.png)
If you gave many people from a century ago a modern IQ test and scored it with today’s norms, a shocking number would look alarmingly below average. That sounds insulting to your great-grandparents, and honestly it should. They built railways, fought wars, raised families, and somehow survived without Wi-Fi. So what is going on?
This puzzle is the Flynn effect: the finding that average IQ test scores rose from one generation to the next across much of the 20th century. James Flynn, the New Zealand political philosopher who gave the effect its name, first documented massive gains in the United States in 1984 and then across 14 countries in 1987. The pattern was simple and bizarre at the same time: each new generation tended to score higher on the same kinds of tests than the generation before it.
And not by a tiny amount. In a major meta-analysis, Trahan, Stuebing, Hiscock, and Fletcher reported in 2014 that IQ scores increased by about 2.31 points per decade across all studies, and by about 2.93 points per decade in modern comparisons using major tests like the Wechsler and Stanford-Binet. That is roughly the classic “3 points per decade” figure you may have heard. Over a lifetime or two, that becomes huge.
This is why IQ tests have to keep being renormed
IQ tests are not scored against some eternal cosmic standard. They are scored relative to a norm group, usually set so that 100 is the average — if you want a deeper look at what that baseline means, our article on what the average IQ is and what it means covers it well — for the population at the time the test is standardized. In other words, 100 is average by design. As later generations get better at the test, the old norms become stale. Keep using them, and people start looking smarter on paper even if nothing dramatic changed inside their skulls overnight.
This is not a technical footnote. It matters in the real world. Trahan and colleagues pointed out that outdated norms can influence high-stakes decisions in education and even death penalty cases. In one legal context, a few IQ points could affect whether someone is considered intellectually disabled. That is not the kind of clerical error you want to shrug off with a “close enough.”
So the Flynn effect is not just “people got smarter.” It is also “our measuring stick keeps drifting.” And once you see that, the next question becomes unavoidable: what exactly was rising?
The biggest gains were often on abstract reasoning
This is where the story gets more interesting. The gains were not evenly spread across every kind of mental task. According to the global meta-analysis by Jakob Pietschnig and Martin Voracek in 2015, fluid intelligence tests tended to show larger increases than crystallized intelligence tests. Their estimates were about 0.41 IQ points per year for fluid ability, compared with 0.21 for crystallized ability.
That sounds dry, but the idea is simple. Fluid tasks ask you to solve new problems, spot patterns, or reason with abstract shapes. Crystallized tasks lean more on vocabulary, facts, and learned knowledge. Tim Folger wrote in Scientific American in 2012 that many of the biggest gains showed up on tasks like similarities and matrices, while vocabulary and arithmetic were much flatter. In other words, modern people did not necessarily become walking dictionaries. We became better at a certain style of thinking.
Flynn had a memorable way to describe this. In a Scientific American interview, he said modern people have put on “scientific spectacles.” His point was that daily life increasingly trains us to classify, compare, abstract, and reason outside immediate concrete experience. Your ancestors might have been better than you at identifying weather shifts from the smell of the air or fixing a machine with wire and stubbornness. But on decontextualized pattern puzzles, you probably have the edge. Congratulations, I guess.
Still, specialists disagree on what this means. Some psychometric researchers argue that the Flynn effect reflects gains in specific skills more than a sweeping rise in g, as we explore in our piece on what general intelligence (the g factor) really means. Recent work by Andrzejewski and colleagues in 2024 even suggests that while some scores are still rising in Austria, the underlying structure of abilities may be changing in subtle ways. That is an active debate, not a closed case.
The strongest explanations all point to the environment
Whatever the Flynn effect is, it almost certainly is not a genetic evolution story. Human genes do not reshuffle fast enough to raise average IQ test performance by roughly 3 points per decade across many countries — if you are curious how much of intelligence is fixed by genetics in the first place, our article on whether intelligence is hereditary explores exactly that. The timescale alone basically sends that idea home early.
The strongest explanations are environmental, and probably cumulative rather than single-cause. Better nutrition is a major candidate. So is better health, especially in childhood. Smaller families may have meant more adult attention and resources per child. More schooling almost certainly helped too, but even Flynn and later reviewers argued that education alone cannot explain gains of this size. As Pietschnig and Voracek concluded in 2015, the most plausible story is a mix of broad environmental changes rather than one magic lever.
There is also the broader texture of modern life. We now grow up surrounded by maps, interfaces, symbols, diagrams, categories, screens, and rule-based systems. Even childhood play changed. A modern child spends years navigating abstract visual worlds before adulthood, sometimes before tying shoes properly (human development is a chaotic masterpiece). That kind of environment may quietly train the exact skills many IQ tests reward.
