The Flynn effect: Are we getting smarter?

The Flynn effect refers to generational IQ test norm changes first systematically described in the 1980s (Flynn, 1987). IQ changes have been observed to be positive (on average 3 IQ points per decade), but differentiated according to the investigated country and the intelligence test domain. Paradoxically, larger gains for fluid than for crystallized intelligence have been observed. In a recent meta-analysis by Pietsching and Voracek (2015), the progress and strength of the Flynn effect since the introduction of psychometric intelligence testing in the early 20^th  century till 2013 was analyzed along with the moderating influences of age, economic growth, sample health status, and sex.

The key findings of the study can be summarized as follows:

• Strong evidence for continuous global generational IQ test score gains in the general population over the past century (2 SD or 0.28 IQ points annually) were observed.
• Gains in fluid IQ were substantially stronger than those in crystallized IQ (4.1 vs. 2.1 IQ points per decade).
• The IQ change trajectories showed robust evidence for decreasing gains in recent decades.
• Stronger gains were observed for adults than for children, showing large effects for fluid and spatial IQ.
• Gross domestic product growth was positively associated with full-scale, crystallized, and spatial IQ but it showed negligible effects for fluid IQ.
• IQ gains were not observed for psychometric g. 

The authors also provided an overview of possible explanations and theories for the observed gains, dividing them into environmental, biological, and hybrid (i.e., interacting biological and environmental). The environmental factors include education, technology, decreasing family size (dysgenic fertility), and test-taking behavior. It is assumed that the availability of education and technology for individuals from different socioeconomic backgrounds had a beneficial influence on the level of intelligence. This also had some influence on becoming more familiar with tests, mainly multiple choice, and by that changing test-taking behavior – principally increasing guessing.

Among biological factors, hybrid vigor is often mentioned, which refers to the mating of individuals from genetically dissimilar subpopulations, thereby increasing allelic heterozygosity and reducing homozygosity.

Hybrid factors include decreased blood lead levels, genomic imprinting (epigenetic inheritance), nutrition, and reduced pathogen stress. Moreover, more complex factors such as reduced IQ variability, effects of social multipliers, and decreasing life history speed (fewer offsprings), have been proposed in the literature.

The end of the Flynn effect?

In more recent decades a stagnation or even a decrease of the Flynn effect, mainly in more developed Western countries, has been reported. It is possible that the beneficial effects have caused the IQ increase to reach a ceiling. Some explanations have also linked these decreases to a higher proportion of immigrants from less developed countries. The table below summarizes the findings.

Study	Country	Findings
Sundet et al. (2004)	Norway	Substantial gains in intelligence were observed from the mid-1950s (test years) to the end 1960s–early 1970s, followed by a decreasing gain rate and a complete stop from the mid-1990s. The gains seemed to be mainly caused by decreasing prevalence of low scorers.
Teasdale & Owen (2005; 2008)	Denmark	Intelligence test results from over 500,000 young Danish men, tested between 1959 and 2004, showed that performance peaked in the late 1990s, and has since declined moderately to pre-1991 levels. 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.
Woodley & Meisenberg (2013)	Netherlands	Sixty-three observations of secular IQ changes were collected from three demographically diverse studies of the Dutch population for the period 1975–2005 (representing the 1950–1990 birth cohorts). Declines due to the anti-Flynn effect were estimated at -4.52 points per decade, whereas gains due to the Flynn effect were estimated at 2.18 points per decade. The N-weighted net of these is a loss of  -1.35 points per decade, suggesting an overall tendency towards decreasing IQ in the Netherlands with respect to these cohorts.
Dutton & Lynn (2013)	Finland	The average IQs of approximately 25,000 18–20 year old male military conscripts in Finland per year were analyzed for the years 1988 to 2009. The results showed increases in the scores on tests of shapes, number and words over the years 1988 to 1997 averaging 4.0 IQ points a decade. From 1997 to 2009 there were declines in all three tests averaging 2.0 IQ points a decade.
Dutton & Lynn (2013)	France	The results of the French WAIS III (1999) and the French WAIS IV (2008–9) were compared based on a sample of 79 subjects aged between 30 years and 63 years who took both tests in 2008–2009. It is shown that between 1999 and 2008–9 the French full scale IQ declined by 3.8 points.
Pietsching & Gitter (2015)	Germany and Austria	A meta-analysis (k=96; N=13,172) showed an inverse u-shaped trajectory of IQ test performance changes (initial increases and  subsequent decrease of performance) over 38 years (1977–2014).

The paradox of the Flynn effect

The cognitive functions that predict psychometric intelligence are thought to be working memory and processing speed. The idea behind the relation between processing speed and g is rather straightforward. Processing speed is viewed as a form of cognitive or neural limitation of processing a simple stimulus.  The most often studied parameters are reaction time and reaction time variability. Similarly, working memory and intelligence are highly related constructs. In a recent confirmatory analysis it was shown that working memory shared 83.4% of variance with fluid intelligence (Chuderski, 2015). Hence, with respect to the Flynn effect, one would expect that reaction time would over the years decrease, while working memory capacity would show an increase. However, the findings do not follow these trajectories.

