Your IQ Matters Less Than You Think

BY DEAN KEITH SIMONTON
In studies of children and historical figures, IQ falls short as a measure of success.

People too often forget that IQ tests haven’t been around that long. Indeed, such psychological measures are only about a century old. Early versions appeared in France with the work of Alfred Binet and Theodore Simon in 1905. However, these tests didn’t become associated with genius until the measure moved from the Sorbonne in Paris to Stanford University in Northern California. There Professor Lewis M. Terman had it translated from French into English, and then standardized on sufficient numbers of children, to create what became known as the Stanford-Binet Intelligence Scale. That happened in 1916. The original motive behind these tests was to get a diagnostic to select children at the lower ends of the intelligence scale who might need special education to keep up with the school curriculum. But then Terman got a brilliant idea: Why not study a large sample of children who score at the top end of the scale? Better yet, why not keep track of these children as they pass into adolescence and adulthood? Would these intellectually gifted children grow up to become genius adults?

Terman subjected hundreds of school kids to his newfangled IQ test. Obviously, he didn’t want a sample so large that it would be impractical to follow their intellectual development. Taking the top 2 percent of the population would clearly yield a group twice as large as the top 1 percent. Moreover, a less select group might be less prone to become geniuses. So why not catch the crème de la crème?

The result was a group of 1,528 extremely bright boys and girls who averaged around 11 years old. And to say they were “bright” is a very big understatement. Their average IQ was 151, with 77 claiming IQs between 177 and 200. These children were subjected to all sorts of additional tests and measures, repeatedly so, until they reached middle age. The result was the monumental Genetic Studies of Genius, five volumes appearing between 1925 and 1959, although Terman died before the last volume came out. These highly intelligent people are still being studied today, or at least the small number still alive. They have also become affectionately known as “Termites”—a clear contraction of “Termanites.”

Now comes the bad news: None of them grew up to become what many people would consider unambiguous exemplars of genius. Their extraordinary intelligence was channeled into somewhat more ordinary endeavors as professors, doctors, lawyers, scientists, engineers, and other professionals. Two Termites actually became distinguished professors at Stanford University, eventually taking over the longitudinal study that included themselves as participants. Their names are Robert R. Sears and Lee Cronbach—and nowhere are they as well-known as Ivan Pavlov, Sigmund Freud, or Jean Piaget, three obvious geniuses in the history of psychology.

Furthermore, many Termites failed to become highly successful in any intellectual capacity. These comparative failures were far less likely to graduate from college or to attain professional or graduate degrees, and far more likely to enter occupations that required no higher education whatsoever. We’re talking only of the males here, too. It would be unfair to consider the females who were born at a time in which all women were expected to become homemakers, no matter how bright. (Even among those women with IQs exceeding 180, not all pursued careers.) Strikingly, the IQs of the successful men did not substantially differ from the IQs of the unsuccessful men. Whatever their differences, intelligence was not a determining factor in those who made it and those who didn’t.

Why not pick a group of obvious adult geniuses, and then try to assess their childhood and adolescent IQs retrospectively from their biographies?

The story goes from bad to worse. Of the many rejects—the children with tested IQs not high enough to make it into the Terman sample—at least two attained higher levels of acclaim than those who had the “test smarts” to become Termites. Here are their stories:

Luis Walter (Luie) Alvarez was born in San Francisco, just up the peninsula from Stanford. He was around 10 years old when he took Terman’s test but scored too low to enter the sample. Yet that rejection did not prevent him from getting his Ph.D. at age 25 from the University of Chicago. Even as a graduate student he began to make important contributions to physics, eventually becoming “one of the most brilliant and productive experimental physicists of the 20th century.” One manifestation of this brilliance was his work on hydrogen bubble chambers for studying elementary particles, which led to his receiving the 1968 Nobel Prize in Physics. No Termite received the Nobel, in physics or otherwise. Oops!

William (Bill) Shockley is the second Termite reject who went on to attain the Nobel Prize in Physics, which he shared with two colleagues in 1956. Born just one year before Alvarez, he grew up in Palo Alto near Stanford, the university his mother had graduated from. Despite his sub-genius score on Terman’s IQ test, he managed to get his B.S. from Cal Tech and his Ph.D. from MIT, both prestigious technical institutions. He then joined Bell Labs and began to publish extensively in solid-state physics, getting his first patent at age 28. Like Luie, Bill got involved in the World War II effort, especially with respect to radar (in his case, bomb sights). After the war, he returned to Bell Labs, where the goal was to find a solid-state substitute for the old glass vacuum tubes that then dominated electronics. The upshot was the transistor.

So there we have it: Little Luie and Bill could have skipped taking the Stanford-Binet and still claim achievements that surpassed Terman’s IQ-certified “geniuses.” But they are not unique among Nobel laureates. Both James Watson, the co-discoverer of DNA’s structure, and Richard Feynman, who worked on the path integral of quantum mechanics, had scores too low to gain membership in Mensa.

