Growing evidence against the idea that boys are inherently better at mathematics

Two decades prior, cognitive scientist Elizabeth Spelke adopted a firm stance in a continuing public discourse.

“There are no distinctions in overall innate talent for science and mathematics between women and men,” the researcher proclaimed.

A recent article in the journal Nature, authored by Spelke and a cohort of European scholars, presents what she describes as “an even more compelling foundation for that assertion.” 

A French government testing project initiated in 2018 provided insights into the mathematical abilities of over 2.5 million schoolchildren over five years. Analyses revealed almost no gender variations at the onset of first grade, when pupils commence formal math education. However, a gap favoring boys emerged after merely four months — and continued to expand through successive grades.

The outcomes corroborate earlier research results derived from far smaller sample sizes in the U.S. “The primary conclusion is that the gender gap arises when structured instruction in mathematics begins,” summarized Spelke, the Marshall L. Berkman Professor of Psychology.

Back in 2005, her viewpoint was shaped by years of examining sensitivity to numbers and geometry in the youngest members of society. 

“My assertion was, ‘Alright, if there were indeed biological differences, perhaps we would observe them during infancy,’” recalled Spelke, who elaborated her findings in a pivotal review for the journal American Psychologist that year. 

“We consistently documented the gender composition of our studies, along with the relative performances of boys and girls,” Spelke continued. “Yet, we were never discovering any distinctions favoring either gender.”

The prospect persisted that disparities in capability or even enthusiasm could manifest later in life.

“The absence of distinctions in infants might suggest that the abilities displaying gender effects might actually appear during preschool,” Spelke noted.

In recent years, the psychologist has been employing her research on initial counting and numeral-recognition abilities through educational interventions, all scrutinized and honed via randomized controlled trials.

As one of the globe’s most influential researchers on early education, Spelke recently collaborated with Esther Duflo, an MIT economics professor and Nobel laureate, to advise the Delhi office of the nonprofit Abdul Latif Jameel Poverty Action Lab (J-PAL). The organization is collaborating with the administrations of four distinct Indian states to create and assess math curricula for preschoolers, kindergartners, and first-graders. 

Alongside her long-time collaborator, cognitive neuroscientist Stanislas Dehaene, Spelke also acts as an adviser on the Scientific Council of the French Ministry of Education. The nationwide EvalAide language and math assessment was rolled out with the council’s assistance in 2018. The initiative’s aim, Spelke detailed, is to establish a baseline measure of every French child’s comprehension of fundamental numeracy and literacy skills while aiding the ministry in its commitment to an evidence-based education for all French pupils.

Spelke co-authored the Nature article with Dehaene and eight other researchers, all operating in France. The study specifically analyzed four consecutive cohorts of predominantly 5- and 6-year-olds entering school between 2018 and 2021. 

As seen in various nations, French girls demonstrated slightly superior results compared to French boys in language as they commenced first grade in the autumn. However, the gender gap was virtually inconsequential regarding mathematics. 

“That definitely relates to the earlier question of whether there’s a biological foundation for these disparities,” Spelke contended.

French first-graders were subsequently reassessed after four months in school, at which point a minor yet significant math gap favoring boys had emerged. The effect quadrupled by the onset of second grade, when schoolchildren were tested once more.

“It was even larger in fourth grade,” stated Spelke, noting that French children are now assessed at the commencement of even-numbered grades. “And in sixth grade, it was even greater.”

For comparison, EvalAide results indicate the literacy gender gap diminished by the first year’s four-month mark and underwent far less change as students advanced to higher grade levels.

What could cause a widening gender gap in mathematics specifically as students spent more time in school? According to Spelke, the paper provides “only negative responses” regarding hypotheses about innate gender differences and societal bias.  

“If there truly was a widespread societal bias, and if parents were influenced by it,” she asserted, “we would anticipate boys to be more inclined towards spatial and numerical tasks upon arriving at school.”

Delving deeper into the data revealed further results that intrigued the researchers. For instance, Spelke’s collaborators could distinguish the findings by month of birth, with the oldest French first-graders celebrating their seventh birthdays in January — almost a year ahead of their youngest peers. The math gap correlated not with age but with the amount of time spent in school. 

Another noteworthy finding involved the COVID-19 pandemic, which eliminated the final 2.5 months of first grade for children who began in fall 2019. “With decreased time in school, the extent of the gender gap grew less than it did in the other years free of long school closures,” Spelke remarked.

The 2019 cohort yielded one additional compelling result. Earlier that year, French schoolchildren had ranked lowest among 23 European nations in the quadrennial Trends in International Mathematics and Science Study. This triggered a national discussion: How could France, the birthplace of the eminent René Descartes, be lagging behind its counterparts in mathematics?

In May 2019, the French Ministry of Education, backed by its Scientific Council, called for the integration of more math curriculum during kindergarten. For the first time, a very slight gender math gap was observed that autumn for those entering first grade. It hadn’t existed in 2018 but remained noticeable in results from the 2020 and 2021 cohorts.

The overall findings, the most definitive to date, indicate it’s time to set aside explanations rooted in biology or bias. Instead, it appears there’s something about early math instruction that gives rise to gender differences. 

“We still don’t know precisely what that is,” said Spelke, who plans to spend a significant portion of her 2025-26 sabbatical year in France. “But now we have the opportunity to uncover that through randomized evaluations of curriculum modifications.”