The Flynn Effect is not on g: A Reply to Mr Dean
Edward Dutton responds to Abel Dean
This is a response to the following article:
Written by Edward Dutton.
Mr Dean argues that the Flynn Effect – the secular rise in IQ scores of about 2.5 points per decade across the twentieth century – is on g, meaning we have become much more intelligent across time. The alternative argument is that the Flynn Effect is on specialised abilities, such as “Similarities,” which very weakly correlate with intelligence, and that this happened so intensely (due to living in an increasingly scientific society) that it has come across on the (imperfect) IQ tests as an IQ rise.
The latter argument is congruous with the various “Woodley Effects” that have been set out in the book At Our Wits’ End: Why We’re Becoming Less Intelligent and What It Means for the Future by myself and Michael Woodley of Menie where this argument is also presented (Dutton & Woodley of Menie, 2018). These “Woodley Effects” include lengthening reaction times, worsening colour discrimination, worsening backward digit span, declining per capita major innovation, and decline across time in the prevalence of alleles associated with high IQ. Mr Dean’s argument begs the question of why we are not seeing the real world effects of rising IQ, such as increasing per capita genius, quickening reaction times, less mental retardation and so on. Thus, the argument in At Our Wits’ End makes sense of the most data.
However, Mr Dean provides a case for his viewpoint by arguing that: “A negative slope implies neither a stall nor a fall of g, but it means merely that the rise of the non-g component is greater than the rise of the g component (the non-g component can be likewise estimated by extending the line to where g-loading is zero).” This is known as the “Brand Effect” and it was addressed by Woodley of Menie and colleagues almost a decade ago (Woodley et al., 2014).
Most of these IQ tests are multiple-choice. People utilise guessing more today than they did in the past: we know this because historically people were more likely to leave items blank if they didn’t know the answer. English psychologist Chris Brand (1943-2017) predicted many years ago that this was the source of the Flynn Effect (e.g. Brand, 1996). He was partially right. He noted, for example, that each correctly guessed Ravens item would raise your apparent IQ by 7 points. What sorts of items are more likely to trigger guessing: easy ones, or hard ones? The answer, of course, is the latter; more g-saturated items are more likely to be guessed at, meaning that gains on these items are illusory.
Woodley of Menie and colleagues (Woodley et al., 2014) showed that this “Brand Effect” - the portion of the IQ gains that are due purely to correctly guessed items - is a strong “Jensen Effect” (it is on g). This makes sense, as the harder the item, the more likely it is to be guessed at. Therefore, you must control the “authentic Flynn Effect” for guessing gains (for example, you strip out the separate “Brand Effect”), and then the vector correlation essentially goes to -1, consistent with the argument that the Flynn Effect is not on g.
A more in-depth reply will be written if and when Mr Dean’s article is published in an academic journal.
Edward Dutton is a researcher based in Oulu, Finland. He has published in leading journals, including Intelligence and Personality and Individual Differences. He is Professor of Evolutionary Psychology at Asbiro University in Lódz, Poland, and an academic consultant to a research group at King Saud University in Riyadh, Saudi Arabia. He runs a popular online show, The Jolly Heretic, in which he explores controversial research, and interviews academics and dissidents. He is the author of many books and you can follow him on Twitter here.
References
Brand, C.R. (1996). The g Factor: General Intelligence and Its Implications. Chichester: Wiley.
Dutton, E. & Woodley of Menie, M.A. (2018). At Our Wits’ End: Why We’re Becoming Less Intelligence and What It Means for the Future. Exeter: Imprint Academic.
Woodley, M.A., te Nijenhuis, J., Must, O. & Must, A. (2014). Controlling for increased guessing enhances the independence of the Flynn effect from g: The return of the Brand effect. Intelligence, 43: 27-34.