The Flynn Effect: Why IQ Scores Rose for Generations

Give people an intelligence test normed decades before they were born, and something strange happens: newer generations often look unusually smart. That recurring rise in raw IQ-test performance is the Flynn effect—and it reshaped how psychologists think about intelligence.

The Flynn effect is the long-term rise in intelligence-test performance across generations. For much of the twentieth century, newer cohorts repeatedly answered more IQ-test items correctly than older cohorts had at the same age. On many widely used tests, the historical gain was roughly 2 to 3 IQ points per decade.

That does not mean the average IQ kept climbing from 100 to 110 to 120 on newly published tests. Modern tests are periodically renormed so that the average IQ remains around 100. The Flynn effect becomes visible when performance is compared across different generations of test norms.

How Was the Flynn Effect Discovered?

Researchers had noticed generational test-score gains before James Flynn, but Flynn's work made the scale of the phenomenon difficult to ignore. In his landmark 1984 paper, The Mean IQ of Americans: Massive Gains 1932 to 1978, he compared successive Stanford-Binet and Wechsler standardization samples.

Every newer American standardization sample had performed better than the earlier one. Across the 46-year span, Flynn estimated a total gain of about 13.8 IQ points—nearly a full standard deviation on a modern IQ scale.

The pattern was later documented across many countries and tests. What initially looked like an American testing curiosity turned into one of the most famous findings in intelligence research.

How Big Was the Flynn Effect?

There is no single rate that fits every test, country, or decade. A 2014 meta-analysis covering 285 studies and 14,031 participants estimated an overall gain of 2.31 standard-score points per decade. When the analysis focused on modern Stanford-Binet and Wechsler comparisons normed since 1972, the estimate was about 2.93 IQ points per decade.

An even broader 2015 meta-analysis examined 271 independent samples, almost four million people, 31 countries, and data spanning 1909 to 2013. It confirmed substantial worldwide gains, but also showed that the size of the Flynn effect differed by cognitive domain and had weakened in more recent decades.

Fluid-intelligence tests showed larger average gains than crystallized-intelligence tests. Spatial and Full Scale IQ measures sat between them. In other words, generations were not improving equally on every kind of mental task.

What Kind of Intelligence Improved the Most?

The largest historical gains often appeared on abstract, unfamiliar problem-solving tasks—especially tests that require people to detect visual patterns, classify relationships, or reason without relying heavily on memorized knowledge.

This is one reason Raven's Progressive Matrices became central to Flynn-effect research. Matrix problems ask you to infer abstract relationships from visual patterns, and gains on this kind of reasoning were often striking.

Vocabulary and other knowledge-heavy abilities also improved in many datasets, but usually less dramatically. That uneven pattern is one clue that the Flynn effect was not simply “everyone learned more facts.” Modern environments may have changed the kinds of thinking people routinely practice.

Why Did IQ Scores Rise?

No single cause has won the debate. Reviews of the Flynn effect have proposed education, better nutrition and health, smaller families, more cognitively demanding work, urbanization, improved parenting environments, greater familiarity with testing, and increasingly complex visual and technological environments.

Several of those explanations have real evidence behind them. Education is especially important: a meta-analysis of 42 datasets and more than 600,000 participants found that an additional year of schooling was associated with roughly 1 to 5 IQ points of benefit, depending on the research design.

But schooling alone does not neatly explain why some cognitive domains rose faster than others or why gains differed across countries and time periods. The most plausible picture is a bundle of environmental changes acting together.

The “Social Multiplier” Idea

James Flynn and economist William Dickens proposed one influential way to understand how modest environmental changes could produce surprisingly large population gains. Their 2001 gene-environment model argued that people both influence and select environments that match their abilities.

Imagine a society becoming slightly more rewarding of abstract reasoning. People who are a little better at it may seek more cognitively demanding work, education, hobbies, and peers. Those environments then give them more practice, which can amplify the original difference. Similar feedback can operate across an entire generation.

The model also explains why the Flynn effect is compatible with substantial genetic influence on individual IQ differences. High heritability within a generation does not prevent environmental changes from shifting the population average. That distinction is explored more fully in Is IQ Genetic or Learned?.

Did People Literally Become 15 Points “Smarter”?

That question is harder than it sounds. A gain of 15 IQ-scale points on an old test does not necessarily mean a later generation has 15 more units of every form of intelligence.

The gains were too uneven across subtests for that simple interpretation. Some kinds of abstract reasoning improved dramatically, while other abilities moved much less. Flynn himself argued that modern people had become better at viewing the world through abstract categories and hypothetical rules—the kind of thinking increasingly rewarded by education, science, bureaucracy, and technology.

So the Flynn effect is unquestionably a change in measured cognitive performance. Exactly how much of that should be described as a rise in underlying general intelligence remains debated. Cognitive Train's guide to what IQ actually means explains why an IQ score combines several abilities rather than measuring one mental substance.

Why Does the Flynn Effect Matter in Real IQ Testing?

Because old norms can make a modern test-taker look artificially high. If a population improves after a test is normed, comparing today's raw performance with an older reference group can inflate the standardized score.

This is one reason major intelligence tests are periodically updated and renormed. It also matters in clinical and legal settings where a few IQ points can affect classification or eligibility decisions.

And it explains a seeming paradox: the population can gain raw points while the average IQ on a newly normed test remains 100. Renorming moves the reference frame.

Is the Flynn Effect Still Happening?

Not everywhere at the old rate. Some countries have shown plateaus or reversals, while others still show gains. In the United States, a 2024 analysis using WAIS-5 validation data found a positive Flynn effect of about 1.2 IQ points per decade—much smaller than the historical three-point rule of thumb.

That newer evidence suggests the story is changing rather than simply ending. Cognitive Train's companion guide Are Americans Getting Dumber? looks at the mixed U.S. evidence, while the next article in this series examines whether the Flynn effect is reversing across populations.

Check Your Own Approximate IQ

The Flynn effect describes population change, not what your own score should be. If you want to sample your 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 is the remarkable rise in IQ-test performance seen across much of the twentieth century. It was large, international, and strongest on some forms of abstract reasoning. The evidence points toward environmental change—especially education and increasingly cognitively demanding environments—but no single explanation accounts for every country, test, and decade.

Most importantly, the Flynn effect shows that population IQ scores are not frozen. Raw cognitive performance can shift substantially across generations even while newly normed IQ scales keep their average centered at 100. For more background, browse the Intelligence & IQ collection or explore Cognitive Train's brain training and cognitive training tools.