His name was William James Sidis, and his IQ was estimated at between 250 and 300 [8, p. 283]. At eighteen months he could read The New York Times, at two he taught himself Latin, at three he learned Greek. By the time he was an adult he could speak more than forty languages and dialects. He gained entrance to Harvard at eleven, and gave a lecture on four-dimensional bodies to the Harvard Mathematical Club his first year. He graduated cum laude at sixteen, and became the youngest professor in history. He deduced the possibility of black holes more than twenty years before Subrahmanyan Chandrasekhar published An Introduction to the Study of Stellar Structure. His life held possibilities for achievement that few people can imagine. Of all the prodigies for which there are records, his was probably the most powerful intellect of all. And yet it all came to nothing. He soon gave up his position as a professor, and for the rest of his life wandered from one menial job to another. His experiences as a child prodigy had proven so painful that he decided for the rest of his life to shun public exposure at all costs. Henceforth, he denied his gifts, refused to think about mathematics, and above all refused to perform as he had been made to do as a child. Instead, he devoted his intellect almost exclusively to the collection of streetcar transfers, and to the study of the history of his native Boston. He worked hard at becoming a normal human being, but never entirely succeeded. He found the concept of beauty, for example, to be completely incomprehensible, and the idea of sex repelled him. At fifteen he took a vow of celibacy, which he apparently kept for the remainder of his life, dying a virgin at the age of 46. He wore a vest summer and winter, and never learned to bathe regularly. A comment that Aldous Huxley once made about Sir Isaac Newton might equally have been said of Sidis.

For the price Newton had to pay for being a supreme intellect was that he was incapable of friendship, love, fatherhood, and many other desirable things. As a man he was a failure; as a monster he was superb [5, p. 2222].

There was a time when all precocious children were thought to burn out the same way that Sidis did. The man most responsible for changing this belief was Lewis M. Terman. Between 1900 and 1920 he was able to carry out a study of about a hundred gifted children, and his observations convinced him that many of the traditional beliefs about the gifted were little more than superstitions. To confirm these observations, he obtained a grant from the Commonwealth Fund in 1922, and used it to sift a population of more than a quarter of a million children, selecting out all those with IQs above 140 for further study. That group has been monitored continuously ever since. Many of the previously held beliefs about the gifted did indeed turn out to be false. The gifted are not weak or sickly, and although the incidence of myopia is greater among them, they are generally thought to be better looking than their contemporaries: They are not nerds.

Nevertheless, in his rush to dispel the erroneous beliefs about the gifted, Terman sometimes made claims not supported by his own data. In fact, in some cases, the data suggests that exactly the opposite conclusion should have been drawn. Terman's own data shows that there is a definite connection between measured intelligence and mental and social maladjustment. The consequences of misinterpreting these data are so grave that it will pay to re-examine them in some detail.

Terman's longitudinal research on the gifted included a constant assessment of mental health and social adjustment. Subjects were classified into three categories: satisfactory adjustment, some maladjustment, and serious maladjustment. Terman defined these categories in the following way.

1. Satisfactory. Subjects classified in this category were essentially normal; i.e., their “desires, emotions, and interests were compatible with the social standards and pressures” of their group. Everyone, of course, has adjustment problems of one kind or another. Satisfactory adjustment as here defined does not mean perfect contentment and complete absence of problems, but rather the ability to cope adequately with difficulties in the personal make-up or in the subject's environment. Worry and anxiety when warranted by the circumstances, or a tendency to be somewhat high strung or nervous–provided such a tendency did not constitute a definite personality problem–were allowed in this category. 2. Some maladjustment. Classified here were subjects with excessive feelings of inadequacy or inferiority, nervous fatigue, mild anxiety neurosis, and the like. The emotional conflicts, nervous tendencies and social maladjustments of these individuals, while they presented definite problems, were not beyond the ability of the individual to handle, and there was no marked interference with social or personal life or with achievement. Subjects whose behavior was noticeably odd or freakish, but without evidence of serious neurotic tendencies, were also classified in this category. 3. Serious maladjustment. a.) Classified as 3a were subjects who had shown marked symptoms of anxiety, mental depression, personality maladjustment, or psychopathic personality. This classification also includes subjects who had suffered a “nervous breakdown,” provided the condition was not severe enough to constitute a psychosis. Subjects with a previous history of serious maladjustment or nervous breakdown (without psychosis) were included here even though their adjustment at the time of rating may have been entirely satisfactory. b.) Classified as 3b were those subjects who had at any time suffered a complete mental breakdown requiring hospitalization, whatever their condition at the time of rating. In the majority of cases the subjects were restored to reasonably good mental health after a brief period of hospital care [6, pp. 99-101].

In 1940, when the group was about 29 years of age, a large scale examination was carried out. Included in that examination was a high level test of verbal intelligence, designated at that time the Concept Mastery, but later re-named the Concept Mastery test form A. Terman found the following relationship between adjustment and verbal intelligence. (These are raw scores, not IQs.)

CMT-A [6, p. 115]

The data show three things. First, that there is a definite trend for the maladjusted to make higher scores on the Concept Mastery test. Second, that women show symptoms of maladjustment at lower scores than men. And third, that 21 percent of the men and 18 percent of the women showed at least some form of maladjustment.

During 1950-52, when the group was approximately 41 years old, another examination was made using a new test, the Concept Mastery test form T. Test scores were again compared to assessments of adjustment. (CMT-T scores are not interchangeable with CMT-A scores. They have different means and standard deviations.)

