IM this article to a friend!

November 6, 2002

Memory miscalculation foils IQ

From: Nature.com, UK
Nov. 6, 2002

Neuroscientists challenge tenets of intelligence testing.

HELEN PEARSON

Many people underscore on IQ tests because the benchmark memory test is inaccurate, a US researcher told the Society for Neuroscience meeting in Orlando, Florida yesterday. Another announced that women's brain size could affect IQ.

In standard intelligence tests, subjects are asked to remember a string of random numbers. The widely quoted average before stumbling - seven, give or take two - is thought to reveal the capacity of our short-term memory.

This 'magic number' is a huge overestimate, claims Mrim Boutla of the University of Rochester in New York. She puts the real size of short-term memory at four digits, plus or minus one - so too do several other studies challenging the gold standard.

Boutla and her colleagues asked 40 deaf and hearing signers to repeat strings of letters in American Sign Language - short, easy signs equivalent to spoken numbers.

Signers recalled fewer digits, but their memory was otherwise equal, she showed.

Unlike signs, spoken digits are taped on the brain's short recording loop, explains Boutla, which may not reflect the underlying ability to process language. Chinese speakers, for example, can store eight or nine clipped digits, while wordy Welsh words are limited to five or six.

Deaf people, and non-native English speakers, are being marked down unfairly in IQ tests, Boutla reckons. Testing working memory - how we recall and use words - is a fairer estimate of language-processing skill, she says. "You have to create new norms."

Sex and size

"There's always controversy about IQ tests," says Debra Kigar of McMaster University in Hamilton, Canada. Joining the fray at the meeting, Kigor presented evidence that women's brain size, but not men's, correlates with IQ. Such suggestions are generally controversial as they can smack of sexual or racial profiling.

Kigar used a unique brain bank: she measured the volume of 100 post-mortem frontal brains from the Witelson Normal Brain Collection based at McMaster University. The brains were donated by cancer patients whose IQ had been carefully tested.

In women, "there's a direct relation between size and function", Kigar and her colleagues found, particularly in language tests. "In theory, if you've got a bigger brain, you'd be a great writer or thinker," she says.

Women are thought to use their entire brain for language; men use one half and either the front or back. Thus, changes in brain volume may have more impact on women's verbal ability than men's, speculates Kigar.

In agreement with previous brain-scanning studies, she also found that brain size decreases with age in men, but not women. Exploring sex differences in brain size and structure could help our understanding of why women tend to recover better after stroke, hopes Kigar


© Nature News Service / Macmillan Magazines Ltd 2002