The Flynn Effect: Why IQ Scores Rose ~3 Points a Decade — And Why They're Now Falling

Give a group of people from 1950 a modern IQ test, scored against today's norms, and they would average well below 100. Do the reverse — score today's population against 1950's norms — and the average leaps to something like 118. Nobody's brain physically changed. The reference point did. That drift is the Flynn effect, and it is the reason every serious IQ test has to be re-normed every decade or two.

Named after political scientist James R. Flynn, who documented it systematically across more than a dozen nations in the 1980s, the effect is one of the most robust findings in the study of what IQ actually measures. It also quietly undermines any test that brags about a fixed, permanent score — because the yardstick itself keeps moving.

Curious where you land against current norms? Our free IQ test (~30 questions, ~20 minutes, verbal and non-verbal, instant result, no email) is normed on a recent sample — which, as you'll see below, is the whole point.

What the Flynn effect is

An IQ score is not an absolute count of anything. It is a rank — a statement of where you sit relative to a reference group whose average is fixed at 100. When test publishers re-standardize a test, they gather a fresh sample and reset that average to 100 again. The Flynn effect shows up in the gap between old and new norms: raw performance keeps climbing, so people scored against an outdated sample look inflated.

Flynn's key evidence came from cases where the same people, or comparable groups, took two tests normed at different times. They consistently scored higher on the test with the older norms — proof that the population had drifted upward relative to the earlier baseline. For the mechanics of how raw answers become a standardized number, see our explainer on how IQ tests are scored.

How big it was — and where the gains landed

The "3 points per decade" figure is a useful average, not a law of nature. The largest formal meta-analysis, by Pietschnig and Voracek (2015), pooled 271 samples and nearly four million participants across 31 countries. It found the gains were uneven across cognitive domains — and, revealingly, largest exactly where formal schooling teaches least directly:

This pattern is the crux of the whole puzzle. If more and better schooling were the sole engine, crystallized knowledge — vocabulary, general facts — should have risen fastest. Instead the steepest gains were on culture-reduced fluid-reasoning tests like Raven's Matrices, where earlier analyses reported gains as high as 5–7 points per decade. The meta-analysis by Trahan and colleagues (2014), across 285 studies, put the overall average at 2.31 points per decade, closer to 2.93 on modern Wechsler and Stanford-Binet tests.

Why scores rose

There is no single agreed cause. The mainstream view, summarized by Neisser (1997) and reinforced by later reviews, is that several environmental forces acted together and reinforced one another:

• Nutrition and health. Better early nutrition, fewer childhood infections, and dramatically reduced lead exposure plausibly raised the biological ceiling on cognitive development — the same forces that made each generation physically taller.
• Schooling. More years of education, reaching more of the population, and teaching increasingly abstract, categorical thinking. Even after controlling for education, though, substantial gains remain.
• Smaller families. Fewer children per household meant more adult attention and verbal interaction per child — a pattern connected to the way scores shift with age and cohort.
• A more abstract environment. Modern life increasingly demands on-the-spot logic, symbol manipulation, and hypothetical reasoning — precisely the mental habits fluid-reasoning tests reward.

Because these gains unfolded within two or three generations — far too fast for the gene pool to change — genetic explanations are ruled out. Whatever drove the Flynn effect, it was environmental.

Flynn's own explanation: "scientific spectacles"

Flynn himself resisted the headline reading that his effect proved everyone was getting smarter. His interpretation was subtler: the 20th century taught people to wear what he called "scientific spectacles" — to treat the world in terms of abstract categories, logical hypotheticals, and formal classification rather than concrete, practical particulars.

Asked how a rabbit and a dog are alike, someone in 1900 might have answered in concrete terms ("you use dogs to hunt rabbits"). A modern respondent reaches instantly for the abstract category ("they're both mammals"). IQ tests reward the second style. On Flynn's account, the population didn't grow more intelligent in some raw sense so much as it grew fluent in the specific abstract idiom that intelligence tests happen to measure.

The reversal: the "negative Flynn effect"

The rise was not permanent. From around the 1990s, average scores stopped climbing and began to fall in several developed countries — a pattern usually called the reverse or negative Flynn effect.

The strongest evidence comes from Norway. Bratsberg and Rogeberg (2018) analyzed cognitive test scores from more than 730,000 male military conscripts born between 1962 and 1991. Scores peaked with the 1975 birth cohort and then declined by roughly 0.2 points per year. Crucially, the decline appeared within families — younger brothers scored lower than their older brothers — which rules out both immigration and any dysgenic genetic story, because siblings share a gene pool. Their blunt conclusion: the Flynn effect and its reversal are both environmentally caused. Similar declines or plateaus have been reported in Denmark, Finland, France, and the UK; a systematic review by Dutton, van der Linden and Lynn (2016) catalogued the trend across countries.

