Now, a "Flynn Effect" for Smart People

There are numerous proposed explanations of the Flynn effect, as well as some skepticism about its implications. Similar improvements have been reported for other cognitions such as semantic and episodic memory.[3] Recent research suggests that the Flynn effect may have ended in at least a few developed nations, possibly allowing national differences in IQ scores[4] to diminish if the Flynn effect continues in nations with lower average national IQs _FlynnEffect_Wikipedia

A brand new study, “The Flynn Effect Puzzle,” currently in press at Intelligence, and led by Jonathan Wai at Duke University, has found an interesting way to assess the right tail of the distribution. By looking at approximately 1.7 million scores of 7th grade students between 1981 and 2000 on the SAT and ACT, as well as scores of 5th and 6th grade students on the EXPLORE test, the psychologists were able to investigate the extent to which the Flynn effect exists in the right tail of the bell curve. The results were clear:

The effect was found in the top 5% at a rate similar to the general distribution, providing evidence for the first time that the entire curve is likely increasing at a constant rate. The effect was also found for females as well as males, appears to still be continuing, is primarily concentrated on the mathematics subtests of the SAT, ACT, and EXPLORE, and operates similarly for both 5th and 6th as well as 7th graders in the right tail.

_Wired

Jonah Lehrer, author of the Wired post linked above, tries to explain why the Flynn Effect might work at the smart end of the curve, but comes up rather empty.

Forbes blogger Alex Knapp comments on Lehrer's inability to explain these findings:

As far as I’ve been able to tell, standardized test use in schools has increased since the 1960s and that use continues to grow apace. I’d suggest that as standardized tests are used more in schools, students get better at grappling with the critical thinking aspect of those tests. Therefore, they perform better, thus increasing the scores over time as more and more students take more and more tests. _Forbes

Knapp's guess is probably closer to the mark than those given by Lehrer in his piece. But no one seems to want to discuss assortative mating, where smart and high achieving males and females more easily find, marry, and mate with each other in medical schools, law schools, engineering schools, other professional and scientific programs, and in the workplace generally -- as more high IQ women enter higher education and the professions.

Explanations for the Flynn Effect generally range from better nutrition, increased literacy, higher use of abstract thinking as populations increasingly move into urban environments, and even the greater use of Caesarean section in childbirth!

One unusual explanation for the Flynn Effect is a higher incidence of Asperger's variant of autistic spectrum disorder. Another interesting explanation given is the relatively accelerated physical development of youngsters today as opposed to youngsters half a century or more ago. Earlier onset of puberty in modern times is a partial support for this argument. If this is true, comparison of today's 12 year olds' scores to the scores of 12 year olds several decades ago, would not be valid. (more here)

The amplitude of the Flynn effect is quite uneven between nations, and even seems to have slowed down or vanished in some nations. While James Flynn himself has expressed high hopes that this effect will lead to the narrowing of racial IQ gaps between whites and blacks, that result appears quite unlikely, given the trend so far.

Modern societies certainly need a healthy proportion of high IQ citizens -- members of the smart fraction -- so as to maintain and advance their high technology infrastructures. But far more public moneys are being spent in remediation and palliation of the much higher numbers of low IQ persons, than are spent to improve the ability of the smart fraction to advance the intellectual and scientific infrastructure of society. Such a trend is typical for a society that wishes to slow down its rate of deterioration, rather than a society that wishes to speed its rate of advancement.

So what good is the Flynn Effect if society doesn't know what to do with it? More on that question later.

Abstract for earlier Jonathan Wai paper looking at sex differences on the smart end of the IQ curve

Literacy Rates, the Flynn Effect, Brain Wiring, Tautologies

In a recent look at the Flynn Effect, we at Al Fin neglected to mention one of the possible important contributors to at least part of the changes in IQ test scores observed over the 20th century: changes in literacy rates.

World Literacy 1900s
In a trivial sense, if a test-taker is unable to read and understand the directions for taking the test, his test score would not be valid. IQ Corner blog presents an interesting discussion about the possible relationship between the Flynn Effect and literacy rates.

