The Flynn Effect Has Slowed and in Some Places Reversed

ByMax Dvornik
flynn effect: Students taking a standardized test in a classroom

The Flynn Effect is the long-run rise in average scores on standardized cognitive tests over the 20th century, and the best current answer is that it has slowed and in some countries and cohorts partly reversed. That does not mean human intelligence suddenly collapsed; it means later test-takers in some datasets are no longer outscoring earlier ones at the old rate, and in some places they score lower on the same kinds of measures.

Psychologist James R. Flynn helped document the pattern across decades of IQ-test renorming, where a test that drifts upward in raw scores has to be recalibrated so the average stays at 100. A rough rule of thumb in the classic literature was gains of about 3 IQ points per decade, though the size varied by country, period, and test.

What the Flynn Effect measures

The Flynn Effect is not a claim that every person got smarter in every sense. It refers to rising test performance on standardized cognitive measures across generations, especially on tasks involving abstract reasoning and visual problem-solving. In plain English: if you gave older test norms to later cohorts, later cohorts often looked better than the norm group the test was built on.

That distinction matters because IQ tests are periodically renormed. If raw scores climb over time, keeping the same norms would make the population average look higher than 100. Renorming resets the scale. The effect is therefore easiest to see by comparing old and new norming samples or by looking at large national datasets over time.

One useful way to think about it is that the effect tracks changes in the environments that shape test-taking performance: schooling, nutrition, health, family size, familiarity with formal testing, and the kind of abstract classification modern life demands. None of those explanations has been proven as the cause, but they are the usual candidates in the research literature summarized by Flynn’s early review and later cross-national work.

A simple derived calculation makes the old trend feel concrete: at the classic rate of about 3 points per decade, a country sustaining that rise for 50 years would see about 15 IQ points of gain on older norms. That is roughly one standard deviation on many IQ scales, a big shift on paper, even if it does not map neatly onto everyday wisdom.

Why scores may be flattening or falling

The newer evidence is why the topic now gets asked about in a different tone. Several datasets, especially from parts of Europe, suggest the old upward march has slowed, stalled, or reversed in some cohorts. The important qualifier is some: this is not a clean global cliff.

Researchers have proposed several explanations, none decisive on its own:

  • changes in education quality or reading habits,
  • demographic and family-pattern shifts,
  • environmental factors such as health or pollution exposure,
  • and the possibility that earlier gains came from improvements that eventually hit diminishing returns.

That last idea is the least dramatic and often the most plausible. If better childhood nutrition, less infectious disease, and broader schooling lifted scores for decades, you would expect the gains to taper once those basics stopped improving so quickly. A trend can flatten without needing a single villain.



Some of the strongest evidence for reversals comes from countries with unusually rich administrative data, including Norway, where researchers studying male conscription records found that IQ-score changes over cohorts were substantially driven by environmental factors within families, not fixed genetic change. That does not prove one cause of decline, but it does cut against simplistic stories that treat the newer pattern as biologically predetermined.

“The causes are environmental and can be found within families,” the authors of the Norwegian cohort study wrote after comparing brothers across time.

What the newer evidence does and doesn’t show

The best summary is that the Flynn Effect is real as a historical pattern, but its recent status is mixed. Some countries still show gains on some tests; others show plateaus or declines; and different subtests can move in different directions. A falling score on one measure is not the same thing as a universal drop in all cognitive abilities.

That is why broad claims like “IQ is reversing everywhere” run ahead of the evidence. The pattern is better described as uneven: long-run gains first, then flattening or reversal in some places and cohorts. The published NovaKnown overview on Flynn Effect current status gets the balance right here: the old effect was robust, but the newer phase is not one story with one cause.

There is also a measurement caution. IQ tests capture performance on specific tasks under specific norms. They are useful instruments, but they are not a complete meter for all forms of thinking, judgment, or knowledge. When scores move, the safest interpretation is that the environments and habits tied to those tested skills have changed.

So: yes, the Flynn Effect appears to be reversing in some datasets, but no, researchers have not pinned that on one proven mechanism, and no, it is not clearly happening everywhere. The honest answer is less tidy than the myth of endlessly rising IQ, and more interesting.

Key Takeaways

  • The Flynn Effect is the long-run rise in average scores on standardized cognitive tests across generations.
  • Classic estimates put the rise at about 3 IQ points per decade, though it varied by place and test.
  • Recent evidence suggests the trend has slowed or reversed in some countries and cohorts, especially in parts of Europe.
  • Researchers have not identified one proven cause of the newer flattening or declines.
  • The strongest careful conclusion is mixed: the historical rise was real, and the current picture is uneven.

Further Reading

Frequently Asked Questions

What is the Flynn Effect in simple terms?

The Flynn Effect is the historical pattern that later generations scored higher than earlier generations on many standardized cognitive tests, as documented in decades of test renorming and review work. It is about average test-score shifts over time, not a claim that every person is smarter in every way.

Is the Flynn Effect reversing?

In some countries and cohorts, yes. The stronger version, that it is reversing everywhere, for everyone, on every measure, is not supported by the evidence.

Why is the Flynn Effect reversing?

No single cause has been proven. Researchers have proposed education changes, environmental exposures, demographic shifts, and diminishing returns from earlier health and schooling gains, but the data do not yet point to one clean explanation.

Does a falling IQ score mean people are less intelligent?

Not necessarily. It means scores on particular tests, under particular norms, have flattened or fallen in some datasets. IQ tests measure useful but limited slices of cognitive performance, not the whole of human intelligence.

References

Last reviewed: 2026-06