At any given moment, a few neurons are ready to take charge of the next piece of information to be stored in the brain's hippocampus.

New research suggests that at any given moment, a few select neurons in the brain are at the ready, primed to take charge of the next piece of information to be stored in the memory-forming region of the brain known as the hippocampus.

The study, conducted in rodents as they explored a new environment for the first time, examined neurons that scientists call place cells: cells that fire when an animal is in a particular location, creating a mental map of its space. The results, published April 14, 2011, in the journal Neuron, offer a new model for how the brain forms memories, according to Janelia Farm group leader Albert Lee, who led the study.

[The brain] has a certain pattern that it wants to have for the next memory that it gets. It doesn’t care so much what the particular details are of that thing, it just wants to assign it.

Albert Lee

Place cells, which have been best studied in rodents, are thought to behave similarly to cells in the human hippocampus that form memories of people, places, facts, and events—a type of memory called declarative memory. Lee says understanding what makes certain cells become place cells in a rodents should help explain how memories are stored in the human brain.

In both humans and rodents, the hippocampus lies deep within the brain, where it receives sensory information relayed by neighboring cells. Scientists first began to suspect that this part of the brain was involved in memory in the 1950s, when they observed that a patient who had part of his brain removed to treat epilepsy was unable to form new declarative memories. Since then, researchers have refined their understanding of the role of the hippocampus. The brain structure is believed to be essential for forming new memories of people, places, and events, as well as the spatial memory people and animals use to navigate their environments. Lee says that the hippocampus is thought to be necessary for converting short-term memories into long-term memories.

In a revealing experiment conducted in 2005, a group of scientists at Caltech found specific cells in the hippocampus of patients that fired whenever the person was presented with an image of—or even just thought about—a particular person or famous building. The same cells remained silent when a picture of a different person or object was presented, suggesting that the brain might assign specific neurons to be responsible for concepts such as “Jennifer Aniston” or “leaning tower of Pisa.” Of the millions of cells in the human hippocampus, it is estimated that about 0.5 percent might fire for a particular object or fact.

Similarly, scientists have found that specific cells in the hippocampus seem to identify selectively with certain spatial locations. As a rodent explores a new environment, such as a maze, the cells in its hippocampus begin to map out the space. In any given maze, about one-quarter to one-half of the cells fire, while the others remain silent—a pattern that becomes more distinct and reliable each time the animal passes through that spot, reinforcing its spatial memory. Lee says recordings of neural activity taken as humans navigate a virtual reality world suggest that they, too, have place cells that function similarly.

Based on the firing patterns researchers had detected in rodents, it was clear that individual cells in the hippocampus take on the role of place cell in some environments, and remain silent in others. Lee was curious what determined which cells become place cells in a particular environment. “We wanted to investigate the cellular basis of this difference,” he says. “What makes some cells fire and some cells not fire in a given place?” For humans, he said, the question might translate to “Why do particular cells end up representing a particular object or person?”

For a neuron to fire, its membrane potential—essentially a difference in charge between the interior and exterior of the cell—must exceed a certain threshold level. On its own, a nerve cell has a baseline membrane potential known as its resting potential. Excitatory inputs from neighboring cells boost this membrane potential bit by bit, and when the threshold is reached—usually thought to require hundreds of inputs—the neuron fires.

Do place cells fire because they receive more input than silent cells? Or do certain cells have some intrinsic predisposition to firing more readily in a particular environment? To find out, Lee wanted to measure the electrical properties of individual cells in the hippocampus as a rodent explored a maze for the first time.

Lee explains that all but a few studies of place cell activity had, so far, only tracked the firing of individual cells. “You can tell whether a cell spikes or not, but you can’t tell what’s going on inside the cell to make that cell spike,” he says. Researchers had not been able to detect the inputs that cells received or how much it took to nudge them above their firing threshold. But as a postdoctoral researcher in 2006 working in Michael Brecht’s lab at the Erasmus Medical Center in the Netherlands, Lee and Brecht had developed a method for attaching electrodes to individual neurons so that their intracellular activity could be detected in a freely moving animal. So Lee set up an experiment in which he and collaborator Jérôme Epsztein, whose lab is at INMED-INSERM in France, could determine the properties of cells in the hippocampus as rats explored a simple, oval-shaped maze for the first time.

By comparing the rapidly firing place cells to the silent cells (which Lee explains do fire, but rarely), he and his team found that the inputs received by place cells were more excitatory than those received by silent cells. More surprisingly, the place cells also had a significantly lower threshold for firing than the silent cells. “That threshold will control how hard it is for a cell to reach a spiking level,” Lee notes.

Lee and his colleagues also tested the cells’ response to a stimulus supplied before the animal began exploring its new environment, and found that future place cells reacted with a distinctly different firing pattern than cells that would remain silent. This suggests some predetermination as to which cells will become place cells, even before the animal experiences the maze, he says.

Taken together, Lee says his team’s findings suggest that there is a bias in the hippocampus for which cells will take charge of the next memory—regardless of what kind of sensory input arrives. “It has a certain pattern that it wants to have for the next memory that it gets,” he says. “It doesn’t care so much what the particular details are of that thing, it just wants to assign it.”

By looking inside place cells for the first time, Lee’s team has offered a more detailed explanation for how cells in the hippocampus form memories—in a way, he says, that may be unexpected to many neuroscientists. At the same time, the findings open up new sets of questions—such as how a memory pattern becomes fixed so that the hippocampus does not assign a new set of place cells to the same location when it is encountered again—that he is eager to explore.

Scientist Profiles

For More Information

Jim Keeley 301.215.8858