This is one of the crisp anecdotes revealed by Ariane Chemin and Marie-France Etchegoin in their Raoult, a French madness (Gallimard), published this Thursday. While the two journalists asked him if he had already “had his intelligence measured” since he seems to be very proud of it, Didier Raoult told them that his intelligence quotient had been assessed at an exceptional height by a child psychiatrist. He was then fourteen years old, and was out of school. “Listen, your kid has 180 IQ. Let him do it. Everything will be fine,” the therapist reportedly told the future microbiologist’s father, a scene that looks a lot like a “rosebud” to understand the professor’s incredible self-confidence. Marseilles. Unsurprisingly, the information ignited social networks and brought the debates around this psychometric test back to the forefront. But an IQ of 180, is that even possible? We asked Franck Ramus, researcher at the CNRS and associate professor at the Ecole normale supérieure. This cognitive scientist deciphers misunderstandings about IQ, and explains why a score of 180, in addition to being unreliable, would be really, but then really exceptional …
L’Express: What is the highest possible IQ?
Franck Ramus: The question of “greatest possible IQ” does not have the meaning most people attribute to it. Many imagine that IQ scores would be on an absolute scale of cognitive performance: that 150 for example would correspond to a well-defined level of performance, the same everywhere and at all times, and double a performance of 75. But this it’s not the case.
Limited offer. 2 months for 1 € without commitment
IQ scores are standardized scores, i.e. relative to a reference population, which establishes the statistical standard of intelligence for a given age group, in a given country, at a given period. By convention, the mean of this reference population gives the score 100, and its standard deviation is 15. An IQ of 100 therefore means that the person does better than 50% of people of the same age, and worse than 50%. . An IQ of 115 means she is doing better than 84%. Having an IQ of 145 means that she is doing better than 99.9% of the individuals in this reference population.
But a score of 145 does not necessarily correspond to the same level of absolute performance in France and Taiwan; nor does it correspond to the same level of performance in France in 2020 and in France in 1970 (because absolute cognitive performance has increased in the meantime); nor does it correspond to the same level of performance at 12, 25 and 70 years old. Because different age groups, different countries, at different times, have different reference populations.
The question of the maximum IQ score is therefore not the question of which cognitive performance is the highest on an absolute scale. It is an essentially methodological question, which depends on the size of the reference population, the difficulty of the test and the number of difficult questions allowing to decide the best.
To what part of the population would an IQ of 180 correspond in theory?
One in 20 million people.
In your opinion, scores above 145 are no longer reliable. Why ?
In practice, the reference populations to which we can compare oneself number at most a few thousand individuals, and they only include a few individuals with a very high level of intelligence. Very high IQs are therefore obtained by extrapolation, with a large margin of error. Wechsler’s batteries, which have the most reliable standards, provide standards for scores up to 160, but with wide margins of error. We can consider that from 145, we are in the extrapolation and that the reliability of these scores is limited. For children, who can sometimes be way ahead of their age, extended battery standards have been created for very high IQs. They go up to an IQ of 210. That said, only one in 10,000,000,000,000 (ten trillion) should theoretically be able to achieve such a score. He may not have been born yet.
How then to explain the higher scores that can be found in certain estimates?
This may be due to the unusual conventions of the standards of the test they passed. For example, the standards of the “Cattell culture fair” test are established with an average of 100 and a standard deviation of 24. With this convention, a score of 145 on a Wechsler battery (and on most other tests) corresponds to a score of 172 at Cattell’s battery. To interpret a score, it is therefore crucial to know which test a person has taken and which scale of scores is used!
It is also possible that certain batteries of tests used on the Internet, by associations or by companies, give scores higher than 160. But until proven otherwise, they do not have standards on a reference population large enough to that these scores are valid.
“Slowing IQ rise is inevitable”
The internet is full of IQ estimates for famous but long-dead figures like Newton and Mozart. However, we know that the notion of IQ was only created at the beginning of the 21st century …
Some people go out of their way to calculate IQ scores for dead “geniuses”. In the absence of testing, these estimates are of little value …
The irony is that Didier Raoult, in a column of Le Point in 2018, was alarmed by “the worrying drop in our IQ”, attributing it to the screens. What about the facts?
There have been a few isolated studies in recent years that have suggested a drop in the average IQ of the population. This feeds a disastrous discourse evoking the effects of junk food and the environment, screens or immigration. But if you synthesize all the studies, you actually see a slowdown in the rise, which means that the IQ is starting to level off. We have gained 30 points in a century, which is huge and has been called the “Flynn effect”. But, at the same time, a slowdown is inevitable since the IQ will not go up endlessly. It is like the increase in the size of individuals. As a result of improved nutrition and health, we have become taller, which does not mean that one day we will reach 3 meters, because there are physiological limits. We have already greatly optimized the environmental factors that condition the development of the body and the brain, the gains that can be made are now limited.
To follow the analysis and decryption wherever you are
Download the app
In conclusion, does the estimate of an IQ at 180 made for the young Raoult seem fanciful to you?
I am not saying that Didier Raoult is lying. But be aware that any score above 160 (on a scale with a standard deviation of 15) is likely to be fanciful and should inspire suspicion.
The best of worlds