The United States shows a messier picture. Dworak, Revelle and Condon (2023), analyzing a large online adult sample from 2006 to 2018, found declining scores in three of four domains — verbal reasoning, matrix reasoning, and letter-and-number series — while spatial (3D rotation) ability actually rose. Lead author Elizabeth Dworak cautioned against reading this as "Americans getting less intelligent," noting the differences may partly reflect test-taking motivation and shifting priorities rather than raw ability. That mixed signal matters: if every domain fell together, you could tell a tidy decline story — but they don't.

Was any of it "real" intelligence?

Here is where the newest work gets interesting — and cautious. A long-running question is whether Flynn gains reflect a rise in the underlying general factor, g, or just improvements on specific, coachable test features. The meta-analytic answer from te Nijenhuis and van der Flier (2013) was largely no: the gains were not primarily on g.

A 2025 re-analysis pushed this further. Nordmo, Norrøne and Lang-Ree re-examined the Norwegian Armed Forces data — where the same three subtests have been administered for decades — using tests for measurement invariance. Their finding: the observed rise and fall failed to achieve scalar invariance across time, meaning the scores can't be cleanly compared as if they measured the same latent trait in every era. Much of the movement traced to specific subtests (figure matrices and word similarities) rather than to general intelligence, and likely reflected educational and cultural shifts. In plain terms: the trend line is real, but reading it as a straightforward rise-and-fall of raw brainpower is not warranted.

This is exactly why we're skeptical of any test — including anyone's — that markets a single permanent, ultra-precise IQ figure. Score meaning always lives inside a population reference system that changes over time. For more on how far to trust a number, see our piece on IQ test accuracy.

What the Flynn effect means for your own score

Three practical takeaways:

• Norm date is everything. A high score against 1990s norms is not the same as a high score against current ones. A test that hasn't been re-normed in decades will systematically flatter you. When you read an IQ score chart, the percentile it implies is only as current as the sample behind it.
• Old family scores don't transfer. A grandparent's 125 from a mid-century test and your 125 today were measured against different populations — they aren't the same rank.
• Environment moves scores. The Flynn effect is the population-scale proof that cognitive test performance responds to environment. That's also the honest, evidence-based backdrop to our guide on how to increase IQ — with realistic expectations about what actually shifts.

Frequently asked questions

Related reading

← Browse all articles on the blog

References

1. Flynn, J. R. (1984). The mean IQ of Americans: Massive gains 1932 to 1978. Psychological Bulletin, 95(1), 29–51.
2. Flynn, J. R. (1987). Massive IQ gains in 14 nations: What IQ tests really measure. Psychological Bulletin, 101(2), 171–191.
3. Flynn, J. R. (2012). Are We Getting Smarter? Rising IQ in the Twenty-First Century. Cambridge University Press.
4. Neisser, U. (1997). Rising scores on intelligence tests. American Scientist, 85(5), 440–447.
5. 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.
6. Trahan, L. H., Stuebing, K. K., Fletcher, J. M., & Hiscock, M. (2014). The Flynn effect: A meta-analysis. Psychological Bulletin, 140(5), 1332–1360.
7. te Nijenhuis, J., & van der Flier, H. (2013). Is the Flynn effect on g? A meta-analysis. Intelligence, 41(6), 802–807.
8. 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(4), 349–362.
9. Teasdale, T. W., & Owen, D. R. (2008). Secular declines in cognitive test scores: A reversal of the Flynn effect. Intelligence, 36(2), 121–126.
10. Dutton, E., van der Linden, D., & Lynn, R. (2016). The negative Flynn effect: A systematic literature review. Intelligence, 59, 163–169.
11. Bratsberg, B., & Rogeberg, O. (2018). Flynn effect and its reversal are both environmentally caused. Proceedings of the National Academy of Sciences, 115(26), 6674–6678.
12. Dworak, E. M., Revelle, W., & Condon, D. M. (2023). Looking for Flynn effects in a recent online U.S. adult sample: Examining shifts within the SAPA Project. Intelligence, 98, 101734.
13. Nordmo, M., Norrøne, T. N., & Lang-Ree, O. C. (2025). Reevaluating the Flynn effect, and the reversal: Temporal trends and measurement invariance in Norwegian armed forces intelligence scores. Intelligence, 110, 101909.
14. Nisbett, R. E., Aronson, J., Blair, C., Dickens, W., Flynn, J., Halpern, D. F., & Turkheimer, E. (2012). Intelligence: New findings and theoretical developments. American Psychologist, 67(2), 130–159.