Wolfgang Amadeus Mozart is arguably the most celebrated child prodigy in history, composing his first pieces of music aged five, his first symphony at eight and his first opera at 11. After a study in 1993 found that listening to Mozart could improve spatial IQ – prompting headlines such as “Mozart makes your brain hum” – he became a symbol for intelligence and brain training.

The study was no doubt interesting. The scientists found that performance on spatial ability tests was improved when their study participants had listened to a Mozart sonata, compared with a relaxation tape or silence. The increase in performance translated to an astounding difference of up to nine spatial IQ points.

Although the effects were temporary, lasting less than 15 minutes, the idea exploded in popular culture. The “Mozart Effect” ignited a lucrative empire of parenting books, self-help manuals and CDs promising to harness the power of Mozart’s music to foster children’s cognitive development. That was despite the fact that the study had been carried out in adults and the evidence for the effect was later overtuned.

The hard fall for the Mozart Effect ultimately highlights the value that society places on intelligence as measured by cognitive tests (like the IQ test). The global market for cognitive assessment and training was valued at about $6.87 billion in 2024 (£5.18bn) but is projected to rise to $35.30 billion by 2032.

Mozart went on to compose over 600 outstanding works in his brief lifetime. But can we reliably predict future success from a child’s performance? Today, IQ tests are often used to spot early academic talent. But are they a good measure? A growing number of scientific studies suggest that IQ measured in childhood might tell us less than we think. Scientists are discovering that children’s IQ scores aren’t as stable as adults’ – they fluctuate substantially.

So why are schools using cognitive assessments? And what other factors can help predict children’s future success?

The rise of cognitive tests to identify potential

Fostering talent is central to human progress. Exceptionally talented individuals drive scientific and cultural innovation and push the boundaries of human knowledge. For over a century, scientists have therefore sought to understand and measure intelligence. This has been partly driven by countries gradually shifting away from mass production and towards becoming knowledge economies.

The Insights section is committed to high-quality longform journalism. Our editors work with academics from many different backgrounds who are tackling a wide range of societal and scientific challenges.

One of the largest and longest running studies of giftedness, the Study of Mathematically Precocious Youth, has followed the lives of intellectually gifted people for over half a century. Over 1,600 talented 13-year-olds were invited to take part in the study if they had scored in the top 1% of ability on a standardised test, the SAT, widely used for US college admission. And indeed, four decades later, many of these young talents had achieved outstanding accomplishments. Some 4.1% had achieved tenure at a major university and 2.3% were top executives at Fortune 500 companies. They had published 85 books and secured 681 patents.

However, it is worth noting that these children were fairly old, already teenagers – and at the absolute top end of achievement. Cognitive tests, however, are taken by a much wider range of children today. Since the 1980s, cognitive ability tests have gradually replaced traditional academic subject exams as school entrance screeners. This was motivated by the idea that a cognitive test could be a more objective assessment of aptitude and potential than a child’s knowledge of the curriculum. Performance on cognitive tests is viewed by many as independent of external influences, such as a more resourceful school or a nurturing home environment.

Schools worldwide, from the US and the UK to Singapore and Vietnam now use standardised tests of cognitive abilities to select students at intake. Admission to many prestigious independent and selective high schools in the UK is often at least partly based on a cognitive ability test, such as the infamous CAT4, that hopeful ten-year-olds sit in the autumn term of their last year of primary school. The CAT4 test is also used in many state secondary schools to help determine sets, predict grades and allocate support and provisions.

The CAT4 takes around 2.5 hours to complete and is divided into four sections. There is verbal reasoning (thinking with words), non-verbal reasoning (thinking with shapes), quantitative reasoning (thinking with numbers) and spatial ability (thinking with shapes and space). Children who score exactly as expected for their age group would be given a score of 100. Scoring between 89 and 111 is considered to reflect “average” performance, while scores of 112 and above or 88 and below indicate above and below average performance, respectively.

Child IQ fluctuates

We know that the human brain is plastic, or changeable, particularly in childhood. It is the only organ in our body that isn’t fully developed when we are born. A newborn’s brain is about a quarter of the size of an adult brain, doubling in the first year of life. By age seven, it reaches 90% of its adult size. Beyond physical growth, our brains refine and consolidate the network of connections between neurons during this time.

Refining and whittling these connections is key to supporting cognitive and behavioural developmental milestones. Recent research shows that it’s possible to identify key “eras” of brain structural change over the life course. The first milestone – the transition from childhood to adolescence – happens at around age nine. From a brain perspective, adolescence lasts for a little over two decades and is defined by greater efficiency of connections across regions. This coincides with a steady increase in cognitive functions, including vocabulary, complex reasoning and learning.

We’ve known for some time that there is a link between intelligence, as measured by cognitive tests, and school achievement. Research from 2015 that combined data from over 100,000 students across 240 different studies did find a substantial association between intelligence and school grades. However, the magnitude of the link differed depending on children’s age. Intelligence was a much better predictor of school performance in secondary school than it was in primary school. This suggests that cognitive abilities might not be stable during the first decades of life, but vary significantly.

