John O'Keefe

Introduction

John O'Keefe (born November 18, 1939) is an American-British neuroscientist trained as a psychologist who discovered that individual neurons in the hippocampus fire when an animal occupies a particular location in its environment. He called them place cells, and the discovery supplied the first direct physiological evidence for the "cognitive map" that Edward C. Tolman had inferred from behavior alone in 1948. For this and related work he was awarded half of the 2014 Nobel Prize in Physiology or Medicine, with May-Britt Moser and Edvard Moser sharing the other half.

It is worth being precise about the framing, because O'Keefe is often described loosely. He is a psychologist by doctoral training — his Ph.D. is in physiological psychology from McGill University — and the Association for Psychological Science described all three 2014 laureates as psychological scientists. But the work itself is systems neuroscience, and O'Keefe has spent his entire independent career at University College London rather than in a psychology department. Calling him simply "a psychologist" understates the neurophysiology; calling him "not a psychologist" loses the fact that his central question — how an organism represents its environment — is one psychology asked first.

O'Keefe's standing rests on more than a single finding. Working with Lynn Nadel, he wrote the 1978 book that turned a collection of single-unit recordings into a general theory of hippocampal function. With Michael Recce he characterized how place cells time their spikes relative to the hippocampal theta rhythm. With Neil Burgess he showed that place fields are governed by the geometry of environmental boundaries and predicted a class of cells that other laboratories later found. All of this happened before the Nobel, and none of it was uncontroversial at the time.

Early Life and Education

O'Keefe was born in New York City to Irish immigrant parents. Neither parent had completed elementary school in Ireland, though his father later earned a high school diploma in New York. He attended Regis High School in Manhattan, a Jesuit school with a demanding classical curriculum. Britannica's biography records that he first studied aeronautical engineering at New York University before enrolling at the City College of New York in 1960 to study the philosophy of mind; he took his B.A. from City College in 1963.

The move from engineering to philosophy of mind to physiology is the interesting part. O'Keefe's original attraction was to the mind-body problem, and he has said in retrospect that he found the philosophical literature unsatisfying because it was disconnected from measurement. City College in the early 1960s was an inexpensive, highly competitive institution that sent a remarkable number of graduates on to doctoral work, and it gave him the credentials to move on.

He went to McGill University in Montreal, then a dominant center for physiological psychology under Donald O. Hebb. There he earned an M.A. in 1964 and a Ph.D. in physiological psychology in 1967, supervised by Ronald Melzack. His dissertation examined the response properties of single units in the amygdala of the freely moving cat. Two features of that training carried forward: the insistence on recording from animals that are actually behaving rather than anesthetized or restrained, and the habit of asking what a single neuron is for rather than merely what it responds to.

Academic and Professional Career

In 1967 O'Keefe moved to University College London as a United States National Institute of Mental Health postdoctoral research fellow, working with the neurophysiologist Patrick Wall. He never left. He was promoted to professor in 1987 and later became the inaugural director of the Sainsbury Wellcome Centre for Neural Circuits and Behaviour at UCL, a purpose-built institute focused on the kind of large-scale circuit recording his own work had pioneered. In 2014, at the invitation of the Mosers, he also took a part-time professorial chair at the Norwegian University of Science and Technology in Trondheim.

The decisive period came early. At UCL he shifted from the amygdala to the hippocampus, a structure that had become a focus of human memory research after the famous case of patient H.M. but whose function in animals was disputed. Using electrodes implanted in freely moving rats, O'Keefe and his student Jonathan Dostrovsky recorded single hippocampal neurons over long sessions. In 1971 they reported that particular cells fired reliably when the rat was in a particular part of the apparatus, and that different cells had different "places." Crucially, the firing was not a simple response to a visual cue: it persisted in darkness and depended on the configuration of the environment, which is what justified the word "map."

Through the 1970s O'Keefe built the case for the hippocampus as a spatial representation system, culminating in the 1978 book with Nadel. In the 1990s he extended the work in two directions. With Recce he showed in 1993 that as a rat crosses a place field the cell's spikes occur at progressively earlier phases of the theta oscillation — "theta phase precession" — a temporal code layered on top of the spatial one. With Burgess he showed in 1996 that place fields stretch, shift, and re-form when the walls of the environment are moved, and he and Burgess predicted the existence of boundary vector cells responding at fixed distances from environmental barriers. Cells with the predicted properties were later reported in the subiculum and medial entorhinal cortex.

His honors include election as a Fellow of the Royal Society (1992) and of the Academy of Medical Sciences (1998), the Feldberg Foundation Prize (2001), the Grawemeyer Award in Psychology (2006, shared with Nadel), the British Neuroscience Association Award (2007), the Federation of European Neuroscience Societies European Journal of Neuroscience Award (2008), the Gruber Prize in Neuroscience (2008), the Louisa Gross Horwitz Prize (2013, shared with the Mosers), and election to the US National Academy of Sciences (2016). He holds honorary doctorates from University College Cork, City College of New York, and McGill.

Major Contributions to Psychology

Place cells

The core finding is that hippocampal pyramidal cells in the CA fields have spatially localized firing fields. A given cell is nearly silent in most of an environment and fires at a high rate in one small region; together, a population of such cells tiles the space the animal occupies. O'Keefe's argument was that this population constitutes an internal representation of location that is allocentric — referenced to the environment rather than to the animal's own body — and that it is constructed rather than merely read off the senses, since it survives removal of any single sensory cue.

