Is the Flynn Effect Reversing?

For decades, IQ-test performance seemed to move in one direction: up. Then Norway, Denmark, Finland, parts of the United States, and other wealthy populations began producing a much stranger pattern—flat scores, smaller gains, or outright declines.

Yes, the Flynn effect has reversed in some populations—but it is not reversing everywhere. Several affluent countries have shown falling cognitive-test scores after decades of gains. Yet recent global research still finds positive Flynn effects in many middle-income and developing countries, and even the newest U.S. Wechsler data show a small positive gain rather than a decline.

The best description is therefore divergence: the historical rise in IQ scores has weakened, plateaued, or reversed in some places while continuing elsewhere. To understand why this is surprising, start with what the Flynn effect is and why IQ scores rose for generations.

Norway Shows a Clear Score Reversal—but What It Means Is Debated

Norwegian military-conscription data are especially valuable because they cover a huge share of young men over many birth cohorts. IQ-test performance rose for decades, peaked among men born around the mid-1970s, and then declined in later cohorts.

A 2018 study by Bernt Bratsberg and Ole Rogeberg went further by comparing brothers born into the same families across three decades of Norwegian birth cohorts. The researchers found that the rise, turning point, and later decline could all be reproduced within families.

That matters because brothers share parents and much of their genetic background. The result strongly argues against explanations based mainly on changing parental composition—such as the idea that the reversal simply reflects people with lower IQ having more children. Bratsberg and Rogeberg therefore interpreted the observed rise and reversal as primarily environmental.

A 2025 reanalysis of the Norwegian Armed Forces data added an important qualification. It found that the test did not achieve scalar measurement invariance across generations. The historical rise was driven heavily by improvement on figure-matrix reasoning, while the later decline was driven more by word comprehension and numerical reasoning. The authors argued that these changing subtest patterns do not clearly demonstrate equivalent changes in the underlying latent general-intelligence factor.

So Norway still shows a genuine generational reversal in observed cognitive-test performance. What is less certain is whether that reversal should be interpreted as an equally clear fall in latent general intelligence.

Denmark Shows a Similar Turning Point

Danish military data tell much the same story. A study following more than 500,000 young men found that cognitive-test performance improved from 1959 through the late 1990s and then began declining.

A later analysis reported that scores fell by roughly 1.5 IQ points between 1998 and 2003–2004, nearly erasing the modest gain made during the previous decade. The decline appeared among men pursuing higher academic education as well as those who were not.

Norway and Denmark therefore provide convincing examples of a reverse Flynn effect: newer cohorts scoring lower than the cohorts immediately before them on comparable cognitive tests.

But the Global Flynn Effect Has Not Disappeared

This is where the story becomes easy to oversimplify. A 2015 meta-analysis covering nearly four million people in 31 countries found that IQ-test gains had weakened in more recent decades, but the century-long global trend remained positive.

An even newer 2023 meta-analysis focused on Standard Progressive Matrices. It combined 1,038 samples, 299,155 people, 72 countries, and data from 1948 to 2020. The overall Flynn effect was still about 2.2 IQ points per decade. Gains were stronger in middle-income countries and among younger age groups.

Another 2023 global analysis described the pattern as a kind of convergence: recent gains remain common in less economically developed countries, while the richest countries increasingly show tiny gains, plateaus, or declines.

So asking whether “the Flynn effect” is reversing worldwide is a little like asking whether world temperatures are rising by looking at one city. The direction depends on where and what you measure.

The United States Is Mixed, Not Simply Reversing

A large 2023 U.S. study found evidence of a reverse Flynn effect in 394,378 adults tested online between 2006 and 2018. Composite ability, matrix reasoning, and number-series scores declined.

But three-dimensional rotation scores increased. And the sample was self-selected rather than nationally representative.

Then the 2024 WAIS-5 comparison produced another wrinkle: Full Scale IQ still showed a positive Flynn effect of about 1.2 points per decade. That is dramatically smaller than the historical three-points-per-decade rule, but it is not a reversal.

Cognitive Train's Are Americans Getting Dumber? article examines those U.S. findings in detail. The short version is that American cognitive trends are increasingly domain-specific rather than uniformly downward.

