Centred Counselling & Mediation
For much of the twentieth century, something remarkable happened.
Generation after generation appeared to perform better on intelligence tests than the generation before it.
The phenomenon became known as the Flynn effect, named after political scientist and intelligence researcher James Flynn, who documented substantial increases in IQ test performance across generations.
For decades, the pattern was surprisingly consistent.
But then something changed.
In a growing number of countries, researchers began finding that the upward trend had slowed, stopped or, in some cases, reversed.
This has led to a provocative question:
Are younger generations actually becoming less intelligent?
The answer is considerably more complicated than the headlines suggest.
The Flynn effect refers to the long-term increase in average scores on intelligence tests across successive generations.
During much of the twentieth century, researchers commonly observed increases of roughly 2–4 IQ points per decade, although the size of the effect varied considerably between countries, age groups and different types of cognitive tests.
This is an extraordinary finding.
IQ tests are designed so that the average score is approximately 100 within the population used to establish the norms. If an intelligence test created decades ago were administered to a later generation using the original scoring system, the newer generation would often obtain substantially higher scores.
It was as though something about the cognitive environment in which children were growing up was producing progressively better performance.
But this immediately raises another question:
What exactly was getting better?
This is where the Flynn effect becomes fascinating.
An increase in IQ scores does not necessarily mean that people's underlying intelligence increased by the same amount.
Intelligence tests measure several different cognitive abilities. Some tests involve:
And these abilities have not necessarily changed at the same rate.
A major meta-analysis covering 271 independent samples, almost four million participants and 31 countries found that twentieth-century gains varied substantially according to the cognitive domain being measured. Fluid intelligence showed particularly strong gains, while crystallised intelligence showed smaller gains. The researchers also found that the size of the Flynn effect had been decreasing in more recent decades.
This distinction is crucial. The Flynn effect is therefore better understood as:
“People have become better at performing certain types of cognitive tasks over generations.”
That is not quite the same statement as:
“Human beings have become inherently more intelligent.”
There is no single universally accepted explanation. The most likely explanation is that multiple environmental changes interacted with one another.
Over the twentieth century, many children experienced:
The world itself became cognitively different.
Consider what a child growing up in 1900 might have encountered compared with a child growing up in 2000. The modern child is surrounded by maps, diagrams, computers, screens, numbers, symbols, timetables, interfaces, abstract categories and constantly changing information.
The brain develops within the environment it experiences.
This is one reason researchers have generally focused heavily on environmental explanations rather than assuming that generational changes in IQ must be genetic.
Indeed, research using Norwegian family data found that the rise, turning point and subsequent decline in cognitive test scores could be recovered within families, providing evidence against explanations based primarily on changes in the genetic composition of the population.
The remarkable part of the story is what happened next.
In several countries, the upward trend began to flatten. And in some datasets, it began moving backwards.
This is sometimes called the negative Flynn effect or reverse Flynn effect.
Research has reported reversals in several countries and populations, although the size, timing and even existence of the reversal varies considerably depending on the country and the test being used.
That last qualification is extremely important.
There isn't one worldwide moment when “human intelligence started declining.” Nor can we simply say:
“The newest generation has a lower IQ than the previous generation.”
The evidence is more complicated.
Norway provides an unusually valuable natural experiment.
For decades, the Norwegian Armed Forces administered cognitive tests to large numbers of young men entering military service. The testing included measures of:
The data showed the familiar pattern: scores increased for several decades, reached a peak, and then declined.
But a fascinating 2025 reanalysis asked a more sophisticated question:
Was general intelligence actually declining — or were particular abilities responsible for the decline in the overall score?
The answer was surprising.
The researchers found that the earlier increase was driven predominantly by improvement in figure-matrix reasoning, a form of fluid reasoning. The subsequent decline, however, was driven primarily by decreases in word comprehension and numerical reasoning.
Those declines were not adequately explained by a corresponding decline in the underlying general mental ability factor.
In other words: The reversal may not mean that young people are simply becoming less intelligent. It may mean that they are becoming less practiced at — or less exposed to — particular forms of thinking that older generations were routinely required to practise.
That is a very different conclusion.
One of the clearest areas of decline in the Norwegian data was numerical reasoning.
This is not necessarily the same thing as “being bad at mathematics”. The test involved forms of mathematical reasoning and calculation that previous generations were more routinely expected to perform manually.
Think about what has happened to everyday arithmetic. Previous generations routinely had to:
Today, many of these tasks can be delegated to calculators, smartphones, spreadsheets and software. That doesn't necessarily make a person less capable of solving a complex mathematical problem. But it can mean that mental arithmetic and manual calculation receive less practice.
The 2025 Norwegian study specifically discusses the possibility that educational changes and the widespread availability of digital technology reduced the emphasis placed on manual arithmetic.
This is an important distinction: A calculator can replace arithmetic execution without replacing mathematical reasoning.
Another area showing decline in the Norwegian data was word comprehension.
This raises an intriguing possibility. Previous generations encountered language differently. They were more likely to spend substantial amounts of time reading:
Modern communication increasingly takes place through:
This does not mean young people cannot understand language. Nor does it mean that vocabulary is universally declining.
But it raises an important question: What happens to certain forms of verbal knowledge when people spend less time encountering complex written language?
The Norwegian researchers point out another complication: some older test items may contain vocabulary or concepts that were more familiar to previous generations because they were embedded in their cultural environment.
