The Flynn effect is the rise of average Intelligence Quotient (IQ) test scores over the generations, an effect seen in most parts of the world, although at greatly varying rates. It is named after James R. Flynn, who did much to document it and promote awareness of its implications. This increase has been continuous and roughly linear from the earliest days of testing to the present. "Test scores are certainly going up all over the world, but whether intelligence itself has risen remains controversial," psychologist Ulric Neisser wrote in an article in 1997 in The American Scientist. The Flynn effect may have ended in some developed nations starting in the mid 1990s although other studies, such as Black Americans reduce the racial IQ gap: Evidence from standardization samples (Dickens, Flynn; 2006), still show gain between 1972 and 2002.
The average rate of rise seems to be around three IQ points per decade. Because children attend school longer now and have become much more familiar with the testing of school-related material, one might expect the greatest gains to occur on such school content-related tests as vocabulary, arithmetic or general information. Just the opposite is the case: abilities such as these have experienced relatively small gains and even occasional declines over the years. The largest Flynn effects appear instead on culture reduced highly general intelligence factor loaded (g-loaded) tests such as Raven's Progressive Matrices. For example, Dutch conscripts gained 21 points in only 30 years, or 7 points per decade, between 1952 and 1982.
Some studies focusing on the distribution of scores have found the Flynn effect to be primarily a phenomenon in the lower end of the distribution. Teasdale and Owen (1987), for example, found the effect primarily reduced the number of low-end scores, resulting in a pile up of moderately high scores, with no increase in very high scores. However, Raven (2000) found that, as Flynn suggested, data reported by many previous researchers that had previously been interpreted as showing a decrease in many abilities with increasing age must be re-interpreted as showing that there has been a dramatic increase in these abilities with date of birth. On many tests this occurs at all levels of ability. Two large samples of Spanish children were assessed with a 30-year gap. Comparison of the IQ distributions indicated that
Taken at face value, these changes are considered large by some. Ulric Neisser, who in 1995 headed an American Psychological Association task force writing a consensus statement on the state of intelligence research, estimates that if American children of 1932 could take an IQ test normed in 1997 their average IQ would have been only about 80. In other words, half of the children in 1932 would be classified as having borderline mental retardation or worse in 1997. Looking at Ravens, Neisser estimates that if you extrapolate beyond the data, which shows a 21 point gain between 1952 and 1982, an even larger gain of 35 IQ points can be argued, however Arthur Jensen warns that extrapolating beyond the data leads to results such as an IQ of -1000 for Aristotle (even assuming he would have scored 200 in his day)
Attempted explanations have included improved nutrition, a trend towards smaller families, better education, greater environmental complexity, and heterosis. Another proposition is greater familiarity with multiple-choice questions and experience with brain-teaser IQ problems.
Many studies find that children who do not attend school for one reason or another score lower on the tests than their regularly attending peers. In the 1960s, when some Virginia counties closed their public schools to avoid racial integration, compensatory private schooling was available only for white children. On average, the African-American children who received no formal education during that period fell back at a rate of about six IQ points per year. Length of average schooling has steadily increased. One problem with this explanation is that if comparing older and more recent subjects with similar educational levels, then the IQ gains appear almost undiminished in each such group considered individually. Mathematics has been proposed as particularly important.
Another explanation is an increased familiarity of the general populace with tests and testing. For example, children who take the very same IQ test a second time usually gain 5 or 6 points by doing so. However, this seems to set an upper limit on the effects of test sophistication. One problem with this explanation and other related to the schooling is, as noted above, that those subsets one would expect to be affected the most show the smallest increases.
Another theory is that many parents are now interested in their children's intellectual development and are probably doing more to encourage it than they did in the past. Early intervention programs have shown mixed results. Some preschool (age 3-4) intervention programs like "Head Start" do not produce lasting changes in IQ, although they do confer other important benefits. The "Abecedarian Project", an all-day program that provided various forms of environmental enrichment to children from infancy onward, showed IQ gains that did not diminish over time; the IQ difference between the groups was still present at age twelve. This very intensive intervention resulted in a gain of five IQ points, and not all such projects have been successful. Also, such IQ gains can diminish until the age of 18. Several other studies have also found lasting cognitive gains.
Still another theory is that the general environment is today much more complex and stimulating. One of the most striking 20th-century changes in the human intellectual environment has come from the increase in exposure to many types of visual media. From pictures on the wall to movies to television to video games to computers, each successive generation has been exposed to far richer optical displays than the one before and may have become more adept at visual analysis. This would explain why tests like the Raven's have shown the greatest increases, since they depend on such analysis. This explanation may imply that IQ tests do not necessarily measure a general intelligence factor, especially not Raven's as often argued, but instead may measure different forms of intelligence that are developed by different experiences. An increase only in particular form(s) of intelligence would explain why the Flynn effect has not caused a "cultural renaissance too great to be overlooked."
Related to this, James Flynn's current explanation (Flynn 2007) is that environmental changes arising from modernization - such as more intellectually demanding work, greater use of technology and smaller families - have meant that people are far more used to manipulating abstract concepts such as hypotheses and categories than a century ago. Substantial portions of IQ tests deal with these abilities. He gives as an example the question 'What do a dog and a rabbit have in common?' - a modern respondent might say they are both mammals (an abstract answer), whereas someone a century ago might say that you catch rabbits with dogs (a concrete answer).
