NYT: Smart brains develop differently, scans show

[marcus]

http://www.nytimes.com/2006/03/30/science/30brain.html?_r=2&oref=slogin&pagewanted=print

Scans Show Different Growth for Intelligent Brains

By NICHOLAS WADE
The brains of highly intelligent children develop in a different pattern from those with more average abilities, researchers have found after analyzing a series of imaging scans collected over 17 years.

The discovery, some experts expect, will help scientists understand intelligence in terms of the genes that foster it and the childhood experiences that can promote it.

"This is the first time that anyone has shown that the brain grows differently in extremely intelligent children," said Paul M. Thompson, a brain-imaging expert at the University of California, Los Angeles.

The finding is based on 307 children in Bethesda, Md., an affluent suburb of Washington. Starting in 1989, they were given regular brain scans using magnetic resonance imaging, a project initiated by Dr. Judith Rapoport of the National Institute of Mental Health.

This set of scans has been analyzed by Philip Shaw, Dr. Jay Giedd and others at the institute and at McGill University in Montreal. They looked at changes in the thickness of the cerebral cortex, the thin sheet of neurons that clads the outer surface of the brain and is the seat of many higher mental processes.

The general pattern of maturation, they report in Nature today, is that the cortex grows thicker as the child ages and then thins out. The cause of the changes is unknown, because the imaging process cannot see down to the level of individual neurons.

But basically the brain seems to be rewiring itself as it matures, with the thinning of the cortex reflecting a pruning of redundant connections.

The analysis was started to check out a finding by Dr. Thompson: that parts of the frontal lobe of the cortex are larger in people with high I.Q.'s. Looking at highly intelligent 7-year-olds, the researchers said they were surprised to find that the cortex was thinner than in a comparison group of children of average intelligence.

It was only in following the scans as the children grew older that the dynamism of the developing brain became evident. The researchers found that average children (I.Q. scores 83 to 108) reached a peak of cortical thickness at age 7 or 8. Highly intelligent children (121 to 149 in I.Q.) reached a peak thickness much later, at 13, followed by a more dynamic pruning process...

Nicolas Wade is a good science journalist but we should keep skeptical about this, it is the first study AFAIK of this sort and it will have to be repeated and there will probably be some critical reaction. but I thought it might be interesting to mention at PF anyway.

 

[sid_galt]

First I would like to see proof that IQ = intelligence.
Almost anyone can improve his IQ through practice. That does not mean practice = intelligence.

 

[marcus]

First I would like to see proof that IQ = intelligence.
Almost anyone can improve his IQ through practice. That does not mean practice = intelligence.

that is something that I don't require, sid. I am not interested here in "intelligence" in any profound sense, only in something they can measure.

general "intelligence" is too vague an idea-----like "virtue" or "creativity" or "sympathy" or "courage"----to be useful in some research like this.

but here they had some rough measure namely performance at age 7 on a certain test (which the kids had presumably NOT practiced and trained for)

this performance measure (whether or not you think it corresponds to anything valuable to the individual or society) CORRELATED with having a thinner layer of cells at age 7 and a scenario of development delayed for several years in reaching max thickness

the ones who scored high on the test had thinner cortical layers and didnt get max thickness until age 12 or 13, whereas the others developed earlier to max at age 7 or 8 IIRC

OR SO THE RESEARCHERS CLAIMED

to me that would be interesting, if true
It strikes me as surprising that, say, a multiple-choice test at age 7 predicts whether a certain cortex layer of cells will thicken rapidly or slowly
that seems really puzzling, could have various explanations

or?
or do you not find it interesting?

 

[zoobyshoe]

or do you not find it interesting?

I find it interesting and more persuasive, correctly or not, to look at the number of neurons and how they prune rather than cruder things like brain size, in the search for a physical correlate of intelligence. This suggestion of this study is that people who end up more intelligent spend a longer time building up more neurons which are then prunned more efficiently. More redundancy to begin with could allow for much more sophisticated pruning, at least in principle.

It suggests the analogy, which may or may not ultimately pertain, to the creation of a film or book: the more preliminary research, or feet of footage shot, the more the writer or editor has to work with to create an effectively compact product.

Of course, that begs more questions about what guarrantees, or, at least, governs, the quality and ultimate greater compactness of the editing? Is quality and trimness somehow an automatic concommitant of the slower buildup of greater quantity? Or does a pre-existing "editor" (whatever that might be, a gene, posssibly) cause the slower buildup of greater quantity and subsequent more dramatic pruning back?

Anyway, I do find the whole thing interesting and am glad you posted it.

 

[sid_galt]

or do you not find it interesting?

I find it very interesting but IMO, to draw any conclusions, the first research that needs to be done is what IQ really means. Is it determined genetically or environmentally? Can it be improved to highs like 160-180 by an ordinary person or is that capability determined at birth? What relation does that capability have to thinking and creativity/passion/skill etc.

Because without that, the whole experiment would have little value.

 

[selfAdjoint]

I find it very interesting but IMO, to draw any conclusions, the first research that needs to be done is what IQ really means. Is it determined genetically or environmentally? Can it be improved to highs like 160-180 by an ordinary person or is that capability determined at birth? What relation does that capability have to thinking and creativity/passion/skill etc.

Because without that, the whole experiment would have little value.

Sid, all that research has been done in spades years ago. Read Arthur Jensen's magesterial book The G Factor. Repace the test-dependent idea of IQ with the intertest statistic g, we can say:

• g is between 50 and 80 percent inherited
• g is very strongly correlated with physical measures like reaction time
• different subpouplations have different g distributions
• g is heavily correlated with ability to learn and plan rationally

Recent studies using functional MRI have shown that differences in g correlate with different amounts of gray matter in the frontal cortex.