The importance of being able to read for performance on an IQ test cannot be understated. Instead of measuring ‘intelligence' in a nonliterate test-taker, the test is measuring that person's inability to read....If increasing literacy were really explaining a number of seemingly different IQ trends, then you would expect to see a few things. First, within a population you should expect increased education of literacy skills to be associated with an increase in the average IQ of that population. Second, IQ gains should be most pronounced in the lower half of the IQ bell curve since this is the section of the population that prior to the education would have obtained relatively lower scores due to their inability to comprehend the intelligence test's instructions. With increased literacy, you should expect to see a change in the skewness of the IQ distribution from positive to negative as a result of higher rates of literacy in the lower half of the IQ distribution (but very little change in the top half of the distribution). You should also expect to see differences on the particular intelligence test subscales, with increased literacy showing the strongest effects on verbal tests of intelligence and minimal differences on other tests of intelligence.

...To test these predictions, Marks looked at samples representative of whole populations (rather than individuals), and used ecological methods to calculate statistical associations between IQ and literacy rates across different countries. Were Marks' findings consistent with the predictions?

Strikingly, yes. He found that the higher the literacy rate of a population, the higher that population's mean IQ, and the higher that population's mean IQ, the higher the literacy rate of that population. When literacy rates declined, mean IQ also declined. Marks also found evidence for unequal improvements across the entire IQ spectrum: the greatest effects of increased literacy rates were on those in the lower half of the IQ distribution. Interestingly, he also found that both the Flynn Effect and racial/national IQ differences showed the largest effects of literacy on verbal tests of intelligence, with the perceptual tests of intelligence showing no consistent pattern.

It must be noted that literacy wasn't the only factor responsible for the Flynn effect. Adopting the Cattell-Horn-Carroll (C-H-C) framework (McGrew, 2005, 2009) Marks found that Visual processing (Gv) and Processing Speed (Gs) also made important contributions. _IQCorner

Recent research suggests that learning to read actually changes the physical structure of the brain, and consequently brain activity and processing. If that is the case, then the effect of literacy on IQ scores may extend far beyond the ability to understand the test instructions.

Reading and writing are cognitive tools that, once acquired, change the way in which the brain memorizes facts and conceptualizes ideas, and these changes stimulate abstract thinking. When psychologists use neuroimaging technology, like MRI, to compare the brains of literates and illiterates working on a task, they find many differences in how their brains work whether or not they are reading. Researcher Alexandre Castro-Caldas discovered that processing between the hemispheres of the brain was different between those who could read and those who could not. A key part of the corpus callosum was thicker in literates, and "the occipital lobe processed information more slowly in individuals who learned to read as adults compared to those who learned at the usual age." Psychologists Ostrosky-Solis, Garcia and Perez tested literates and illiterates with a battery of cognitive tests while measuring their brain waves and concluded that "the acquisition of reading and writing skills has changed the brain organization of cognitive activity in general… not only in language but also in visual perception, logical reasoning, remembering strategies, and formal operational thinking." Literacy — a human invention — rewires the human mind. _The Technium

Kelly goes on to point out that the study of music can also alter brain structure and brain function. Then he suggests that the internet and related communication and information technologies may be causing similar changes in the way the brain is wired, thus changing how (and how well) people think.

Children must be exposed to language by a certain age, or they will lose the ability to become skilled in its use -- presumably forfeiting any advantages in brain processing that language confers. The same is true for the timely learning of music, and the skillful development of musical talents. In fact, developmental windows exist for many cognitive skills -- both known and likely unknown (for now) skills.

This raises intriguing questions about what we could be doing -- but aren't -- to facilitate improved thinking skills for our children, and perhaps ourselves. Very few persons, if any, are consistently functioning at their peak cognitive levels. What training methods are we missing which we should be aware of?