There is even more direct evidence for the environmental view. In a 2018 PNAS study, Bernt Bratsberg and Ole Rogeberg analyzed Norwegian conscript data and found that both the rise and the later decline in IQ scores could be recovered within families, comparing brothers born in different years. That matters because siblings share parents and much of their background. If the pattern appears within families, it becomes much harder to blame changing national genetics or immigration. Their conclusion was blunt: both the Flynn effect and its reversal were environmentally caused.
The modern picture is messier than the classic story
Now we hit the part that ruins the neat headline. If the first half of the story was “modern life lifted scores,” the second half is “not everywhere, not forever.” In several countries, especially parts of Northern Europe, the pattern has slowed, flattened, or even reversed.
A review by Dutton, van der Linden, and Lynn in 2016 identified nine studies across seven countries reporting negative Flynn effects. Bratsberg and Rogeberg’s Norwegian data showed gains for cohorts born from the early 1960s into the mid-1970s, followed by declines afterward. Pietschnig and Voracek also found evidence that gains decelerated in more recent decades.
And it gets messier. A 2019 study by Platt, Keyes, McLaughlin, and Kaufman using more than 10,000 U.S. adolescents found that the Flynn effect was not uniform at all. Thirteen-year-olds showed gains of about 2.3 points per decade, but by age 18 the trend had flipped to a decline of about 1.6 points per decade. High-scoring adolescents gained, while the lowest-scoring group actually lost ground. So when someone talks about “the Flynn effect” as if it behaved identically everywhere, for everyone, at all times, that is your cue to squint politely.
At the same time, the effect has not vanished everywhere. Andrzejewski and colleagues found positive Flynn effects across all IQ domains in Austrian data from 2005 to 2018. Earlier work by Pietschnig, Formann, and Voracek in 2010 also found substantial gains in German-speaking samples. So the modern picture is mixed: some places still rise, some stall, some dip.
So are we actually smarter than our grandparents?
Yes, no, and annoyingly, both.
We are clearly better, on average, at many of the tasks IQ tests ask people to do. The evidence for that is strong. Across millions of people and many countries, score gains were real. But that does not automatically mean we are superior in every meaningful sense of intelligence.
Remember what we saw earlier: the largest gains often appeared on abstract and nonverbal tasks. That detail matters. It supports Flynn’s view that modern culture changed the style of thinking people bring to problems. We may be better at classifying, comparing hypotheticals, and manipulating symbols. That is important. It is not fake. But it is also not the same thing as proving that humans now have more wisdom, better judgment, deeper creativity, or stronger common sense. If only IQ tests could measure “not posting nonsense online at 2 a.m.,” civilization would be thriving.
If you want to explore the generational side of this in more depth — including the real-world legal stakes and a practical formula for adjusting your own IQ score — our article on how every generation is becoming more intelligent than the previous one covers those angles directly.
This is also why some researchers resist saying the Flynn effect shows a simple rise in general intelligence. The gains may partly reflect better alignment between modern minds and modern tests. That is still fascinating. It means intelligence scores are more shaped by culture and development than people once thought.
Why psychologists care so much about this
The Flynn effect changed the field because it disrupted a comfortable assumption: that IQ scores were stable snapshots of a fixed mental trait. Instead, the scores turned out to be surprisingly sensitive to historical change.
As psychologist Timothy Salthouse showed in 2015, these generational shifts even complicate how we study cognitive aging. If later-born adults score higher than earlier-born adults at the same age, some apparent age differences can reflect cohort effects rather than pure aging. In other words, the Flynn effect can sneak into studies and make time itself look guilty.
And the practical implications keep coming. Old norms can inflate scores, delay special education identification, distort clinical judgment, and affect legal decisions. That is why test publishers keep updating norms, and why good psychologists care about which version of a test was used, not just the final number. An IQ score without context is like a bathroom scale on a tilted floor: still a measurement, but maybe not one you want to build a court case on.
The Flynn effect also leaves us with a larger, more hopeful idea. Human cognitive performance is not sealed off from the world. Change the world enough—nutrition, schooling, health, complexity, expectations—and minds change too. That should make us cautious about grand claims of fixed limits, but also cautious about easy optimism. Because as we said earlier, gains can stall. Environments can help minds grow, and they can also stop helping.
That, to me, is the real lesson. The Flynn effect is not just a curiosity about IQ tests. It is a reminder that human thinking is historical. Our minds are shaped by the worlds we build around them. Which raises an uncomfortable next question: what, exactly, is our current world training us to be good at?
.png)