Some years after Terman had begun his study of 1,528 high-IQ boys and girls, he acquired a new graduate student named Catharine Cox. Because her mentor’s investigation was already well in progress, she found it difficult to carve out a portion that might serve as her doctoral dissertation. So she tried a bold alternative. If Terman was going to see if high-IQ kids grew up to become adult geniuses, why not do the opposite? In particular, why not pick a group of obvious adult geniuses, and then try to assess their childhood and adolescent IQs retrospectively from their biographies?

Coming up with a list of geniuses is the easy part. For example, nowadays we would just google “famous scientists” or “famous artists” (try it). In Cox’s pre-Internet era, the equivalent would be to compile a list from biographical dictionaries and other (paper) reference works. Fortunately, she found an already published list, from which she extracted the most famous names. She ended up with 301 historic creators and leaders (192 and 109, respectively). No doubt that her sample included some of the top figures in the history of modern Western civilization. Besides the eight mentioned above, she would study big-time creators like Isaac Newton, Jean-Jacques Rousseau, Miguel de Cervantes, Ludwig van Beethoven, and Michelangelo (as well as leaders like Napoleon Bonaparte, Horatio Nelson, Abraham Lincoln, and Martin Luther)—all of them folks who can boast extensive Wikipedia biographies.

The hard part was estimating IQ scores for all 301 geniuses. How is that even possible?

Happily, Terman had already shown, just one year after devising the Stanford-Binet test, how IQs could be estimated from biographies. Back in those days IQ was defined as a literal “intelligence quotient,” namely a child’s mental age divided by his or her chronological age, the arithmetic result then multiplied by 100. The mental age is determined by performance on intellectual tasks that are age graded. Accordingly, if a 5-year-old could do well on tasks more suitable for 10-year-olds, the IQ quotient would become 200 (= 10/5 × 100). Pretty straightforward, no?

Terman applied this method to the early intellectual development of one of his heroes, Francis Galton, the very first scientist to investigate genius. For instance, Francis wrote the following little letter to his older sister: “I am 4 years old and I can read any English book. I can say all the Latin Substantives and Adjectives and active verbs besides 52 lines of Latin poetry. I can cast up any sum in addition and can multiply by 2, 3, 4, 5, 6, 7, 8, [9], 10, [11]. I can also say the pence table. I read French a little and I know the clock.” The two numbers in brackets had been obscured, one by an erasure that made a hole and the other by a more effective paper patch. The young Galton apparently saw that he was claiming too much—an act that itself could be considered evidence of a higher mental age. Now what is the normal expectation for 4-year-old children? Only this: be able to give their gender; name a key, knife, and penny placed before them; repeat back three numbers just told them, and compare two lines in front of their eyes. That’s it! Galton, were he average, would not even be able to count four coins until age 5, give his age until 6, copy a written sentence until 7, or write from dictation until 8. In any event, using additional biographical evidence like this, Terman inferred that Galton’s IQ approached 200. His mental age was almost twice his chronological age.

Cox decided to apply the same method to the 301 but, moving beyond her mentor’s scope, added methodological improvements such as compiling detailed chronologies of intellectual growth from multiple biographical sources and having independent raters make IQ estimates from those chronologies.

Whatever their differences, intelligence was not a determining factor in those who made it and those who didn't.

She also took an additional step. Not all of her creative geniuses achieved the same magnitude of eminence. On the contrary, many were also-rans who would most likely be unknown except to cognoscenti. Examples include the French philosopher Antoine Arnauld, the Swedish chemist Jöns Jacob Berzelius, and the Scottish writer William Robertson. At the same time, her geniuses sometimes exhibited sub-genius IQs—at times too low even to qualify for Mensa. Among these less stratospheric intellects are creators like the Spanish writer Miguel de Cervantes, the Polish astronomer Nicolaus Copernicus, and the French painter Nicolas Poussin. Because all of her geniuses had already been previously ranked on achieved eminence according to the amount of space devoted to them in the reference works—the French general Napoleon came out # 1 while the English writer Harriet Martineau ranked # 301 (ouch!)—Cox could easily correlate the IQ scores with the ranks (inverted, of course). She obtained a statistically significant correlation, and the correlation remained significant even after correcting for data reliability (meaning that the biographical information wasn’t equally good for all geniuses). Furthermore, this positive relation has been replicated multiple times since her own 1926 demonstration. Hence, achieved eminence is associated with superlative intelligence. Her mentor Terman thus seems vindicated!

So far, so good. Scoring high in IQ would seem to increase the odds of attaining acclaim. That being said, four problems cast some doubt on this conclusion.

Problem #1: The Intelligence-Eminence Correlation

The relation between IQ and achieved eminence is not huge or even large. Most statisticians would classify it as a “moderate” relationship. In practical terms, that means that there’s ample room for exceptions at either end. The highly eminent can have IQs lower than average and supremely high IQs can be associated with relative obscurity. I’ve already given three examples of the former, so who illustrates the latter? How about Paolo Sarpi, the Venetian historian? Although his estimated IQ was as high as 195, making him one of the very brightest among the 301, his eminence ranking put him in the lower 20 percent, that is, 242nd!

