The Journey to the Cognitive Map

The British Neuroscience Association celebrates its 50^th birthday this year, and in honor of the occasion, Nobel Prize winner John O’Keefe delivered a special anniversary lecture at the organization’s annual meeting in Edinburgh earlier this month.

O’Keefe was one of the founding members of the British Neuroscience Association (BNA), and later acted as its president from 1977 to 1980. He told the story of how the BNA was established, and then went on to describe his early work on spatial navigation, which eventually led to him being awarded the 2014 Nobel Prize in Physiology.

He obtained his Ph.D. from McGill University in 1967, under the supervision of psychologist Ronald Melzack, then moved to University College London almost immediately afterwards, to accompany Melzack as a postdoctoral fellow.  Several years earlier, Steven Rose, Patrick Wall, and John Wolstencroft had set up the Brain Research Association, which aimed to bridge the gap between anatomy and physiology, and convened for small, informal meetings in an upstairs room at the Black Horse pub on Rathbone Place.

O’Keefe and Melzack got involved in the organization soon after their arrival to London. “I started off as the gopher for these meetings, carting the screen and the slide projector around,” O’Keefe recalled. With additional funding from the Sloan Foundation, this loose organization of local groups soon expanded to include researchers from yet more disciplines such as psychology.  “We needed a more formal national organization, to represent British neuroscience to other societies around the world,” he said.

“I graduated to being secretary of the London branch and then went on to being national secretary and national chair,” O’Keefe said. There was the sense that a whole new discipline was emerging and that we were all part of it.”

Eventually, the organization changed its name to reflect its wider remit, and thus the BNA was born.

At that time, O’Keefe had become interested in the role of the hippocampus in memory, due largely to Brenda Milner’s work with the amnesic patient H. M. (See “One Man’s Continuing Contribution to the Science of Memory“). David Hubel, Vernon Mountcastle, and others had developed methods of recording the activity of single cells in the brains of live animals, but believed that recording from cells deep inside the brain would be futile, due to the complexity of the sensory systems, and restricted themselves to studying neurons involved in the earliest stages of sensory processing.

“I wanted to go straight for the jugular and stick the electrodes straight into the hippocampus,” said O’Keefe, “but everybody said I’d never make sense of something that was so far along a pathway which has [so many] thousands of converging synapses.”

His persistence quickly paid off. O’Keefe and his colleague John Dostrovsky recorded the activity of 60 neurons in the rat hippocampus, and in 1971 the pair published preliminary evidence of spatial cells.

“It took us a while to realize that the cells weren’t interested in what the animals were doing, but where the action was taking place,” said O’Keefe. “Not what, not why, but where the action is taking place. As the animal wanders around, the cells are silent most of the time, but when it goes over this patch, one of the cells wakes up and starts to fire.”

O’Keefe and Dostrovsky hypothesized that these neurons—now known as place cells—are one component of the brain’s navigational system. “We immediately jumped to the conclusion that this might be the neural substantiation of a cognitive map, on the basis of the small number of cells. In retrospect, it was quite a jump.”

The findings were initially met with opposition, but O’Keefe was undeterred, and pursued the idea nonetheless. He elaborated on the idea in a 1978 book called The Hippocampus as a Cognitive Map, co-authored with Lynn Nadel, and, working with lab members and collaborators, devised new ways of recording from cells in the brains of freely moving animals, and of monitoring and testing the animals’ behavior.

Using these new methods, O’Keefe and his colleagues continued to characterize the properties of place cells, and evidence for the cognitive map began to mount. In 1982, they reported that place navigation is impaired in rats with hippocampal lesions, and went on to predict the existence of other types of spatial cells in the hippocampus and surrounding areas.

Many of the predictions have since been borne out, and the initial discovery of place cells has led to a host of other discoveries about the brain’s navigational system. We now know, for example, that the hippocampus and surrounding brain areas contain head direction cells, which fire only when the animal is facing a certain direction, as well as boundary vector cells, which fire only when it is a given distance from borders in its environment.

Another important discovery came in 2005, when Edvard and May-Britt Moser and their colleagues at the Norwegian University of Science Technology in Trondheim and their colleagues identified grid cells, which fire periodically as the animal traverses a space. Grid cells are arranged in an orderly fashion, such that those located toward the bottom of the brain have a larger “scale” than those further up.

Place cells and grid cells thus comprise key components of the brain’s global positioning system, and O’Keefe and the Mosers shared the 2014 Nobel Prize in Physiology or Medicine in recognition of their discoveries, which have taught us a great deal about the cellular basis of spatial navigation.

Others have furthered our understanding of how the navigational system operates in three-dimensional space, and there is now evidence that some of the cell types involved in spatial navigation are fully formed before animals have had a chance to explore their environment, suggesting that the cognitive map is innate, or present at birth.

Yet, we still don’t know how these spatial cells cooperate to generate maps of the environment and help us to navigate it successfully. The initial discovery of grid cells suggested that they form some kind of metric system that feeds relevant information to place cells. Subsequent research suggests that this is not the case, however, and so, despite all the progress that has been made so far, the journey to the cognitive map is still strewn with obstacles.

“We all thought grid cells were going to be the brain’s metric system, but some of our key assumptions about them turn out to be invalid,” said O’Keefe. “Their scale changes from one environment to another, and you can’t have a ruler that changes as you move around the world, so there’s still quite a bit of work to be done.”