Silverman (2010) analyzed simple visual reaction time (RT) obtained in a study conducted by Francis Galton in the late 1800s with the RTs obtained in subsequent studies. In Galton’s study, the median RTs obtained by men and women between ages 18 and 30 were 183.0 and 187.0 ms, whereas the RTs obtained in more recent studies (3,836 men and 3,093 women) were for men 250.43 ms (SD = 46.53) and 277.71 ms (SD = 30.76) for women. Moreover, the RTs obtained in the comparison studies were all longer than the RTs obtained by Galton. Out of several possible causes for longer RTs, Silverman suggested that the most tenable were that RT has been increased by the buildup of neurotoxins in the environment and by the increasing numbers of people in less than robust health who have survived into adulthood.

Based on Silverman’s findings, Woodley et al. (2013) in their meta-analytic study of RT concluded that the Victorians were cleverer than modern populations. Woodley and colleagues used the data on the slowing of simple reaction time described in a meta-analysis of 14 age-matched studies from Western countries conducted between 1889 and 2004. Using psychometric meta-analysis, they computed the true correlation between simple reaction time and g, yielding a decline of −1.16 IQ points per decade or −13.35 IQ points since Victorian times. Further, the difference between the meta-regression trend-weighted present (2004) simple RT mean (270.18 ms) and the trend weighted 1889 mean (193.17 ms) was 77 ms. Woodley and colleagues  further concluded that the most likely reason for this finding, also put forward by Silverman, was that those with poorer health and slower RTs surviving into adulthood are more numerous in the modern era than in the past.

Miller (1956) suggested that the typical short-term memory capacity (STMC) of an adult is 7 plus or minus two objects. Cowan (2005) suggested that the typical working memory capacity (WMC) of an adult is 4, plus or minus one object. Were these numbers lower in the past? This would be expected based on the Flynn effect. To answer this question, Gignac (2015) analyzed digit span (forward and backward) across 85 years of data (Ns of 7,077 and 6,841). The mean adult verbal STMC was estimated at 6.56 (±2.39), and the mean adult verbal WMC was estimated at 4.88 (±2.58). No increasing trend in the STMC or WMC test scores was observed from 1923 to 2008. This finding may be considered surprising, since memory span is so intimately related to fluid intelligence. At the moment they still lack a plausible explanation.

We can conclude that (Gignac, 2015, p. 93) “it may be prudent to acknowledge that the magnitude, pervasiveness, and true nature of the Flynn effect remains a substantially open question”.

References

Chuderski, A. (2015). The broad factor of working memory is virtually isomorphic to fluid intelligence tested under time pressure. Personality and Individual Differences, 85, 98–104. http://doi.org/10.1016/j.paid.2015.04.046

Cowan, N. (2005). The magical number 4 in short-term memory: A reconsideration of mental storage capacity. Behavioural and Brain Sciences, 24, 87–185.

Dutton, E., & Lynn, R. (2013). A negative Flynn effect in Finland, 1997–2009. Intelligence, 41(6), 817–820. https://doi.org/10.1016/j.intell.2013.05.008

Dutton, E., & Lynn, R. (2015). A negative Flynn Effect in France, 1999 to 2008–9. Intelligence, 51, 67–70. https://doi.org/10.1016/j.intell.2015.05.005

Flynn, J.R. (1987). Massive IQ gains in 14 nations: What IQ tests really measure. Psychological Bulletin, 101, 171–191.

Gignac, G. E. (2015). The magical numbers 7 and 4 are resistant to the Flynn effect: No evidence for increases in forward or backward recall across 85 years of data. Intelligence, 48, 85–95. https://doi.org/10.1016/j.intell.2014.11.001

Miller, G.A. (1956). Themagical number seven, plus or minus two: some limits on our capacity for processing information. Psychological Review, 63(2), 81–97.

Pietschnig, J., & Gittler, G. (2015). A reversal of the Flynn effect for spatial perception in German-speaking countries: Evidence from a cross-temporal IRT-based meta-analysis (1977–2014). Intelligence, 53, 145–153. https://doi.org/10.1016/j.intell.2015.10.004

Pietschnig, J., & Voracek, M. (2015). One Century of Global IQ Gains: A Formal Meta-Analysis of the Flynn Effect (1909–2013). Perspectives on Psychological Science, 10(3), 282–306. https://doi.org/10.1177/1745691615577701

Silverman, I.W. (2010). Simple reaction time: It is not what it used to be. The American Journal of Psychology, 123(1), 39. https://doi.org/10.5406/amerjpsyc.123.1.0039

Sundet, J. M., Barlaug, D. G., & Torjussen, T. M. (2004). The end of the Flynn effect? A study of secular trends in mean intelligence test scores of Norwegian conscripts during half a century. Intelligence, 32, 349–362.

Teasdale, T. W., & Owen, D. R. (2005). A long-term rise and recent decline in intelligence test performance: The Flynn Effect in reverse. Personality and Individual Differences, 39(4), 837–843. https://doi.org/10.1016/j.paid.2005.01.029

Teasdale, T. W., & Owen, D. R. (2008). Secular declines in cognitive test scores: A reversal of the Flynn Effect. Intelligence, 36(2), 121–126. https://doi.org/10.1016/j.intell.2007.01.007

Woodley, M. A., & Meisenberg, G. (2013). In the Netherlands the anti-Flynn effect is a Jensen effect. Personality and Individual Differences, 54(8), 871–876. https://doi.org/10.1016/j.paid.2012.12.022

Woodley, M. A., te Nijenhuis, J., & Murphy, R. (2013). Were the Victorians cleverer than us? The decline in general intelligence estimated from a meta-analysis of the slowing of simple reaction time. Intelligence, 41(6), 843–850. https://doi.org/10.1016/j.intell.2013.04.006