A 2024 analysis that combined data from 205 different studies including over 85,000 participants across 29 countries supports this view. The researchers set out to investigate how stable cognitive abilities are (whether they fluctuate) across the human lifespan and whether stability changes with age.

They discovered that the stability of cognitive abilities increased exponentially with age – and was low in the first decade of life. This means that each child’s positioning compared to their peers changes significantly in childhood. So a child’s IQ score might indeed change substantially during this time. The stability, however, increased throughout childhood and adolescence, plateauing around age 20 and remaining high throughout adulthood and old age.

But even when IQ starts stabilising, in adolescence, it can still fluctuate by up to 20 points. Somebody increasing their IQ score from 100 to 120 would move from the 50th percentile to about the 91st percentile – a 41% improvement. Indeed, one study, albeit with a small sample of students, could link such fluctuations to physical changes in the brain over time.

This means that it can be tricky to infer long-term consequences, such as later grades, from cognitive tests. Basing school intake, or more broadly selection into educational programmes on a single, unstable metric is likely to lead to systematic errors and unreliable decisions.

Worryingly, it may also result in attempts to manipulate the metric, potentially perpetuating systemic inequalities. This may be true of other tests too, but IQ tests are often seen as an exception. But research shows that you can actually train yourself to boost your IQ test score by roughly eight IQ points, for example by retaking the test. Parents with a lot of resources might be better placed to help prepare their children.

The myth of the child prodigy

Recent research has backed all this up by questioning the widely accepted myth of the child prodigy as someone destined for greatness, like Mozart. One 2025 study, which combined data from over 34,000 elite performers, from Nobel laureates and chess players to music composers and athletes, found that exceptional performance in childhood was a limited predictor of elite performance in adulthood.

In fact, about 90% of those who achieved elite performance in youth did not achieve equivalent adult status. Similarly, 90% of top performing secondary school students were no longer top performers at university. And even more strikingly, several Nobel laureates and elite athletes actually had lower childhood performance than their peers.

The routes leading children and adults, respectively, to world-class performance also differed. Exceptional talent early in development was associated with intensive, discipline-specific progress at a young age. But adult world-class performance was more often achieved through extensive multi-faceted practice and gradual advancements.

This means that educational and talent programmes that prioritise early identification of intelligence may overlook a large proportion of future world-class innovators.

Environmental exposure

The idea behind identifying talent as early as possible so that it can be nurtured is founded on the belief that exposure to an enriched environment can impact ability and vice versa. Half a century of scientific discovery supports this proposition. Perhaps the most famous example is a study published in 1979 by paediatrician Herbert Needleman and his colleagues. This study provided the first robust evidence that exposure to the metal lead, even at levels previously considered negligible, could significantly impair a child’s cognitive performance.

By comparing children with high and low lead levels in baby teeth, while controlling for potentially confounding factors – such as the mother’s IQ and socio-economic status – the study showed that children with higher lead levels scored roughly four points lower on IQ tests. The evidence presented influenced major public health policies, including the removal of lead from gasoline and interior paint in the US.

A large number of other environmental exposures have been positively linked with cognitive development, from walking in nature to exercise and nutrition, albeit with mixed results. However, arguably the most successful environmental intervention to improve cognitive ability is administered every year to more than 85% of children worldwide: education.

By combining data collected across multiple studies from over 600,000 individuals, researchers found that education has a direct effect on the development of cognitive abilities. The study found that each year of education results in a gain of about one to five IQ points. These effects were remarkably robust, appearing across different cognitive domains and persisting throughout the lifespan. In fact, significant benefits were still measurable into people’s 80s and 90s. While a few IQ points per year may seem small, their cumulative impact at a societal level has been shown to be of great consequence.

Environmental factors that shift population IQ even modestly — like lead exposure, nutrition or education — carry enormous economic consequences. Economists have calculated that each gained IQ point is associated with roughly a 2% increase in lifetime earnings.

In the year 2000, a single IQ point gained or lost across the US population translated to between $110 and $319 billion in aggregate economic output. More recent analysis of the global economic impact of lead exposure on childhood IQ estimated the total cost of IQ loss at US$1.4 trillion globally in 2019, mainly affecting low and middle-income countries.

The role of parents

From the moment a child is born, parents invest vast amounts of energy, time and resources to promote their children’s physical and cognitive development. Not all parenting practices are supported by scientific evidence, nor is the Mozart Effect the sole parenting myth that has been busted. However, research has shown that parenting can nevertheless have profound effects on children’s early cognitive development.

Studies have found that the environment that parents provide for their children by reading to them, engaging them in stimulating activities and conversation, and maintaining a warm and organised household, has a significant positive effect on early cognitive development. This is particularly the case for the first five years of life. What makes early investment especially powerful seems to be that the benefits compound. Fostering a child’s early cognitive competence makes it easier for children to acquire new skills down the line.

However, the pathways to parental investment are complex. Reflecting on my own childhood illustrates this point. I was born in the mid-80s to parents in their early twenties. At the time, my mother was in medical school and my father designed and produced bespoke furniture. As a child, I had several ear infections which meant that I had to have regular checks with a specialist. One warm, sunny morning in early April, my mum and I set off for my otolaryngologist appointment, just the two of us. As the eldest of four children, this was a rare and special occasion.