As suggested by Castro-Caldas et al. (1998), learning to read and write adds a visuographic dimension, based on operation of matching phonemes and graphemes, to the internal representational system for spoken language, so learning a specific skill during childhood may partially determine the functional organization of the adult brain. _Can Literacy Change Brain Anatomy? Ostrosky-Solis 2004 IJP 39 (1) 27-35

Of course, since reading comprehension is part of what comprehensive IQ tests measure, pointing out that increased literacy rates leads to higher IQ scores is a somewhat tautological -- or circular -- argument. Nevertheless, the human brain is inherently circular in function, and the evidence for a significant change in brain structure due to the acquisition of reading and writing should not be dismissed.

As has been pointed out, the general spread of affluence (in the sense of increased food, education, information technologies, and leisure time) over the 20th century -- at least in the countries and over the time spans registering a significant Flynn Effect -- would lead to a cluster of changes in the pre-natal and post-natal environment of children. Better nutrition, a richer stimulatory environment, increased literacy, better prenatal and childhood medical care, increased exposure to music and music training, etc., could all lead to changes in brain structure and brain function in these upwardly evolving societies.

And yet, it is relatively easy to "cut to the chase" by looking at members of societies who are already at the top of the heap, the affluent socioeconomic classes. By looking at "pre-evolved," already affluent members of a given population group, one can anticipate what the maximum effects of any Flynn Effect are likely to be on a given population.

Rather than to tell you what the results of such studies have been, it might be better to allow you to cogitate on the matter first.

Over time, we are able to incorporate more factors of potential importance into our concepts of the foundations of human cognition. We are stymied by politically correct limitations on what can be studied and tested, but despite the ongoing corruption of science by political correctness, the underlying structures will be worked out piece by piece.

"Flynn Effect" Contradicts Heritability of Intelligence?

One of the quickest paths to career suicide for an academic or any public figure, is to assert that human intelligence is genetically bounded, and roughly 50 - 80% heritable. The preponderance of scientific evidence points toward a genetic limiting of the "g factor", commonly associated with intelligence, with substantiating research coming from China. And yet the existence of the "Flynn Effect" -- the apparent rise in IQ test scores across several national populations over the past century -- has led many people to suggest that the IQ measure is too changeable to be a true indication of genetically "determined" intelligence.

Average IQ scores in many nations have apparently risen by between 5 points and 25 points per generation, over much of the past century. The increase has come largely in the measure of "fluid intelligence," which is one measure of problem-solving capacity.

Explanations for this general, but uneven, trend include better nutrition, better education and familiarity with testing, a more stimulating childhood environment, better control of childhood infectious disease, and an increase in hybrid vigour from less inbreeding of populations. Changes in variance for the different sample distributions can also introduce problems in interpreting such a trend.

The issues of general human intelligence, and the genetics of intelligence, are complex and multi-factorial. Many genetic and epigenetic factors are involved in the "biological bounding" of human intelligence, as well as other biological and environmental factors which bear on gene expression.

It is never a question of "either genetics or environment," but is always both. Thus it is not a question of genetic "determination" of intelligence, but rather a genetic bounding of intelligence. It can be easier to understand that measures such as height potential or "tallness" are "genetically bound" (in the absence of hGH and other exogenously administered growth factors). Similarly, intelligence potential is genetically bound in the absence of NZT or other hypothetical efficacious brain boosters.

So are people really getting smarter? The Flynn effect does not seem to be particularly relevant to the high IQ end of the distribution. So the smarter people do not seem to be getting smarter. But part of the lower end of the distribution seems to have moved toward the middle, and parts of the middle may have moved toward the higher end, on IQ test scores.

"Flynn Effect" improvements in scores seem to have slowed consiberably in advanced nations over the past 20 years. Whatever the cause of the trend, it may have largely played itself out in those nations. But if better nutrition, higher levels of stimulation, and lower infectious disease load are at least partially responsible for the trend of rising IQ scores, we should expect improvements in scores across the more impoverished areas of the third world which are able to move out of poverty.

How much of an improvement? One possible answer would be to look at the children of affluent persons in third world nations, as a "preview of coming scores." Rich people's children are generally provided with the best medical care, education, and generally more intellectually stimulating environments. Look at the top end of performance for the most advantaged children in these countries. What do you find?

