted a coordinate system that enabled spatial navigation (Figure 2). The Mosers also showed that the grid pattern, which had not been seen before, did not arise out of a simple transformation of motor or sensory signals but were a result of complex network activity.
The grid system thus provided a means to measure distances traversed or was a metric to the spatial maps in the hippocampus. Recordings from multiple grid cells in different parts of the cortex also showed that the grid cells were organised as functional modules with different grid spacing, or equivalently the scales for the coordinate system, which covered distances of a few centimetres to metres, thus covering small to large environments.
Exploring further, the Mosers showed that there was a reciprocal influence between the grid cells and the place cells. Together with other cells of the cortex that recognise the direction of the head and the edges of the place field, the grid cells form circuits with the place cells in the hippocampus. That is, these cells in the cortex may be even contributing to the firing pattern in the hippocampus through a kind of feedback mechanism. This circuitry thus forms a comprehensive GPS-like inner positioning system in the brain (Figure 3). The Mosers’ groundbreaking discovery of grid cells, a coordination system with a metric and the identification of the cortex as the computational hub for measuring distances and generating a cognitive map with appropriate scales has opened up new avenues to advance the understanding of the neural mechanisms underlying spatial cognitive functions and other cognitive processes such as memory, thinking and planning.
These types of cells have now been found in mammals beyond the experimental animal systems of rats and mice. Humans in particular have large hippocampal-entorhinal structures in the brain that have been previously associated with spatial learning and episodic memory. Studies in the immediate years after the Mosers’ work also suggested that the human brain had a similar spatial coding system. But direct evidence of both place cells in the hippocampus and the grid cells in the entorhinal cortex came when signals from neurons in the brains of epileptic patients during post-surgical investigations were recorded using functional magnetic resonance imaging (fMRI). It has been found that hippocampal-entorhinal structures in the brains of all mammals are very similar. Existence of hippocampal-like structures in non-mammalian vertebrates with navigational capacity has also been observed. These together suggest, from an evolutionary perspective, that the grid-place cellular network is a functionally robust system that has been conserved in vertebrate evolution.
This Nobel Prize-winning work has unravelled the brain’s positioning and navigation system and marks a paradigm shift in the understanding of the neurological basis for higher cognitive functions. More importantly, the work also has relevance to medicine and therapy for certain brain disorders such as dementia and Alzheimer’s disease. These diseases frequently affect the hippocampus and the entorhinal cortex regions of the brain at an early age. Such patients often lose their way and cannot recognise the environment. Their episodic memory also gets affected. O’Keefe and co-workers have demonstrated in a mouse model with Alzheimer’s disease that the degradation of place cells correlated with the deterioration of spatial memory in the animal. Knowledge about the underlying system has definitely helped in the understanding of the basis for the loss of spatial memory in people affected with these brain disorders. However, this has not yet translated into clinical research or practice, let alone curative therapy. But the research certainly gives one hope that that day may not be very far away.
How the brain processes and stores the idea of time, however, has remained much more elusive in physiology and neuroscience. Essentially what is known about the representation of time in the brain so far is twofold. One, how overlapping events are separated into discrete episodes and, two, that the brain does sequentially order these events in a kind of temporal framework. It has also been hypothesised, largely on the basis of empirical physiological findings on patients with lesions in the hippocampus, that the hippocampus does play a critical role in this parsing and sequential ordering of events. This suggests that the hippocampus is involved in more than simply representing the spatial layout of an environment. In fact, in 2011, Howard Eichenbaum and co-workers of Boston University did identify a different group of cells in the hippocampus, which they called “time cells”. Their studies also revealed that time cell activity could uniquely code successive events and disambiguate overlapping events into temporally organised episodes.
Although there is now evidence that both spatial environment and the temporal sequence of events are governed by similar neural structures, it is still an open question whether the Minkowski-Einstienian paradigm of space-time unity in physical dynamics is reflected in a kind of generalised cognitive process in the physiology of the brain.
