Scientists' Experiments Confirm Theory Central to Memory Research
Sept. 20, 2002
Scientists have experimentally supported a central tenet of a theory regarding how new memories are converted from a short-term, unstable form, into stable, long-term memory. Experimental evidence at the University of Arizona in Tucson shows that during quiet time or sleep, some brain cells replay recent events, possibly to consolidate memory collected from various sources.
For some time, neuroscientists have believed that new, or "short-term memories" consist of separate parts, which are stored in different regions of the cortex, depending on the type of information involved, and that consolidation into "long-term memory" involves formation of direct associative links among these different parts.
According to this theory, these individual parts of the memory are initially linked, indirectly, by virtue of the common connections they share between each cortical region to a higher region called the hippocampus. As the memory is replayed during rest, scientists believe that direct links form among the various parts, thus making the memory independent of the links with the hippocampus, which then become reused for linking new memories.
This theory tries to explain why old memories are resistant to disruption by damage to the hippocampus, such as might occur as a result of brain injury or degenerative diseases like Alzheimer's disease, as opposed to new memories which simply tend to be lost.
Kari L. Hoffman and Bruce L. McNaughton report ("Science," Sept. 20) new evidence supporting this theory. Hoffman and McNaughton found that when one part of a recent memory is replayed in its cortical location, the other parts are, in fact, concurrently replayed in their cortical locations. This "concurrent reactivation" of brain cells is thought to be essential if the correct pieces of memory are to be linked together into a coherent whole. How the UA researchers studied this phenomenon is as interesting as the subject matter itself.
Memory is a shorthand term for the process the brain uses to encode, store and retrieve information. Most organisms, including humans, depend on remembering learned behavior and past events, and the breakdown of any of these processes reflects memory failure.
Humans are especially able to recall memories, even from the distant past, often in vivid detail, suggesting that we have a very high-capacity storage system. But recent memories are susceptible to disruption, especially in the first minutes to days after an event. The researchers say this period of instability "may be a consequence of the way memory traces are stored throughout the cortex." How these fragile new memories become stabile, long-term memories has long been a puzzle.
Researchers have theorized that brain cells in the hippocampus that were active while an event occurred will reactivate, sometimes repeatedly, as a way of assembling information for long-term memory storage. This activity may elicit related activity from lower level brain cells in the cortex that also were active at the time of the event.
"Through repeated coactivation, these lower-level ensembles may create the connections necessary to encode the memory trace efficiently and to sustain it, or some approximation of it, independently of top-down input," says McNaughton, a professor of psychology at the UA.
McNaughton says two critical predictions follow the theory. One is that "Patterns of neural ensemble activity expressed during an experience should be spontaneously reactivated during subsequent periods of behavioral inactivity." The second holds that "The distributed components of the reactivated memory trace should appear concurrently within the relevant cortical sites."
Using a sophisticated array of micro electrodes to monitor brain activity during the administration of various tasks, Hoffman, a graduate student at the UA, and McNaughton were partly able to address the question. They theorized that if reactivation occurs, "cells that were active together during the task should tend to be coactive afterward, and cells active at different times during the task should not be coactive afterward."
The results of their study demonstrated that memory trace reactivation occurs in a "coherent, distributed manner across much of the neocortex." The results, they say, indicate that "the observed effect is not simply an uninterrupted persistence of a previously expressed activity state, but rather reflects the reemergence of recent patterns."
Also significant is the technology used to study this and other ongoing memory research at the University of Arizona. McNaughton has been conducting this and related research for several years, using microelectrode arrays and computers he and his colleagues have developed specifically for this kind of inquiry.
The experimental results reported in "Science" was made possible by an array of 144 independently advanceable microelectrodes that were able to measure input from four regions of the cortex. Each array consisted of a 12 x 12 lattice of electrodes spaced about 6 and a half millimeters apart. A total of 800 cells were recorded from over nine recording sessions, producing a total of more than 21,000 cell pairs for study. The ability McNaughton and his collaborators have to measure such large collections of individual brain cell activity is unmatched in any other laboratory, and is a necessary prerequisite to performing the sort of analysis they undertook.
While Hoffman and McNaughton were able to demonstrate that memory trace reactivation is temporally ordered and concurrent across large areas of the neocortex, what remains to be shown are the actual mechanisms leading to this phenomenon, and that this coherent memory trace reactivation is actually involved in memory consolidation. Their findings, however, are an important step in the process of confirming the trace-reactivation theory of memory.
University of Arizona in Tucson