You will find the "gifted" and "advantaged" in countries across Subsaharan Africa topping out at 1 standard deviation below average populations of the advanced world and about 2 standard deviations below gifted children in the advanced world. In most of tribal Asia, gifted and advantaged groups top out at roughly 1 standard deviation below their counterparts in advanced countries -- higher than Africans but significantly lower on average than East Asians and Europeans.

Similarly, you can look at the segment of African American children in the US which is most advantaged and gifted, and find persistent IQ score gaps between this group and far less advantaged European and East Asian children -- who nonetheless score higher on IQ and aptitude tests than their more advantaged black cohorts.

It is wrong to assume that the Flynn Effect will raise the overall average human IQ score of Earth, and it is wrong to assume that it will equalise IQ scores between population groups which have maintained stable and long-lasting differences in IQ scores over time.

Whenever nutrition, education, medical care, and intellectual stimulation are roughly equal, the genetic component in bounding intelligence comes to the forefront. Of course there is always room to improve on all those counts -- and on other determinants of intelligence which have not yet been discovered or delineated. But underneath it all, is the genetic component which will never be equalised, but which may well be compensated for with clever science and technology, in the future.

Al Fin cognitive scientists and psychometricians feel that the nutritional and infectious disease components were of limited relevance to the Flynn Effect in the advanced world since the 1940s -- except among truly impoverished populations, which are rare in the western developed world. Other explanations for the increase in scores which are more likely include increased early childhood stimulation and a greater familiarity with test taking.

The "critical window of development" theory is likely to be quite pertinent to the Flynn Effect in western populations over the past century. If a child is exposed to language, music, math, executive function training, and certain types of abstract thinking within the critical developmental timeframe, it is more likely that he will emerge from that period with better developed skills.

IQ gains in childhood that come from a highly stimulatory intellectual environment may not be permanent, however. And not all children will tolerate such external manipulations well. But the more cognitive and behavioural competencies the child is given at an early age (within reason), the more likely he will be to retain at least some of them. Each child is an individual and should be treated as such.

In conclusion, the Flynn Effect is a variable and limited trend in the rise of IQ scores across many nations. It should be yet another reminder of the remarkable complexity of human intelligence and behaviour, all of which derive ultimately from the human genome.

Megafoundation links and views on intelligence

Black-White Test Score Gap

"The Black White Test Score Gap: An Introduction" by Jencks and Phillips is a book length look at the problem.

African Americans currently score lower than European Americans on vocabulary, reading, and mathematics tests as well as on tests that claim to measure scholastic aptitude and intelligence...the typical American black still scores below 75% of American whites on most standardized tests. __TBWTSG Chapter1

Nothing perpetuates the racial divide within societies so much as the "achievement gap." It begins early, and persists--even widens--throughout the lifetime. Despite society's best efforts to narrow the racial test score gap, it persists.

It is well documented that scores on cognitive tests taken by adolescents are predictive of future labor market outcomes, such as educational attainment and earnings.1 Even test scores at age seven have been shown to be correlated with measures of labor market success.2 These findings have led many researchers to assign a large role to "premarket factors" in explaining adult earnings inequality, where premarket factors are broadly interpreted to represent endowed ability, the effects of family background and the influence of schools. Premarket factors are also thought to be an important part of the explanation for racial disparities in test score performance.3 __(PDF) Source

Is it possible that tests are biased against some races, in favour of others?

The FDNY test was not the only test to show a gap. So did the SAT, GRE, GMAT, MCAT, LSAT, bar exam, police exams, IQ test - every test of aptitude. All of them are supposed to be racist....All of them show about the same gap - what psychometrists call one standard deviation. And all of them show that Asians are about a third of a standard deviation above whites.

Consider how difficult it must be to maintain this gap year in and year out if it were not real, but the result of a conspiracy...Every year, the test makers must work to make that the gap remains the same - and not just one one test, but for everyone from children taking the SAT and AP exams to doctors taking medical boards.

