Generation Z may become the first generation in many decades to score lower on certain cognitive ability tests than its predecessors. This is indicated by studies in the U.S. and Europe, international educational tests, and data on the so-called Flynn effect reversal.
The debate has reignited following the publication of the article “Is Gen Z the First Generation Less Intelligent Than Their Parents?” by the popular science website RathBiotaClan. The author, Shibasis Rath, compiled research showing a decline in scores on specific measures—mathematical and verbal reasoning, attention span, memory, and problem-solving.
One of the catalysts for the article’s publication was a hearing held by the U.S. Senate Committee on Commerce, Science, and Transportation on January 15, 2026. Neurobiologist and education expert Jared Kuni Horvath presented data to senators showing that over the past two decades, a number of developed countries have seen stagnation or a decline in literacy, math skills, attention, and the ability to engage in complex reasoning.
For most of the 20th century, IQ test scores consistently improved from one generation to the next. This phenomenon became known as the Flynn effect. The increase was attributed to improvements in nutrition and healthcare, expanded access to education, smaller family sizes, and the growing complexity of the environments in which people grow up and work.
However, in some developed countries, this trend first slowed down and then reversed.
One of the most well-known studies was conducted by Norwegian scientists Bernt Bratsberg and Ole Røgeberg. In a paper published in the *Proceedings of the National Academy of Sciences* in 2018, they used data on military conscription and administrative records spanning several decades. By comparing brothers within the same families, the authors concluded that both the initial rise and the subsequent decline in outcomes were largely linked to environmental changes rather than genetic changes in the population.
In July 2026, Bratsberg and his colleagues published a new study in PNAS based on data from 579,379 Norwegian men born between 1967 and 1991. It showed that the expansion of secondary education may have masked a more general downward trend in cognitive test scores for a time. Once the growth in educational attainment slowed, the decline became more pronounced across various social groups.
American data yield a similar, albeit more complex, result.
Researchers at Northwestern University analyzed the results of 394,378 Americans who took online tests as part of the SAPA project between 2006 and 2018. Scores on verbal reasoning, matrix tasks, and number sequences declined over time. At the same time, spatial reasoning, on the contrary, improved.
That is why one of the study’s authors, Elizabeth Dvorak, cautioned against claiming that Americans are simply “getting dumber.” According to her, changes in test scores do not necessarily indicate a corresponding change in overall intellectual ability. It may reflect changes in skills, motivation, education, or the ability to perform a specific type of test task.
This is a fundamentally important caveat for evaluating Generation Z as well. Today’s young people may perform worse on some types of cognitive tasks and, at the same time, better on others—for example, when processing visual information or using digital tools.
Another major source of data is the OECD’s PISA program, which assesses the knowledge and ability to apply that knowledge among 15-year-old students.
Between PISA 2018 and PISA 2022, the average score in mathematics across OECD countries fell by a record 15 points, and in reading by 10 points. At the same time, science scores remained largely unchanged. The OECD emphasizes that the decline in reading scores began even before the COVID-19 pandemic, so it is impossible to attribute the entire trend solely to school closures in 2020–2021.
At the same time, the OECD itself does not attribute the problem solely to smartphones or computers. The organization notes that the reasons for the decline in educational outcomes are numerous, and that the competent use of digital technologies by teachers can, on the contrary, contribute to the development of digital literacy and the ability to critically evaluate information.
This is where the most controversial part of the discussion begins.
Horvat attributes the decline in certain indicators to the rapid proliferation of laptops, tablets, and other digital devices in the education sector. In his written testimony to the Senate, he argues that excessive screen-based learning can impair concentration, the depth of information processing, and memory retention—especially if digital technology simply replaces traditional instruction without offering improved pedagogical methods.
However, based on the available data, it cannot be concluded that smartphones are the sole or even the proven primary cause of the Flynn effect reversal.
The results may be influenced simultaneously by the quality of school education, changes in curricula, the social environment, the pandemic, sleep, physical activity, nutrition, patterns of information consumption, and shifts in motivation to take standardized tests.
Therefore, it is more accurate to speak not of a proven “decline in the intelligence of Generation Z,” but rather of an observed decline in performance across a range of cognitive and educational indicators in several developed countries.
Even this more cautious phrasing has serious economic implications.
Skills such as reading complex texts, mathematical analysis, concentration, and solving non-standard problems directly impact human capital. As artificial intelligence becomes more widespread, the importance of these skills may not diminish but rather increase: people need not only to receive a ready-made answer from a machine but also to evaluate its correctness, spot errors, and formulate complex tasks on their own.
If the decline in certain cognitive indicators is indeed sustained, the consequences may manifest in labor productivity, the quality of vocational education, the ability to master complex professions, and the economy’s innovative potential.
But there is also the opposite possibility: the digital environment does not so much reduce intelligence as it changes the structure of skills, making certain forms of information processing less in demand while developing others. This is precisely the question that remains open today.
The RathBiotaClan article is useful primarily as a popular science overview that brings together in one place several studies and public presentations by Jared Kuni Horvath. RathBiotaClan positions itself as an Indian platform for science education and media, registered in India as a micro services enterprise under the MSME/Udyam system. The resource’s founder, Shibasis Rath, specializes in popularizing research in biology and related sciences.
RathBiotaClan is not a peer-reviewed scientific journal. Therefore, it is more appropriate to use this publication as a starting point rather than as standalone evidence that Generation Z has become less intelligent.
The most important primary sources for this topic are studies in PNAS and the journal Intelligence, OECD PISA statistics, and materials from U.S. Senate hearings. These sources make it possible to verify the main claims of the popular publication.
DIGITALIZATION, EDUCATION, Generation Z, intelligence, RESEARCH