After my check-up, we took a tram to Milan’s State University, where we attended a conference on HIV infections in vulnerable populations – the topic of my mother’s thesis. I remember sitting in the beautiful auditorium, admiring the frescos on the ceiling, and slowly adjusting a pair of disposable headphones to listen to the real-time translation of the talks. The panel of female scientists discussed the topic so eloquently and clearly that even a ten-year-old girl could grasp their main message.

I was hooked. It must be the best job in the world, I thought. It was only a quiet thought then, one that I never had the courage to privately contemplate or publicly share. That came much later, when I found the confidence to admit that a career in scientific research was for me. But this specific episode in my childhood was not an isolated peak. It was the pinnacle of many simpler, everyday moments when my parents invested time and effort to provide us with a nurturing and stimulating environment.

However, seeing these as merely environmental exposures would only provide part of the picture. Perhaps, the science-enriched environment that my mother created for us depended, at least in part, on her own, partly genetically driven, scientific aptitude.

The nature of nurture

Scientists have named this amalgamation of nature and nurture gene-environment correlation, or more intuitively, the nature of nurture. Parents who provide their children with intellectually stimulating environments may also pass on a greater disposition to doing well in school or performing well in cognitive tasks. Research has shown that accounting for genetic effects shared between mothers and children resulted in a reduction in the effect of parenting on educational attainment.

However, cognitively stimulating parenting remained a significant predictor of children’s educational outcomes beyond direct genetic inheritance and socio-economic status. It ultimately contributes to channelling children’s dispositions and translating them into academic outcomes.

The important role of the family environment is also highlighted by decades of twin studies. These show that environmental exposures that are common to siblings, including growing up in a more resourceful home, school and neighbourhood, explain about a third of differences between children in cognitive ability.

Randomised control trials have demonstrated that early interventions are likely to lead to the greatest returns. Investing in children early — through parenting, stimulating environments and good nutrition — pays back far more than trying to catch up later. Every year of delay makes it harder to close the gap.

Interventions created to bridge this gap in groups of disadvantaged children through high-quality preschool education, such as the Perry Preschool Project, can lead to meaningful gains in cognitive performance. Interestingly, while the benefits on children’s cognitive performance faded over time, their long-term educational, economic and social benefits were remarkably far-reaching. So a high quality school education could indeed lead to better job prospects and higher salaries, regardless of IQ.

It follows that boosting cognitive ability may not be the only way to lasting educational, economic and health benefits. Non-cognitive skills — such as motivation, curiosity, self-regulation and social skills — are equally important.

What IQ tests fail to capture

Cognitive tests have never been viewed as instruments to capture the entire set of skills necessary for succeeding in school and life. In 1916, Alfred Binet and Theodore Simon, the inventors of the first IQ test, wrote that things other than intelligence also mattered to academic success, arguing “one must have qualities which depend especially on attention, will and character”.

Decades of research have shown that children who are emotionally stable, motivated and capable of regulating their attention and impulses do better at school, regardless of their level of cognitive ability. These important characteristics have been broadly described as “non-cognitive skills”.

Recent research by my own team shows that the importance of non-cognitive skills for learning also changes over the school years. We analysed data collected from over 10,000 children born in England and Wales who were followed throughout compulsory education, from age seven to 16. Non-cognitive skills not only predicted academic achievement at every developmental stage, but their role increased as the children got older. Still, at all ages, skills such as curiosity, creativity, motivation and self-efficacy predicted success in school in addition to what was predicted by cognitive abilities.

Similar to cognitive ability and learning, differences in non-cognitive skills are a complex product of nature and nurture. Partly based on their genetic dispositions, children encounter and select environmental experiences that contribute to the development of their motivation and curiosity. This in turn leads to differences in school achievement.

Ultimately, cognitive tests are thought to offer an objective measure of a child’s natural ability, one that is largely unaffected by upbringing or circumstances. But research shows that a range of factors, from environmental exposures to toxic agents, nutrition, differences in parenting and educational interventions, can change cognitive performance, particularly as the brain develops.

During childhood, when the brain is rapidly growing, cognitive test scores can fluctuate considerably from one year to the next. This means that a single test taken on a single day in primary school is not a reliable enough indicator for decisions as consequential as which school a child attends or which academic track they are placed on. These are decisions that can shape the entire course of their education.

Even later on, cognitive tests only capture part of what it takes to do well in school and in life. Curiosity, motivation and the belief that you can improve with effort are crucial to educational success, yet most education systems pay them little attention. Rather than treating a test score as a fixed marker of a child’s future, mounting evidence invites us to treat it as one factor among many. The best approach would be to invest in all children’s cognitive and non-cognitive development alike.

So don’t read too much into Mozart’s journey. He may have been a child prodigy destined for greatness, but chances are he was an exception rather than the rule.

For you: more from our Insights series:

To hear about new Insights articles, join the hundreds of thousands of people who value The Conversation’s evidence-based news. Subscribe to our newsletter.