Science Watch

Can you remember the details of your wedding? Do you recall what you had for breakfast? If so, you have your hippocampus to thank. A ridge at the bottom of each side of the brain, this small structure encodes experiences for storage in other brain regions.

Many people with damage to their hippocampus are able to recall memories prior to their brain damage but cannot create new ones--a phenomenon that's led some scholars to believe that this brain structure serves as the gateway to all aspects of autobiographical memory. But new research published in the July issue of Behavioral Neuroscience (Vol. 118, No. 3) suggests that some aspects of memory may bypass the hippocampus entirely.

"We found [rats] without the hippocampus are still able to learn to associate smells with rewards after a single experience," says study author Emma Wood, PhD, a neuroscience lecturer at the University of Edinburgh in Scotland.

This means that areas of the brain previously not considered to be involved in encoding memories, such as a more primitive part of the brain above the hippocampus known as the perirhinal cortex, may encode or store simple sensory associations, Wood notes.

And these findings could lead memory researchers beyond the hippocampus to the study of interactions between brain structures, says Raymond Kesner, PhD, a psychology professor at the University of Utah. The line of research, Kesner notes, holds promise for a broader understanding of memory, potentially bringing new insight to the processes of Alzheimer's disease, which can devastate the hippocampus and surrounding areas.

Simple associations

One of the best ways to determine the function of brain structures is to damage them and then observe the deficits that result, says Wood. However, tapping into an animal's memory for specific events can be challenging, she notes.

"This kind of memory really lends itself to verbal descriptions," says Wood. "So we had to create a task where the rats could express knowledge gained from a single event."

To do this, Wood and her colleagues began by operating on 19 rats, injecting a neurotoxin into some of the hippocampi and injecting saline into the brains of the control group.

After the animals recovered, the researchers filled eight cups with sand, each scented with a different odor, such as allspice, coffee or dill. The researchers buried a Froot Loop in the sand of three of the cups, then gave each rat a chance to dig in all of the cups. After that, they waited 30 seconds to two hours, then tested the animals' memory by presenting them with the five odors not previously associated with Froot Loops and one that was. If the rat remembered the earlier experience, it would dig first in the rewarded odor's cup, Wood explains.

"We initially hypothesized that the animals with hippocampal damage wouldn't be able to do [the task], and we were quite surprised when they could," says Wood.

Both groups dug in the cup with the previously rewarded scent about 75 percent of the time when the delays were 15 minutes or less, she reports. However, when the delay was extended to two hours, all of the animals performed at near-chance levels.

Creating context

These findings, however, do not knock the hippocampus from its perch as the most important brain structure involved in creating memories, say some researchers.

"Wood's study shows that the hippocampus may not be involved in creating simple associations," says Kara Agster, a third-year graduate student in Brown University's neuroscience program. "But we know that it is required for learning contextual information, like place and time."

Agster demonstrated such an effect in a study published in a 2002 issue of Nature Neuroscience (Vol. 5, No. 5). In this research, rats performed a task similar to the one used in Wood's trials. However, success in Agster's task required memory not only for a single smell, but for the order in which a smell occurred in a succession of scents, she explains. Before operating on 10 rats, the researchers let the animals dig in six scented cups of sand, all containing a Froot Loop. Then, the researchers allowed the rats to dig in cups containing six different smells, each of which also held a reward. After this training, the researcher damaged half of the rats' hippocampi and presented all of the animals with a scent from each of the two groups. For example, if group one contained scents A, B, C, D, E and F and group 2 consisted of L, M, C, D, P and Q, an animal might sniff scents A and L but only find a reward in scent A's container.

After a short delay, the researchers presented another two scents--this time it might be scents B and M, with scent B rewarded. To correctly solve this puzzle, rats had to remember that scent B occurred in the same group as scent A, and make a connection between all the scents in the first group and rewards.

The researchers found that rats with working hippocampi were able to discern this pattern, getting between 70 and 95 percent of the tasks correct depending on the time lag between trials. The animals with hippocampal damage, however, got only about 50 percent correct, regardless of the delay.

"With this task you have to use memory of an individual smell in context with other information," says Agster. "This suggests that the hippocampus is required to place an event in time and understand its relationship with other memories."

Agster and Wood's findings dovetail with those of Kesner at the University of Utah. According to Kesner, the hippocampus may serve to integrate information recorded in many different areas of the brain. For example, he says, the perirhinal cortex may be responsible for forming basic associations between smell and reward, but the hippocampus could pair that association with other kinds of information, such as where the smell occurred or when it happened--allowing the animal to apply past learning flexibly to new situations.

"All of this research points to the idea that the hippocampus is not unique in terms of forming associations--other parts of the brain can do that as well," he says. "What makes the hippocampus so special is that it integrates stimuli--visual, auditory and spatial."

These elements, Kesner notes, are the ones that, in humans, make up what we consider episodic memory--a recollection of one's personal life events.

And determining the way the brain encodes events may eventually shed light on memory breakdowns when they occur, suggests Agster.

"With these studies we may eventually answer questions about how memory is processed and tap into more of what is going on with Alzheimer's disease," she notes.

Further Reading

  • Fortin, N., Agster, K., & Eichenbaum, H. (2002). Critical role of the hippocampus in memory for sequences of events. Nature Neuroscience, 5(5), 458-462.

  • Kesner, R., Gilbert, P., & Barua, L. (2002). The role of the hippocampus in memory for the temporal order of a sequence of odors. Behavioral Neuroscience, 166(2), 286-290.

  • Wood, E., Agster, K., & Eichenbaum, H. (2004). One trial odour-reward association: A form of event memory not dependant on hippocampal function. Behavioral Neuroscience, 118(3), 526-539.