Mensa’s idea of ‘intelligence’ is plain stupid

Mensa turns 80 this year. To celebrate, British Mensa has launched a six-part Channel 4 television series called Secret Genius, hosted by Alan Carr and Susie Dent, in which contestants from across the UK tackle intelligence tests and games based on the challenges Mensa uses to measure IQ. There was a two-day event in Oxford at the weekend, where the organisation was born. Additionally, a brand new online IQ test is being rolled out to mark the occasion.

None of this changes the fact that Mensa is built on a foundation that stopped being defensible decades ago. The organisation exists to sort human beings by a single metric – timed pattern recognition – and then convert that statistical sorting into a social hierarchy. Time and time again, such testing has proved to have very little to do with actual intelligence.

Undoubtedly, IQ tests measure something tangible. They capture processing speed, working memory and the ability to identify abstract patterns quickly. These are useful cognitive traits. But calling them ‘intelligence’ is like calling grip strength ‘fitness’. You have measured one narrow quality and made a judgement on the whole athlete.

What IQ tests fail to measure is one’s capacity to construct original frameworks, to identify hidden assumptions, to reason under ambiguity, or to adapt when your environment negates the rules of the game that you were trained on. They do not measure strategic depth, nor the quality of your questions. The tests only concern themselves with the speed of your answers.

Mensa’s entry requirement is a score at or above the 98th percentile on a standardised intelligence test. This sounds impressive until you examine what it actually selects for. Nobel Prize-winning theoretical physicist Richard Feynman, who achieved the highest score in the nation on the notoriously difficult Putnam mathematics exam, scored 125 on his school IQ test. He would not have qualified for Mensa. James Watson, co-discoverer of the structure of DNA, scored 124. William Shockley, whose Nobel prize-winning work on semiconductors launched the modern electronics industry, tested at 129.

Shockley, along with Luis Alvarez, another Nobel laureate in physics, did not have sufficient IQs in their youth to even qualify for Lewis Terman’s famous longitudinal study of ‘gifted’ children. Terman screened 168,000 children in the 1920s, selecting those above a strict IQ threshold for lifelong tracking. Though they didn’t make the cut, Shockley and Alvarez went on to receive the most prestigious scientific award in the world. None of the children Terman selected for his study came near that level of achievement in adulthood. The metric that was supposed to identify genius had, in fact, filtered it out.

This is not a fringe criticism. The literature on IQ and achievement has been clear on this issue for decades. Above a moderate threshold, IQ ceases to predict who will produce original work. That’s because originality relies on persistence, tolerance for ambiguity, willingness to challenge premises, and what psychologists sometimes call ‘openness to experience’. These are traits that no timed multiple-choice exam has ever captured, because to do so would be impossible.

Mensa should know this better than most. The Stanford-Binet test, on which Mensa’s original selection criteria were based, was never designed to identify brilliance. Alfred Binet created it in 1905 to identify children in French schools who needed additional support. It was a diagnostic tool for struggling students, not a crown for gifted ones. That Mensa repurposed a remedial-screening instrument as the entrance exam for an elite intellectual society isn’t simply ironic – it is structurally absurd.

And yet the brand persists. Visit Mensa’s website and you will find smiling faces at social gatherings, invitations to local meetups, and language engineered to comfort rather than challenge. The free Mensa IQ Challenge is a 35-question online quiz to be completed in 25 minutes. At the bottom, it reads: ‘If you’re pleased with your score, you might want to consider taking a properly administered and supervised IQ test.’ Take the quiz, feel flattered by a respectable score, and receive the pitch: for an annual membership fee, you can join the ranks of the certified clever. American Mensa charges $107 per year. The product being sold is not intellectual development, but the feeling of intellectual superiority.

This matters because it shapes how the public understands intelligence. When an organisation with 140,000 members and eight decades of brand recognition tells the world that IQ equals intelligence, people believe it. Parents test their children. Schools sort their students. Employers filter applicants. And at every stage, the metric penalises the slow, recursive, divergent thinking that actually produces breakthroughs. The kid who stares out the window for an hour and then asks a question that reframes the entire problem is not the kid who scores 145. But she might be the one who revolutionises a field.

Mensa could have been an institution that cultivated deep reasoning, funded unconventional research, challenged its members to produce rather than merely perform. Instead, it became a members’ club for people who once did well on a test, complete with a magazine, lanyards and an annual five-day symposium of talks, games and buffets. Far from encouraging any real building, Mensa’s ‘intelligence’ test appears to determine only one thing: who gets invited to the party.

Michael Aaron Cody is an independent researcher published in the European Physical Journal Plus, AI and Society, Philosophy and Cosmology, and the Journal of Modern Physics. His work spans physics, systems theory and defence policy. Follow him on X: @MichaelCody.