Has the Flynn effect actually reversed?

The short answer is yes — but with a footnote. Average IQ scores rose steadily across the industrialized world from the early 1900s through the mid-1990s, an increase of roughly 3 IQ points per decade in many countries. That rise is the Flynn effect, named after political scientist James R. Flynn, who documented it in a 1987 Psychological Bulletin paper showing massive gains across 14 nations. Since the 1990s, that curve has bent. Several countries have flatlined, and a handful — Norway, Denmark, the UK, France, parts of Australia and the US — show modest declines per decade. Researchers call this the reverse Flynn effect, or sometimes “negative Flynn effect.”

The Norwegian military data is the cleanest example. Jon Martin Sundet and colleagues at the University of Oslo published a 2004 Intelligence paper tracking the test scores of Norwegian conscripts from the 1950s through 2002. Gains were steady for half a century, then stopped after the mid-1990s, with numerical reasoning sub-tests actually declining. Thomas Teasdale and David Owen found the same pattern in Danish draft records, and published a 2005 paper titled “A long-term rise and recent decline in intelligence test performance: The Flynn effect in reverse.” A 2015 survey by Edward Dutton and Richard Lynn reported declines across most US states.

Why the headline is more nuanced than it sounds. A 2018 meta-analysis by Bernt Bratsberg and Ole Rogeberg, published in PNAS, looked at 29 OECD countries and argued that part of the apparent reversal is a measurement artifact. Modern IQ tests are restandardized every 10–15 years — the Wechsler scales were updated in 1949, 1974, 1991, 2003, and 2014 — so a raw score of 100 in 2020 doesn’t mean the same thing as a raw score of 100 in 1990. Once you control for that, the meta-analysis found the Flynn effect continues, just at a slower rate. The APA’s 2022 Monitor on Psychology summary notes that the question of whether we are actually getting dumber, or whether the tests are drifting, is still genuinely contested.

Why this matters beyond the trivia. The two strongest hypotheses for the original Flynn gains were better nutrition (especially childhood nutrition) and longer formal schooling. Stuart Ritchie, a psychologist at the University of Edinburgh, published a 2018 review covering nearly 30 studies and roughly 600,000 people; the consistent finding was that each additional year of schooling raises measured IQ by 1–5 points. If the gains were driven by education, then any reversal would suggest education quality is no longer improving at the same pace — or that something is offsetting it. The leading candidate for an offsetting factor in recent years is changes in screen time and reduced deep-reading habits during childhood, though the causal evidence is still thin.

The practical takeaway. You don’t need to worry that you or your children are suddenly getting dumber in a meaningful sense. The data does not show a collapse — it shows a plateau with mild downward drift in some developed countries. What is genuinely worth taking from the research is the Stuart Ritchie finding on schooling: the most durable lever on cognitive development is more and better formal education, not brain-training apps or supplements.

The reverse Flynn effect is real, documented, and modest. Whether it reflects an actual decline in cognitive ability, a drift in the tests themselves, or a combination is the part the experts are still arguing about.