The hippocampus as a cognitive map

With Nadel, O'Keefe argued in 1978 that the hippocampus implements Tolman's cognitive map: a Euclidean, map-like representation supporting flexible navigation, including shortcuts and detours that could not have been learned as fixed response chains. The book went further, proposing that in humans the same system had been extended to support episodic memory, so that the amnesic syndrome following hippocampal damage could be understood as a loss of the spatial-temporal context in which events are placed. This is the part of the theory O'Keefe has continued to develop, and it is why he has long linked his work to Alzheimer's disease, where disorientation in familiar surroundings is often an early sign.

Theta phase precession and temporal coding

The 1993 work with Recce established that place cells carry information in the timing of spikes relative to the ongoing theta rhythm, not only in their firing rate. As an animal enters a place field the cell fires late in each theta cycle; as it crosses the field, firing shifts to progressively earlier phases. The result is that each traversal compresses a sequence of positions into a single theta cycle, a format well suited to synaptic plasticity and to the ordering of experience in memory.

Boundary geometry and predicted cell types

The 1996 work with Burgess was important methodologically because it made a quantitative prediction. By modeling place fields as a function of distance and direction from environmental boundaries, they predicted cells that would fire at a fixed distance from a barrier regardless of the animal's heading. The subsequent identification of boundary and border cells in the subiculum and entorhinal cortex turned that prediction into confirmed anatomy, and it fed directly into the Mosers' grid cell work and later models of the entorhinal-hippocampal circuit.

Major Publications

O'Keefe, J., & Dostrovsky, J. (1971). The hippocampus as a spatial map: Preliminary evidence from unit activity in the freely-moving rat. Brain Research, 34(1), 171–175.
O'Keefe, J., & Nadel, L. (1978). The Hippocampus as a Cognitive Map. Oxford: Clarendon Press.
O'Keefe, J., & Recce, M. L. (1993). Phase relationship between hippocampal place units and the EEG theta rhythm. Hippocampus, 3(3), 317–330.
O'Keefe, J., & Burgess, N. (1996). Geometric determinants of the place fields of hippocampal neurons. Nature, 381, 425–428.

Readers should be cautious with two titles sometimes attributed to O'Keefe. The Nobel Prize in Physiology or Medicine 2014 was awarded one half to O'Keefe and one half jointly to May-Britt Moser and Edvard Moser "for their discoveries of cells that constitute a positioning system in the brain." The Kavli Prize in Neuroscience 2014, awarded by the Norwegian Academy of Science and Letters, went to O'Keefe together with Brenda Milner and Marcus Raichle, "for the discovery of specialized brain networks for memory and cognition" — a different citation, and different co-recipients, from the Nobel. The 2006 Grawemeyer Award in Psychology was shared with Lynn Nadel.

Impact and Legacy

O'Keefe's discovery rewired a field. Before 1971, the hippocampus was discussed mainly through lesion studies in humans and rats; afterward it became possible to ask what individual cells in the structure represent, and the tools he developed — chronic tetrode recording in freely moving animals, rigorous behavioral control, quantitative spatial analysis — became standard. The work also made spatial navigation a mainstream model system for studying how circuits compute, and it gave Alzheimer's research a cellular handle on why patients get lost. It is no exaggeration that the modern fields of spatial cognition, memory consolidation, and entorhinal circuit physiology descend from this one line of experiments.

Theoretical resistance has been substantial, and it is worth stating plainly rather than smoothing over. The cognitive map theory has never achieved universal acceptance. In 1993 Howard Eichenbaum and Neal Cohen proposed relational memory theory, arguing that the hippocampus binds arbitrary relations among the elements of experience and that spatial coding is one application of a general relational mechanism rather than the structure's defining job. Empirical support came from recordings showing that hippocampal cells encode non-spatial variables — odors, objects, task phase, elapsed time — in tasks where navigation is irrelevant, and from findings that place fields are not simply coordinates in a unified map: most are controlled by subsets of cues, and they can remap independently of one another. Jacobs's 2003 parallel map theory offered yet another alternative, and reviews by Eichenbaum and Cohen in 2014 concluded that the two literatures remain only partly reconciled.

O'Keefe and his allies have answers to each point. O'Keefe and Nadel had already conceded in 1978 that language might extend the hippocampal system to more abstract relations in humans, so the existence of non-spatial coding was not, for them, a refutation. O'Keefe's own boundary-vector work is itself evidence that he never held a naive "the cell is a GPS coordinate" view: his 1996 model is explicitly about how fields are constructed from geometric inputs. And in interviews he has pushed back on the popular "inner GPS" label, insisting that the cognitive map is a framework for memory and not merely a positioning device. On the other side, Eichenbaum has argued that calling these neurons "relational cells" rather than place cells is the more accurate description, and that a theory flexible enough to encode relations among anything risks explaining everything and predicting little.

The honest verdict is that the discovery is settled and the interpretation is not. Place cells, theta phase precession, grid cells, and boundary cells are all replicated findings. Whether the right summary sentence is "the hippocampus builds a map of space" or "the hippocampus builds a map of relations, of which space is one instance" remains an open, productive dispute — and it is a dispute that exists only because O'Keefe's experiments made the question empirically tractable.

Learn More

References

O'Keefe, J., & Nadel, L. (1978). The Hippocampus as a Cognitive Map. Oxford University Press.