The Flynn Effect was originally noted by Rundquist (1936) and Lynn (1982) and then Flynn (1984). Credit should probably go to Runquist, but a happy compromise is to call it the FLynn effect, in honour of the two major researchers. The history has been described by Lynn, in part of a Special Issue on the Flynn Effect in 2013. All the chapters in that issue are summarised here:
Far from ignoring the effect, psychometricians have been working to understand it, and in the three years since the special issue further advances have been made in tracking down possible causes. Most recently, the theory that the Flynn effect results in part from an increase in the use of abstract reference frames in solving cognitive problems has been supported. I heard Jim Flynn propose that in 2007 and thought it unlikely, but the Estonian results support his hunch.
The most comprehensive recent summary is Pietschnig, J., & Voracek, M. (2015). One Century of Global IQ Gains: A Formal Meta-Analysis of the Flynn Effect (1909-2013) and can be found here:
The effect is strongest for fluid intelligence, and least on crystallised intelligence, though that has gone up as well. The bad news is that the effect is reducing in size, and even reversing in some countries.
Now to expert opinion. Experts should know more, be brighter than average, and to have practiced their trade for longer. Few will reach that level, and fewer will excel.
To be considered an expert for this survey, experts had published articles in or after 2010 in journals on intelligence, cognitive abilities and student achievement. The journals included Intelligence, Cognitive Psychology, Contemporary Educational Psychology, New Ideas in Psychology and Learning and Individual Differences. Notice of the study was also emailed to members of the International Society for Intelligence Research (ISIR), and posted to the web site for the International Society for the Study of Individual Differences (ISSID). ISIR and ISSID support intelligence research and host professional conferences with intelligence researchers. Finally, the study was announced at the 2013 ISIR conference in Melbourne, Australia.
Are such people expert enough? First, here is the paper and the link:
Survey of expert opinion on intelligence: The FLynn effect and the future of intelligence. Heiner Rindermann, David Becker, Thomas R. Coyle. Personality and Individual Differences 106 (2017) 242–247
Experts on intelligence, cognitive ability and student achievement were surveyed for their opinions on the causes of the 20th century rise in intelligence test results called the “FLynn effect”, on the causes of a possible end of the FLynn effect and on the future development of IQ in different world regions. Ratings from N=75 experts attributed the secular IQ rise to better health and nutrition, more and better education and rising standards of living.
Genetic changes were seen as not important. A possible stagnation or retrograde of the FLynn effect was attributed to asymmetric fertility (genetic and socialization effects), migration, declines in education and the influence of media. Experts expected 21st century IQ increases in currently on average low-ability regions (+6 to +7 IQ points, in Latin America, Africa, India) and in East Asia (+7 IQ), but not in the West (a stagnation, below +1 IQ), with a small decline in the US (−0.45 IQ). Similar results were obtained for all experts and experts on the FLynn effect itself (mean r=0.90 to 0.97; N=17). The results correlated strongly with and confirmed a recent meta-analysis on the causes of the FLynn effect (r = 0.65 to 0.71; Pietschnig & Voracek, 2015).
We received a total of 265 responses from May 2013 to March 2014, at which time the survey was closed. The response rate was 20% of all invitations. The present article focused on the three questions on FLynn effects, which were answered by 75 respondents.
Incidentally, some commentators have suggested the sample size of experts is too small. It is a pity that many avoid giving their opinions, despite the anonymity of the survey. That said, the sample is probably too big. To be a real expert is difficult. Publishing a few papers is not enough, though you still know more about the topic than most people. Although I have read some of the literature, and have published on the subject, and edited the special issue of Intelligence, in my eyes I am not an expert. One needs to publish more than that, and over a longer period. Perhaps 50 at most can claim to be expert, and that may be too many. These authors selected 17 whom they judged to be really expert, and that feels about right. Some samples are better when they are small. The survey closed in 2014 the authors quote the results and then compare them to Pietschnig and Voracek’s 2015 meta-analysis as a benchmark. For once, we can work out whether the experts are really expert.
The authors say: For example, secular declines have been found for mental speed, digit span backwards, the use of difficult words, and color acuity, all of which are related to intelligence (Madison, Woodley of Menie & Sänger, 2016; Woodley et al., 2013; Woodley of Menie & Fernandes, 2015; Woodley of Menie, Fernandes, Figueredo, & Meisenberg, 2015; Woodley of Menie & Fernandes, 2016).
Three decades ago, Flynn (1987) argued that genetic changes could not explain the massive IQ gains in the 20th century. Our expert survey supports Flynn’s view. Experts tended to attribute the FLynn effect not to genetic factors but to environmental factors such as better health, more and better school education, better nutrition, rising standards of living, better education in families and better educated parents. Health improvements were rated as the most important cause.
Why is the Flynn effect petering out, or even reversing in the West?
“Having more children” and migration seem to be the main causes. Education systems and mass media also come in for criticism. I am not sure about these.
Tempting fate, the authors pressed experts to guess what the future of intelligence would be for different regions of the world.
Poor countries are predicted to keep raising their game, and their intellects, richer countries less so. The USA is the only country predicted to decline in ability, presumably because of mass migration. The real experts take an even more jaundiced view, and hold out little hope for The West. These predictions will be partly testable within one generation, so pin this table to your study notice board, and test for goodness of fit in 2040.
The authors conclude:
The strong correlations between our expert ratings and the results of Pietschnig and Voracek’s (2015) study confirm the validity of expert opinions. Such opinions can stimulate future research and help to approximate truth (Rindermann et al., 2016). (Our survey was conducted before Pietschnig and Voracek’s (2015) article, which could not have anticipated our results). Of course, majority is not verity and expert opinions are no guarantee of truth, but experts tend to be more accurate than laypeople. In this regard, the similar results between the FLynn experts and all other experts are noteworthy. Ratings from the two groups (FLynn experts vs. general experts) correlated at r= 0.91 for FLynn effect causes, r = 0.91 for FLynn end causes and r = 0.88 for the prediction of future development (on average r = 0.90). Compared to all (only general) experts, FLynn experts rated education more highly for the FLynn effect; rated dysgenic effects of asymmetric birth rates and migration more highly for the end of the FLynn effect; and were more pessimistic about future gains in intelligence in the 21st century, especially in the West.
The delightful paradox of the FLynn effect is that all deliberate policies to boost IQ have failed, and the only apparent IQ boost has been achieved inadvertently as an indirect consequence of improving general health and education. This is good news.
I have forgotten how I answered the survey. My own view at the moment is that the closer a mental task approaches the requirements of a ratio scale, with a non-arbitrary zero point, as explained by SS Stevens, the less it is affected by IQ inflation. There is no overall Flynn Effect on digit span and little on maths, though Digit Span forwards may have increased a little, and digit span backwards decreased a bit.
Doing mental puzzles is a better-known activity than it used to be decades ago, so there is a familiarity with testing approaches, but I doubt that generalizes to new, real life problems. Tests are approximations, after all, and if they get to be too well known in their general approach the less they represent the shock of really new problems, which are the ultimate intelligence test. People are more willing to guess on tests. Earlier generations were more cautious. There is a small contribution because of increases in brain size. Although the experts give much emphasis about the positive effects of longer education for more people, I agree that there is convincing data in support of that effect, but am a bit unsure about the long-term magnitude of the effect.
So, a summary would be that the improvements in health and education that have been enjoyed by most people in the world over that last century have probably boosted their mental ability, but in richer countries the effect has levelled off, and in several it has gone down.
Keep watching the skies.