How the brain builds a memory

Reuters News

Activation of particular parts of the brain -- and the levels of activity in these brain regions -- determine which facts and events will be remembered, according to two studies published this week in the journal Science.

The findings "may give us a new tool to objectively measure what is memorable for the brain," said Dr. John Gabrieli, senior author of one study and associate psychology professor at Stanford University in Stanford, California. He believes the study results have implications for "everything from curriculum development in schools to advertising."

Scientists have long wondered why -- and how -- the brain retains a memory of some incoming data, but allows other information to be forgotten. The recent development of functional magnetic resonance imaging (fMRI), which allows scientists a highly-focused view of brain activity over time, is providing new answers to these old questions.

In one study, Gabrieli and others at Stanford focused their efforts on visual memories. Subjects placed under fMRI viewed and then re-viewed a series of pictures. The study authors found that activity levels in two brain regions -- the right prefrontal lobe and the parahippocampal cortex -- measure "how well a particular visual experience is encoded and therefore predicts whether it will be remembered well."

A second study, led by Dr. Anthony Wagner of the Harvard Medical School in Charlestown, Massachusetts, focused on verbal, or linguistic memory. This time, subjects were asked to remember words, either by their meaning or by their appearance (upper or lower case spelling).

Reflecting the results of the Stanford study, the Harvard team found that fMRI imaging showed that activity levels in the left prefrontal cortex and the parahippocampal cortex dictated which words were remembered or forgotten in subsequent tests. Furthermore, they discovered that words were much more likely to be remembered when subjects concentrated on semantics (meaning), rather than on their appearance. "Subsequent memory was superior following semantic (85% recognized) than following nonsemantic (47% recognized) processing," the authors report.

Dr. James Brewer, lead author of the visual-memory study, points out that a word's meaning could be triggered by a myriad of impulses. "Perhaps one person would see a photograph of Zion National Park and think, 'Hey, I just visited that place on my way to California!' while another would think, 'Outdoor desert scene... what's the next scene going to be?" He believes the first viewer would be more likely to remember the image later on.

The findings may help researchers find new treatments for diseases that affect the memory, such as Alzheimer's disease. "These findings might give us a tool to examine the very earliest effects" of the disease, Brewer explained in a statement released by Stanford.

Gabrieli agrees, adding that new research should determine whether memory-imaging techniques might be used in "older people at risk for Alzheimer disease (to) see if they predict who will and who will not get the disease."

SOURCE: Science 1998;281:1185-1187, 1188-1190.