Improved nutrition is another explanation. Today's average adult from an industrialized nation stands much taller than the comparable adult of a century ago. That increase in stature, almost certainly the result of general improvements in nutrition and health, has come at a rate of more than a centimeter per decade. Available data suggest that these gains have been accompanied by analogous increases in head size, and presumably by an increase in the average size of the brain. One problem with this, and other explanations arguing that the IQ gains reflect real gains in general intelligence, is that if all of the gains are real, then this would imply changes in average mental ability too great to seem plausible. A 2005 study presented data supporting the nutrition hypothesis, which predicts that gains in IQ will predominantly occur at the low end of the distribution where nutritional deprivation is most severe.
Possibly related to the Flynn effect is change in cranial vault size and shape during the last 150 years in the US. These changes must occur by early childhood because of the early development of the vault.
Flynn earlier argued that the very large increase indicates that IQ tests do not measure intelligence well but only a minor sort of "abstract problem-solving ability" with little practical significance.
Dickens and Flynn in 2001 presented a model for resolving several contradictory findings regarding IQ. They argue that the measure "heritability" includes both a direct effect of the genotype on IQ and also indirect effects where the genotype changes the environment, in turn affecting IQ. That is, those with a higher IQ tend to seek out stimulating environments that further increase IQ. These reciprocal effect result in gene environment correlation. The direct effect can initially have been very small but feedback loops can create large differences in IQ. In their model an environmental stimulus can have a very large effect on IQ, even in adults, but this effect also decays over time unless the stimulus continues (the model could be adapted to include possible factors, like nutrition in early childhood, that may cause permanent effects). The Flynn effect can be explained by a generally more stimulating environment for all people. The authors suggest that programs aiming to increase IQ would be most likely to produce long-term IQ gains if they taught children how to replicate outside the program the kinds of cognitively demanding experiences that produce IQ gains while they are in the program and motivate them to persist in that replication long after they have left the program. However if the Flynn Effect is caused by intellectual stimulation, this may suggest that the Flynn Effect is unrelated to g because according to Jensen "the preponderance of evidence argues that variance in the level of g is not a psychologically manipulable variable, but rather a biological phenomenon under the control both of the genes and of those external physical variables that affect the physiological and biochemical functioning of the central nervous system, which mediates the behavioral manifestations of g...Anything less than very early and intensive intervention, including medical and nutritional advances, during the preschool years (and also prenatally) is probably inadequate to cause a lasting increase in the child's level of g. However, Dickens and Flynn's paper, which was written after Jensen's book, disputes Jensen's claims, for example arguing that using Jensen's methodology the Flynn effect is found to be substantially due to genetic improvements, an extremely unlikely cause. Their theory explains such contradictions.
Some researchers, such as J. Phillipe Rushton argue the Flynn effect largely has not changed the general intelligence factor (g), which would mean practical significance of the effect would be limited. More recent studies have found that g has improved substantially
Studies that make use of multigroup confirmatory factor analysis test for "measurement invariance." Where tenable, invariance demonstrates that group differences exist in the latent constructs the tests contain and not, for example, as a result of measurement artifacts or cultural bias. Wicherts et al. (2004) found evidence from five data sets that IQ scores are not measurement invariant over time, and thus "the gains cannot be explained solely by increases at the level of the latent variables (common factors), which IQ tests purport to measure". In other words, according to this study, some of the inter-generational difference in IQ is attributable to bias or other artifacts, and not real gains in general intelligence or higher-order ability factors.
A 2003 study looking at the Flynn effect in Kenya between 1984 and 1998 found that the increase was best explained by parents' literacy, family structure, and children's nutrition and health.
A 2006 study from Brazil looked at data from testing children in 1930 and 2002-2004, the largest gap ever considered. The results are consistent with both the cognitive stimulation and the nutritional hypotheses.
In the end a number of varied phenomena may be contributing to the Flynn effect.
The Flynn effect may have ended in some developed nations starting in the mid 1990s. Teasdale & Owen (2005) "report intelligence test results from over 500,000 young Danish men, tested between 1959 and 2004, showing that performance peaked in the late 1990s, and has since declined moderately to pre-1991 levels." They speculate that "a contributing factor in this recent fall could be a simultaneous decline in proportions of students entering 3-year advanced-level school programs for 16–18 year olds."
In 2004, Jon Martin Sundet of the University of Oslo and colleagues published an article documenting scores on intelligence tests given to Norwegian conscripts between the 1950s and 2002, showing that the increase in scores of general intelligence stopped after the mid-1990s and in numerical reasoning subtests, declined.
Some have claimed that the Flynn effect was masking a dysgenic decline in human reproduction and that in developed countries the only direction that IQ scores will now move is downwards. However, even if there is a decline, then this may have other causes than dysgenics. Genetic changes usually happen relatively slowly. For example, the Flynn effect has been too rapid for a genetic explanation. Researchers have warned that constantly greater exposure to industrial chemicals shown to damage the nervous system, especially in children, in industrialized nations may be responsible for a "silent pandemic" of brain development disorders.
Also, if the Flynn effect has ended for the majority, it may still continue for minorities, especially for groups like immigrants where many may have received poor nutrition during early childhood. For example, William T. Dickens and James R. Flynn write in their 2006 paper Black Americans reduce the racial IQ gap: Evidence from standardization samples that blacks have gained 5 or 6 IQ points on non-Hispanic whites between 1972 and 2002. Gains have been fairly uniform across the entire range of black cognitive ability. J. Philippe Rushton and Arthur R. Jensen have disputed their findings, calculating a mean gain for Blacks of only 3.44 IQ points, and questioned the exclusion of four independent tests that showed low or negative IQ gains.