Some people are better than others at remembering what they have just seen — holding mental pictures in mind from moment to moment. An individual’s capacity for such visual working memory can be predicted by his or her brainwaves, researchers funded by the NIH’s National Institute of Mental Health have discovered.

A key brain electrical signal leveled off when the number of objects held in mind exceeded a subject’s capacity to accurately remember them, while it continued to soar in those with higher capacity, report University of Oregon psychologist Edward Vogel, Ph.D., and graduate student Maro Machizawa, in the April 15, 2004, Nature.

Analogous to a computer’s RAM, working memory is the ever-changing content of our consciousness. It’s been known for years that people have a limited capacity to hold things in mind that they’ve just seen, varying from 1.5 to 5 objects. “Our study identifies signals from brain areas that hold these visual representations and allows us to coarsely decode them, revealing how many objects are being held and their location in the visual field,” explained Vogel.

To find out if the amplitude of detectable signals reflects the number of object representions held in visual memory, the researchers presented 36 subjects with a series of trials containing an increasing number of objects. Subjects briefly viewed a picture containing colored squares, followed by a one-second delay, and then a test picture. They pressed buttons to indicate whether the test picture was identical to — or differed by one color — from the one seen earlier. The more squares a subject could correctly identify having just seen, the greater his/her visual working memory capacity. Subjects averaged 2.8 squares.

Electrodes on the scalp recorded neural activity during the one-second delay to pinpoint signals reflecting activity of brain areas involved in holding the images in working memory. Asking subjects to remember just one of two sets of colored squares that appeared on the left and right sides of the screen revealed signals near the opposite rear side of the head as emanating from the brain area involved.

The researchers found that the more squares a subject correctly identified, the higher the spike of corresponding brain activity — up to a point. Amplitude of the signal for correct trials was much higher than incorrect ones, suggesting that the delay activity specifically reflects the maintenance of successful representations in visual memory. Neural activity of subjects with poorer working memory scores leveled off early, showing little or no increase when the number of squares to remember increased from 2 to 4, while those with high capacity, who correctly remembered more squares, showed large increases.

Using a similar task with functional magnetic resonance imaging (fMRI), a research team at Vanderbilt University reports in the same issue of Nature that the posterior parietal cortex, an area at the top rear part of the brain, is the brain area responsible for holding representations in visual working memory — and likely source of the signal in the Oregon study.

"Simply by measuring the amplitude increase across memory array sizes, we can accurately predict an individual's memory capacity,” said Vogel.

Since working memory capacity is strongly predictive of performance on a broad array of cognitive abilities — reasoning, language, flexible problem solving — Vogel foresees the physiological measure as finding applications in assessing individuals who are behaviorally or verbally impaired, such as in cases of stroke or paralysis. The technique has also been used to study development of cognitive abilities in pre-verbal children.

NIMH is part of the National Institutes of Health (NIH), the Federal Government's primary agency for biomedical and behavioral research. NIH is a component of the U.S. Department of Health and Human Services.

Signal Rises with Memory Capacity Graph (http://www.nimh.nih.gov/ImageGallery/Press/prwmprowress1.jpg)