Professor John O'Keefe's work on how the brain tracks location stands as one of the clearest examples of single-neuron recordings turning a vague idea into a measurable system. Born in 1939 in New York to Irish immigrant parents and raised in the South Bronx, he followed a route that mixed engineering, philosophy, and physiology before settling into decades of experiments at University College London.

The 1971 finding that certain hippocampal neurons fire only when a rat occupies one spot in its environment gave the field its first concrete component of an internal positioning system. That observation, made with student Jonathan Dostrovsky, appeared in a short communication in Brain Research and set the stage for everything that followed.

O'Keefe arrived at UCL in 1967 after earning his doctorate at McGill, where he had recorded from the amygdala. He shifted focus to the hippocampus after noting how damage there disrupted spatial tasks in rats. The lab setup involved chronically implanted electrodes that let the team monitor individual cells while the animal moved freely through simple enclosures. Place cells emerged when the team noticed firing rates tied tightly to the animal's position rather than to its speed, direction, or the task itself.

One dataset that still shapes expectations comes from the original recordings: a given cell might activate across a few dozen square centimeters in a small box, with neighboring cells covering adjacent patches so the population collectively tiles the space. That pattern held across repeated trials in the same environment but remapped when the box shape or landmarks changed.

Photo by Brett Jordan on Unsplash

A collaborator in a neighboring lab once described spending weeks reconstructing electrode tracks after a postdoc left with the raw files on a single external drive. The episode illustrates how much of the early place-cell work depended on meticulous note-taking and shared lab notebooks rather than automated pipelines. Replicating those 1971 conditions today still requires the same patience with signal-to-noise ratios and the same willingness to discard sessions when movement artifacts swamp the spikes.

The 1978 book with Lynn Nadel laid out the broader claim that the hippocampus builds a cognitive map usable for navigation and memory. The text spelled out predictions about how place cells should behave under cue rotation or in the dark, predictions that later experiments tested one variable at a time.

Head-direction cells, which signal facing angle independent of location, turned up in O'Keefe's lab in the mid-1980s. Boundary cells that mark proximity to walls appeared later. These additional cell classes showed the positioning system was not a single map but a set of interlocking signals.

Grid cells, discovered by May-Britt and Edvard Moser in 2005 in the medial entorhinal cortex, supplied the metric layer. Their regular hexagonal firing fields provide distance and direction information that place cells can anchor to specific sites. The three researchers shared the 2014 Nobel Prize in Physiology or Medicine for mapping these components of the brain's GPS.

Photo by Vitaly Gariev on Unsplash

What this means for any lab recording from freely moving animals is straightforward: stable behavioral tracking and stable single-unit isolation remain the rate-limiting steps. Ranges matter more than single numbers. Place-field sizes vary from roughly 20 cm in small arenas to several meters in larger ones, and stability across days sits between 60 and 90 percent depending on the environment and the interval. The practical habit is still the same one O'Keefe's group used: run the animal on the same track or in the same box for enough trials to separate signal from drift, then archive both the spike times and the position data in the same folder with a one-page methods note.

In 2026 O'Keefe published a reflective account of the 1971 discovery in the journal Hippocampus, walking through the sequence of experiments that ruled out alternative explanations such as simple sensory responses. The piece underscores how incremental the work was and how many control sessions were required before the place-field pattern became convincing.

The system matters outside the lab because the same hippocampal circuits are among the first affected in Alzheimer's disease and other disorders that impair wayfinding. Patients lose the ability to form new spatial memories long before other cognitive domains decline, consistent with the selective vulnerability of place and grid networks.

One concrete next step is to read the 2026 Hippocampus paper and compare its description of the original electrode placements with current silicon-probe methods. The contrast shows both how much the core observation has held up and how much the tools have changed.