Why Would the Flynn Effect Reverse?

There is no established single cause. Researchers have proposed changes in schooling, nutrition, health, family structure, migration, environmental exposures, technology use, reading habits, test familiarity, and the kinds of cognitive tasks modern life rewards.

Some explanations are more persuasive than others. The Norwegian sibling study is particularly important because it makes a simple genetic-composition explanation much harder to defend. The decline occurred within families: later-born brothers tended to follow the same cohort trend as the wider population.

But the 2025 measurement-invariance reanalysis changes how strongly that result can be interpreted. It suggests that the shifting Norwegian scores may reflect changing performance in particular cognitive domains rather than a uniform movement in latent general intelligence. Environmental cohort change remains central to the score pattern, but the exact cognitive construct that changed is less settled than the raw trend alone suggests.

Could the Original Flynn Effect Simply Be Reaching a Ceiling?

This is one plausible part of the story. Many twentieth-century improvements—universal schooling, better nutrition, reduced childhood disease, smaller families, more cognitively demanding jobs, widespread visual media—cannot keep improving at the same pace forever.

If a wealthy country has already captured much of the cognitive benefit available from those changes, future gains could naturally shrink. That could help explain why newer studies often find stronger Flynn effects in middle-income countries than in the richest ones.

A ceiling does not automatically produce a decline, however. It explains why gains might slow toward zero. Something additional would still be needed to explain sustained negative trends.

Are Smartphones or the Internet Causing IQ Scores to Fall?

There is not enough evidence to say that. Digital technology is an obvious suspect because its spread overlaps with the period in which some countries' scores flattened or fell. But timing alone is not causation.

Technology may reduce practice in some skills while increasing practice in others. The recent U.S. finding—declines in matrix and series reasoning alongside gains in three-dimensional rotation—is a good reminder that cognitive abilities can move in different directions at the same time.

The next article in this cluster, Why Are IQ Scores Falling in Some Countries?, looks more closely at the competing explanations rather than treating any one of them as settled.

Does a Reverse Flynn Effect Mean People Are Becoming Less Intelligent?

It means newer cohorts are performing worse on the cognitive measures showing the reversal. Whether that equals a decline in broad underlying intelligence depends on the test and on which abilities changed.

The original Flynn effect itself was uneven: abstract reasoning often rose faster than knowledge-heavy abilities. The reversal can also be uneven. This is why what IQ tests actually measure matters when interpreting generational trends.

It is also different from individual aging. A population trend compares people born in different years; IQ changes across the lifespan track what happens as one person grows older.

So, Is the Flynn Effect Reversing?

In some countries and on some tests, clearly yes. Norway and Denmark provide strong evidence of genuine reversals after decades of gains. Several other wealthy populations show plateaus or negative trends, and some recent U.S. abilities have declined.

Globally, no. Large modern meta-analyses still find positive generational gains overall, especially in middle-income countries. The historic Flynn effect appears to be changing from a broadly rising phenomenon into a patchwork of gains, plateaus, and declines.

Check Your Own Approximate IQ

A population trend cannot predict your personal score. If you want to sample your own verbal, logical, numerical, and visual-spatial reasoning, Cognitive Train's free IQ-Style Test gives a non-clinical, approximate estimate based on your performance on the test itself.

For individual measures of reasoning, memory, attention, and processing speed, browse Cognitive Train's brain tests and cognitive assessments.

The Bottom Line

The Flynn effect has not simply gone into reverse worldwide. Affluent countries increasingly show reverse Flynn effects in observed cognitive-test scores, while many less-developed populations still show rising scores. Norway's within-family evidence makes changing family composition an unlikely main explanation for its score trend, while the 2025 measurement-invariance reanalysis cautions that the rise and decline do not map cleanly onto equivalent changes in latent general intelligence. Researchers still do not have one agreed explanation for what changed.

The emerging picture is more interesting than a universal rise or fall: population cognitive performance responds to changing environments, and different societies may now be at different points in that process. Browse the Intelligence & IQ collection or explore Cognitive Train's brain training and cognitive training tools for more ways to understand and measure cognitive performance.