A more contemporary example is Marilyn vos Savant, who was once listed in the Guinness Book of World Records as having the highest recorded IQ. Reportedly, she had taken a revised version of the Stanford-Binet when she was just 10 years old, and got a perfect score! Although there’s some debate about how best to translate that performance into a precise IQ estimate, it is certainly arguable that she is more intelligent than the brightest Termite and any member of Cox’s 301. Yet what is her main accomplishment? Becoming famous for her super-high IQ! Exploiting that distinctive status, she writes the Sunday column “Ask Marilyn” for Parade magazine. That column doesn’t come close to the writing in Don Quixote or On the Revolutions of the Celestial Spheres, which her two intellectual inferiors, Cervantes and Copernicus, managed to pull off! An extra 60 IQ points or more didn’t buy her any creative edge at all.

Problem #2: The Creative Domain

IQ’s relevance to achievement appears dependent on the domain of achievement. Some domains seem to place far less emphasis on intelligence relative to other domains. For example, famous leaders tend, on the average, to have lower IQs than famous creators. The low IQs of commanders (generals and admirals) is really conspicuous—in Cox’s 301, about 20 points lower than everybody else! The most distinguished military leader in the sample was certainly Napoleon, yet with the highest estimate for his IQ at only 145 he would have been among the less intelligent Termites. Sometimes an excessively high IQ can work against effective leadership: too much of a good thing. Being a “man (or woman) of the people” often implies having an intellect closer to their level. Comprehension is more persuasive than competence. No wonder, then, that the presidents of the United States don’t do much better than the commanders! These results help us understand why the epithet genius seems more likely to be assigned to great creators than to great leaders. Leaders may exhibit charisma, perhaps, but creators are more likely to display genius.

Problem #3: Personality and Persistence Matter

Because the IQ-eminence correlation is so low, even if positive, other psychological factors must be involved that have nothing to do with intelligence. Cox herself revealed as much. Besides assessing her 301 geniuses on IQ, she also took a subset of 100 geniuses for whom the biographical data was particularly good and then measured them on 67 personality traits. Motivational traits emerged as especially critical—persistence standing out above the rest. As she put it: “High but not the highest intelligence, combined with the greatest degree of persistence, will achieve greater eminence than the highest degree of intelligence with somewhat less persistence.” In a sense, the highly eminent are overachievers, attaining more distinction that would be expected from their IQs alone. Natural ability entails not just intelligence, but both passion and perseverance—or what some contemporary psychologists call “grit.”

Problem # 4: Deceptive Assessment

Cox cheated! Not deliberately, I mean, but she cheated nonetheless. Her IQ scores cannot really be equated with Terman’s IQ scores. It’s not just that one set is too high or too low relative to the other, but that the two sets don’t really measure the same thing, at least not most of the time. On the one hand, the Stanford-Binet gauges a person’s acquisition and development of basic cognitive skills, such as memory and reasoning, and rudimentary scholastic skills, such as the proverbial three Rs of “reading, ’riting, and ’rithmetic.” Almost everybody would be expected to possess those elementary skills by the time they reach adulthood. What makes one person smarter than another is mainly the speed at which those skills are acquired. A 5-year-old with an IQ of 200 has somehow managed to master what the average person wouldn’t get until age 10, but otherwise there’s little difference. On the other hand, Cox’s IQ estimates were very often based on skills that would be very rare even in grown-ups. Because these skills are highly specific to a particular domain of creativity, the resulting scores would contrast like apples and oranges, or perhaps even celery and onions.

In most dictionaries, the entry for “genius” provides multiple definitions. And getting a score of 140 on an IQ test is not the only one. Here’s another also given by the American Heritage Dictionary: “Native intellectual power of an exalted type, such as is attributed to those who are esteemed greatest in any department of art, speculation, or practice; instinctive and extraordinary capacity for imaginative creation, original thought, invention, or discovery.” That definition definitely fits the top creators in Cox’s 301, but it fails to apply to any of Terman’s 1,500-plus Termites. If you’re smart enough to score 140 or better on an IQ test, then all by all means go that route. Given that you can take this test when you’re only 2 years old, this may be the best choice if doable. A 2-year-old doing what 3-year-olds can do is not that difficult. You can get it over with while still a toddler and then spend the rest of your life basking in the glory of certified geniushood.

But if you don’t succeed, even after multiple retesting, there’s no need to despair. Just pick some “department of art, speculation, or practice,” and then achieve eminence for some “imaginative creation, original thought, invention, or discovery.” Admittedly, this second course seems much more arduous, and may even take a whole lifetime to accomplish, but at least you can avoid taking any IQ test whatsoever! Plus, your claim to genius status just might withstand the test of time. Authentic genius leaves an impact longer than a testing session, creating a pervasive impression that endures for decades, even centuries.

原文传送门 Nautil