So a lower score does not necessarily mean: “Young people have lost the ability to think.” It may partly mean: “Young people are less familiar with the particular language and knowledge that this test rewards.”
Research from the United States provides another interesting piece of the puzzle.
A large study examining more than 394,000 adults between 2006 and 2018 found evidence consistent with a reverse Flynn effect in composite cognitive scores. Declines were observed particularly in:
But there was an important exception: Three-dimensional rotation performance generally increased.
This is extremely important. If younger people were simply becoming globally less intelligent, we would expect cognitive abilities to move broadly downward.
Instead, researchers found something much more interesting: some abilities went down while another went up. That looks less like a collapse in intelligence and more like a change in the cognitive profile of the population.
The finding concerning three-dimensional rotation is particularly intriguing.
Three-dimensional rotation tasks involve mentally manipulating objects in space. They require someone to imagine: “If I rotate this object, what will it look like from another angle?”
Modern environments may provide considerably more exposure to spatial information than previous generations experienced. Video games, 3D environments, digital modelling, navigation systems and interactive graphics all require different forms of spatial processing.
The U.S. study found that three-dimensional rotation scores generally increased while several other cognitive domains declined.
This is a powerful reminder that technology may not simply be reducing cognitive ability. It may be redistributing which abilities receive practice.
Perhaps the most useful way to understand the reversal is to stop thinking about intelligence as a single muscle.
Imagine instead that cognition is an ecosystem. Different activities exercise different parts of it:
The environment determines which skills receive repeated practice. And repeated practice matters.
Not necessarily.
This is where popular discussions about the Flynn effect often go too far. The research does not establish that smartphones, social media or the internet have caused the reversal.
There are correlations and plausible mechanisms, but identifying the exact causes is considerably more difficult. There are many potential influences:
And different countries have experienced different patterns.
This may be one of the most overlooked parts of the discussion.
An intelligence test is not completely independent of culture.
Imagine giving someone today a test containing dozens of questions about technologies, occupations and cultural references from 1950. A lower score could reflect unfamiliarity with the cultural material rather than lower reasoning ability.
This issue has been demonstrated in research examining alleged IQ declines in France. Researchers found that apparent declines were concentrated in subtests heavily influenced by culturally dependent knowledge. They concluded that some apparent declines could be explained by changes in education, culture and test characteristics rather than an actual reduction in intelligence.
This is why researchers have to be extremely careful when comparing IQ scores across generations.
A 2024 study comparing German-speaking samples from 2005 to 2024 found another interesting pattern.
The researchers observed changes in the relationship between different cognitive abilities — what psychologists call the positive manifold. Interestingly, they found that some individual cognitive test scores increased even while the overall relationship among abilities weakened.
Again, this complicates the idea of: “IQ is going down.”
Human cognitive ability does not appear to be moving in one simple direction. Different components may be changing differently.
The most defensible conclusion is not: “Young people are becoming less intelligent.”
It is: The cognitive profile of newer generations may be changing.
Some abilities appear to have weakened in certain populations. These include evidence of reductions in:
At the same time, some abilities have remained stable or improved, including evidence for stronger performance on certain three-dimensional spatial tasks.
And importantly, the latest Norwegian research suggests that some of the apparent decline in overall IQ scores may reflect specific skills rather than a decline in underlying general mental ability.
Perhaps the biggest lesson from the Flynn effect isn't that technology is making people less intelligent.
It is that the brain adapts to the environment in which it operates:
This does not mean the ability disappears. It means that the cognitive environment changes.
And this may be particularly important during childhood and adolescence, when cognitive abilities are still developing.
The answer isn't to reject technology. It is to make sure technology supplements cognitive development rather than replacing it.
Children still benefit from opportunities to:
In other words: Give the brain something to do.
The Flynn effect initially seemed to tell us that every generation was becoming smarter. The reversal appears to tell us something different.
Human cognition is remarkably responsive to its environment. Change the environment and the cognitive profile can change with it.
The twentieth century may have rewarded certain forms of abstract reasoning, formal education and problem-solving. The twenty-first century may be rewarding a somewhat different combination of skills — rapid information processing, technological fluency, visual navigation and digital problem-solving.
The danger is not necessarily that technology makes us stupid. The danger is that we may stop practising skills simply because technology makes them unnecessary in everyday life.
And that distinction matters. Because intelligence isn't simply something we possess: It is also something we exercise.
There is a tendency whenever generations change to say:
“Young people aren't what they used to be.”
Psychology should make us more cautious than that. The evidence does not justify dismissing an entire generation as less intelligent. Instead, it gives us a much more interesting possibility.
Perhaps the question isn't: “Are they smarter or dumber?”
Perhaps the question is: “What kind of thinking is their environment teaching them to become good at?”
That is a much more useful question. Because if certain skills are weakening, the answer isn't to panic about an allegedly “dumber generation”. It is to identify the skills that are receiving less practice — and deliberately give them space again.
The goal isn't to raise an IQ score. The goal is to raise a generation that can still think deeply when there isn't a shortcut.
The “reverse Flynn effect” is a genuine research finding in several populations, but it should not be presented as proof that today's children or young adults are universally less intelligent than previous generations. Results differ by country, age group, test and cognitive domain. Some apparent declines may also reflect changes in education, culture and test characteristics.
The strongest recent evidence suggests that specific cognitive skills can move in different directions rather than general intelligence simply rising or falling as a whole.