At the same time, East Asians must score better than whites - again, always by the same margin. The White Racist Conspiracy allows Asians to score higher because… why is it again? If the tests are developed by whites, why are whites failing to score higher than East Asians?...To make sure that hundreds, maybe thousands, of different tests conform perfectly to each other in terms of racial/ethnic/gender averages, and achieve those unfair differences through racist cheating for dozens of years, we would expect that there would be a conspiracy of hundreds of thousands of people over the generations since these tests were implemented.

Would it be racist to suggest that East Asians have higher IQ scores than whites? Would it be racist to say that Asians are better educated than whites? Would it be racist to say that Asians have significantly lower crime rates than whites? That they save more and waste less of their income than whites?...So why then is it racist to say all those things about whites as compared to blacks when all the tests and statistics show that these statements are factually accurate?

White people don’t break out in tears when they hear that they have lower IQ than Asians, on average of course. They do not feel that now they can’t go to school or that society is against them. They do not suffer low self-esteem as a result of knowing this fact. __Source

Researchers Charles Murray and James Flynn debated the question of whether the black-white gap is likely to every close. Here is the link for viewing the debate. H/T Reason

The test score gap has been persistent over the time period that aptitude testing has been commonplace. In Nazi Germany, IQ tests were outlawed when it was discovered that Jews were routinely outscoring Germans on measures of intelligence. Something similar has occurred in many institutions in North America.

But sweeping inconvenient data under the rug does not mean that the underlying phenomena no longer exists. That is the fallacy of Wikipedia censorship, and the censorship practised by Google in China. The phenomena are still there. All they accomplish by covering it up is to keep the masses of people clueless about the reality they are dealing with.

If there is ever to be a solution to problems of poverty, crime, civil disorder, lack of achievement, etc. that emanate from low IQ, humans must face problems squarely and study them intently. The long term aim is to raise the IQ of everyone, regardless of race and ethnicity. Many of the most severe problems facing humans today come from low achievement and poor problem solving. Low IQ and decadent culture both contribute to low achievement and poor problem-solving. Both of these lower level problems are likewise based upon problems even lower.

It is time to honestly get to the base of these problems so that we can open the way to a more widely participatory future, where human problems are solved rather than used as political hockey pucks.

Image Source

Addendum 20 April 08: The above posting takes no stand on the nature vs. nurture debate, regarding underlying causes for achievement and test score gaps between populations. Here is a brief, readable summary of both sides of the issue--nature and nurture.

For those drawn to more international comparisons, this Wiki article contains a table that gives comparisons of estimated average population IQ scores for 81 different countries.

IQ, Executive Function, and Emotional Intelligence

There are many components of a person's life that will contribute to his success--or lack of success. Three of the most important ingredients for success are IQ, executive function, and emotional intelligence. IQ is moderately correlated with executive function, and executive function can be very highly correlated with emotional intelligence--depending upon how the two ideas are defined.

You would expect that children with above average IQ, executive function, and emotional intelligence, would be more successful in school, socially, and career wise.

In this study 141 healthy children between the ages of three and five years took a battery of psychological tests that measured their IQs and executive functioning. Researchers found that a child whose IQ and executive functioning were both above average was three times more likely to succeed in math than a kid who simply had a high IQ.

"[The fact] that executive function, even in children this young, is significantly related to early math performance suggests that if we can improve executive function, we can improve their academic performance," says Adele Diamond, professor of developmental cognitive neuroscience at the University of British Columbia.

SciAm

Improving IQ, executive function, and emotional/social intelligence would seem to be worthy pursuits of the education/child care professions. But how amenable are these "metrics" to improvement through training and education?

James Flynn is one optimistic researcher of human intelligence, who believes that "IQ gaps" can be narrowed (or eliminated), and that success in life should not be held hostage to an IQ test score.

