December 18, 2006

Flynn says Flynn Effect winding down

From the London Times (via GNXP):

We are about as smart as we're going to get, says IQ pioneer
Alexandra Frean, Education Editor

It is a common refrain, repeated in response to every new television reality show and every bumper crop of school exam results: society is dumbing down. Scientists have long argued the opposite, pointing to the now widely accepted “Flynn effect”, which shows that over the past century average IQ scores have improved across the developed world, irrespective of class or creed.

Now the man who first observed this effect, the psychologist James Flynn, has made another observation: intelligence test scores have stopped rising. Far from indicating that now we really are getting dumber, this may suggest that certain of our cognitive functions have reached — or nearly reached — the upper limits of what they will ever achieve, Professor Flynn believes. In other words, we can’t get much better at the mental tasks we are good at, no matter how hard we try. If we are to make any further progress, we will have to start exercising different parts of our brain, particularly the parts controlling language acquisition and empathy, according to Professor Flynn, an emeritus professor at the University of Otago in New Zealand.

In a lecture in Cambridge yesterday, he said that the study of intelligence has for too long been asking the wrong question: “The questions are not ‘Are we getting smarter?’ and ‘Are our children really smarter than we are?’ If the rise in IQ scores meant that we were smarter, that would mean our grandparents were dull and our great grandparents idiots, which is clearly not the case. The question should be, ‘Have certain cognitive skills risen?’ And the answer to that is yes.”

What accounts for our rise in intelligence test scores, Professor Flynn believes, is social and environmental changes that have given us the opportunity to exercise the kinds of skills that IQ tests measure. We increasingly fill leisure time with cognitively demanding pastimes, such as puzzles and computer games.

We have also developed a more scientific way of viewing the world. “In 1900 if you’d asked a child what do a dog and a rabbit have in common, they might have replied with a concrete answer like, ‘Dogs are used to hunt rabbits’. Today a child would be more likely to say, ‘They’re both mammals’. We classify things scientifically.”

Another factor in rising test scores concerned our ability to deal with complex abstract ideas. This is demonstrated in our ability to absorb abstract “shorthands” — for example the term “market” to signify laws of supply and demand.

Professor Flynn believes there is no reason to believe IQ gains will go on for ever. He points out that although gains are still robust in America, they have stopped in Scandinavia.


My published articles are archived at iSteve.com -- Steve Sailer

9 comments:

Anonymous said...

"He points out that although gains are still robust in America, they have stopped in Scandinavia."

This might be due to the greater socioeconomic inequality in the USA, depressing IQs in lower socioeconomic groups, which then means there's more 'slack' at the low end that can rise as societal wealth rises.
It might also indicate a declining education system in Sweden and/or a general decline in Swedish culture. Or it could be a statistical artefact. :)

-Simon

JD said...

Finally somebody acknowledging the biggest gotcha in IQ tests (in both the good and the bad sense). They measure your capability for "abstract symbol manipulation."

As an analogy, a would-be athlete would not become what he is unless he exercised rigorously for years to shape up and gain the specific motor coordination skill required in his field (say football.) That doesn't exactly mean a 4'7", 100 lbs. girl would be the equal of a 6'4", 220 lbs. guy -- even if the former exercised 6 hours a day and the latter was a couch potato. It just means given two guys each 6'4" and 220 lbs., the one that is a couch potato has no chance vis-a-vis the one who exercises 6 hours a day.

Half Sigma said...

Here's a link to SAT Data.

Scores over the last 40 years of fallen, not risen.

If people were really getting as smarter as the Flynn Effect supposedly says, SAT scores surely would have risen, even with a higher percentage of the population taking the test.

Mark Seecof said...

If that newspaper has represented Flynn's views accurately, and if Flynn meant to include our posterity in the notion that "our cognitive functions have reached — or nearly reached — the upper limits of what they will ever achieve," then Flynn is walking on air. We have no basis to think that our brains cannot ever be smarter--some of our children may be just one SNP away from a higher IQ.

Furthermore, the fact that we have millions of high-IQ people in the world proves that high-IQ people are possible. Just shifting the population up the ol' bell curve would improve national IQ scores.

So far as I can see, in existing higher-IQ countries, the main drags on national IQ are low-IQ immigration and the difference in fertility between lower- and higher-IQ portions of the population.

So even if we have run out of ways to boost individual IQ's by providing stimulating environments, we cannot say that puts any limit on population IQ's.

Anonymous said...

"So far as I can see, in existing higher-IQ countries, the main drags on national IQ are low-IQ immigration and the difference in fertility between lower- and higher-IQ portions of the population."

This may be the case, and I understand why Americans especially worry about immigration, but something I noticed from Lynn's "Race & IQ" is that white north Europeans' median IQ seems out of step with, and higher than, what you would expect from a purely genetic comparison. Comparing eg brain size, you would expect north-east-Asians to score about 10 points higher than Europeans, not 5 points, and Europeans' difference from sub-Saharan Africans (even in the West, ie no major malnutrition effects) is also greater than what you'd expect from a physical comparison.

This is easily overlooked because us white north Europeans tend to set ourselves as the baseline, and it's further obscured by arbitrarily excluding non-north-European Caucasians (eg Arabs), who have lower median IQs, from the 'European' group the way Lynn does. The way Lynn groups "races" makes it look like it's designed to make Europeans look good, with them getting the second-highest median IQ of the major races. If he included Turks, Arabs, Berbers, Persians, Pathans et al, the median Caucasian IQ wouldn't be much if any over the global median IQ of around 90.

It seems an important point - from Lynn's data, it looks to me that the 'natural' or genetic European IQ is thus around 90, not 100, and that Europeans thus have a roughly 10 point cultural IQ advantage vs the rest of the world. Put another way, the difference between white Americans' natural ability, & that of (eg) Hispanic immigrants, may be much less than the IQ data indicate.

This would then indicate that where there is assimiliation of culture, cultural factors may significantly raise the IQ of non-European immigrants into European-culture nations. But this requires cultural assimilation; where immigration does not involve integration into the European host culture, the cultural IQ boost is lessened or eliminated.

Anonymous said...

Above was me, Simon.

-Simon

Floccina said...

JD wrote:

“As an analogy, a would-be athlete would not become what he is unless he exercised rigorously for years to shape up and gain the specific motor coordination skill required in his field (say football.) That doesn't exactly mean a 4'7", 100 lbs. girl would be the equal of a 6'4", 220 lbs. guy -- even if the former exercised 6 hours a day and the latter was a couch potato. It just means given two guys each 6'4" and 220 lbs., the one that is a couch potato has no chance vis-a-vis the one who exercises 6 hours a day.”

This is only half way true studies have shown in non skill things like strength and endurance all previous years of work do not affect performance. If fact there was a case of twins where one had been training for and running triathlons for years and was world class it took her sister how had never trained before 6 months to catch up with her. Athletes are born not made.

David said...

Wasn't cousin marriage virtually manditory worldwide for most of history?

Anonymous said...

@HalfSigma,

IQ tests, and SATs, are constantly re-normed. That's the main discovery that Flynn made - every time IQ tests were re-normed, the new norming group did better on the old test than the old norming group did.