Once you break intelligence down into its autonomous components, many things become clear. For example, the Nation’s Report Card shows that today’s children are ahead of their parents in reading at early ages and then the gains fade away by the age of 17. How is that possible? The children are doing much better on heavily g loaded IQ tests like the WISC at all ages. Should not brighter people be able to read adult novels better?....

Cato Unbound

Here is a short list of executive functions:

1. * planning for the future and strategic thinking
2. * the ability to inhibit or delay responding
3. * initiating behavior, and
4. * shifting between activities flexibly

Source

Here are some components of emotional intelligence:

1. Delay Gratification: Take right action even though there may not be an immediate reward.
2. Prioritize: Bring to the forefront of your mind to a “checklist” of what is more important so that you can weigh decisions and actions.
3. Manage Feeling: Use the simple feelings to act.. not react.
4. Optimism: Recognize that you have choices, that you can make a difference, that you are an important part of a living whole.
5. Accountability: Hold yourself to high standards and do what is right, even when it seems hopeless

Emotional Intelligence

What are the chances of being able to increase IQ, EF, or EI? Here, many people stumble upon the concept of "heritability." When researchers state that IQ is up to 80% heritable, executive function is up to 99% heritable, and emotional intelligence is also significantly inherited, many people despair of ever escaping the "gene trap."

People should indeed take inheritance and genetics seriously when looking at the effects of inherited factors on societal success and prosperity. But not at the expense of optimistic creativity in designing training and educational systems to maximise the realisation of "potential talents" that everyone possesses in varying amounts. Children cannot raise themselves, they cannot carry themselves to mastery on their own. But it is clear that our current system of government education and university indoctrination is not properly developing and refining the innate talents of youth.

Educational alternatives are important--from the earliest ages up through university and adult education. Fortunately, more people in positions of influence--although outside the government/education/industrial complex--are becoming aware of the critical problems and opportunities in this area. While the educational system of the US appears mired in a quagmire of political inbreeding and theoretical "dead-endism", fruitful ideas are being generated from the outside. Outside the sanctioned edu-establishment, and outside the US itself.

It is important not to get stuck on "g", or IQ, as the be all and end all of life success determinants. But it is also important to understand that even with EF and EI, genetics and heredity play a significant role. We absolutely must face the reality of the genetics of human behaviour and potential, but we must also understand that we are nowhere close to being able to optimise the potential of all students. The goal is optimal performance, not the impossibility of equal performance.

IQ Debate Between Charles Murray and James Flynn

Charles Murray and James Flynn debated on the topic of race and IQ at the American Enterprise Institute. You can view the full video of the debate here.

The debate is quite interesting in contrasting the different approaches to understanding the reasons for the persistent 1 standard deviation gap between mean IQ scores in the US between resident blacks and whites. I recommend listening to both debaters with an open mind, rather than only listening with the aim of gathering ammunition to defeat the point of view with which you may disagree.

It is in the interest of all North Americans that the well documented IQ gap between the populations classified as "white" and "black" be amenable to closing--by lifting the black IQ average to the white IQ average. The question before the debaters is whether such a closing of the gap is possible.

The topic is complex, with a long history that documents less than laudable actions and beliefs. Yet it is of vital importance that the issue be faced, rather than sidestepped, or glibly explained away. The disturbing tendency of ideological groups to attempt to influence what can be studied within science should be discontinued.

Also, see this recent Steve Sailer commentary on a recent NYT article on the Flynn effect.


Readers of Al Fin will understand that the quest for the Next Level involves raising the average intelligence of all persons to unprecedented levels. In the search for ways to do this, it is likely that there will be genetic "borrowing" from persons of all racial classifications. East Asian populations test higher on average on IQ tests than European populations, who test higher than Arab or SubSaharan African populations. There is no doubt that genetics is involved in these differences, even though environmental factors are likely also at work.

So to improve the intelligence of human populations across the board, genetic therapies will need to find more global optima, where currently it is local optima that determine population IQ differences.

I will not comment on either Flynn or Murray in this post, in the hopes that most readers will listen to both authors and thinkers with an open mind., Please do not comment on either debater unless you have listened to his full presentation above.