CMT-T [7, p. 50]

Similar conclusions can be drawn from these data as well. Again, there is a definite trend shown for the maladjusted to make higher scores than the satisfactorily adjusted. Again, women show symptoms of maladjustment at lower scores than men. But the most alarming thing of all is that the percentage of maladjustment shown for both sexes rose in the 12 years since the previous examination. The percentage of men showing maladjustment having risen from 21 percent to 29 percent, and the figure for women having risen from 18 percent to 33 percent! Nearly double what it was before!

How did Terman interpret these data? Terman states:

Although severe mental maladjustment is in general somewhat more common among subjects who score high on the Concept Mastery test, many of the most successful men of the entire group also scored high on this test [7, p. 50].

In other words, Terman deliberately tried to give the impression that the relationship between verbal intelligence and mental and social maladjustment was weak and unreliable. He did this by misdirection. He gave a truthful answer to an irrelevant question. Terman failed to realize that a small difference in means between two or more distributions can have a dramatic effect on the percentage of each group found at the tails of the distribution. The relevant questions should have been “what is the percentage of maladjustment found at different levels of ability, and does this show a trend?” Terman's data can be used to find answers to these questions.

The method used to solve this problem is a relatively simple one but tedious in detail. (See appendix.) The results, however, are easy to understand. Using CMT-T scores for men as an illustration, and pooling the data for some maladjustment and serious maladjustment, the following percentages can be obtained.

PERCENTAGE OF MEN SHOWING SOME OR SERIOUS MALADJUSTMENT AT SIX LEVELS OF ABILITY

By comparison, the Triple Nine Society averages 155.16 on the CMT-T, and the average score for Prometheus Society members is 169.95 [1, 2]. The implications are staggering, especially when it is realized that these percentages do not include women, who show more maladjustment at lower CMT-T scores than men do. Perhaps this is one of the reasons why super high IQ societies suffer so much from schisms and a tendency towards disintegration. In any event, one thing is certain. The currently accepted belief that verbal intelligence is unrelated to maladjustment is clearly a myth.

Nevertheless, while Terman's data do provide a prima facie case for a connection between verbal intelligence and maladjustment, they fail to explain the causal mechanism involved. To obtain such insight requires close observation by a gifted observer. Fortunately, those insights are available to us in Leta S. Hollingworth's book, Children above 180 IQ. Hollingworth not only observed her subjects as children, she also continued to maintain some contact with them after they had reached maturity. So although her book is ostensibly about children, it is in fact laced throughout by her observations on exceptionally gifted adults as well.

Before examining Hollingworth's findings, however, it is necessary to explain how childhood IQs are related to adult mental ability. As a child ages, his IQ tends to regress to the mean of the population of which he is a member. This is partly due to the imperfect reliability of the test, and partly due to the uneven rate of maturation. The earlier the IQ is obtained, and the higher the score, the more the IQ can be expected to regress by the time the child becomes an adult. So although Hollingworth's children were all selected to have IQs above 180, their adult status was not nearly so high. In fact, as adults, there's good reason to believe that their abilities averaged only slightly above that of the average Triple Nine member. Evidence for this conjecture comes from the Terman research data. Terman observed the following relationship between childhood IQs on the Stanford-Binet and adult status on the Concept Mastery test form T.

CONCEPT MASTERY SCORES ACCORDING TO CHILDHOOD STANFORD-BINET IQ [7, p. 58]

The average childhood IQ score for those with childhood IQs above 170 was 177.7 for men, and 177.6 for women. That's quite close to the 180 cutoff used by Leta Hollingworth in selecting her subjects. Note that Terman's subjects who scored above 170 IQ as children averaged 155.8 on the CMT-T at age 41, a score quite close to the 155.16 made by the average Triple Nine member. Such a close match makes it reasonable to generalize Hollingworth's findings to members of both the Triple Nine Society and the Prometheus Society.

Hollingworth identified a number of adjustment problems caused by school acceleration. As this is rarely practiced in today's educational system, these are no longer problems and will not be discussed. There still remain, however, four adjustment problems that continue to perplex the gifted throughout their lives, two applying to all levels of giftedness, and two applying almost exclusively to the exceptionally gifted–i.e. those with childhood IQs above 170, or adult Concept Mastery test (T) scores above 155.

One of the problems faced by all gifted persons is learning to focus their efforts for prolonged periods of time. Since so much comes easily to them, they may never acquire the self-discipline necessary to use their gifts to the fullest. Hollingworth describes how the habit begins.

Where the gifted child drifts in the school unrecognized, working chronically below his capacity (even though young for his grade), he receives daily practice in habits of idleness and daydreaming. His abilities never receive the stimulus of genuine challenge, and the situation tends to form in him the expectation of an effortless existence [3, p. 258].

But if the “average” gifted child tends to acquire bad adjustment habits in the ordinary schoolroom, the exceptionally gifted have even more problems. Hollingworth continues:

Children with IQs up to 150 get along in the ordinary course of school life quite well, achieving excellent marks without serious effort. But children above this mental status become almost intolerably bored with school work if kept in lockstep with unselected pupils of their own age. Children who rise above 170 IQ are liable to regard school with indifference or with positive dislike, for they find nothing in the work to absorb their interest. This condition of affairs, coupled with the supervision of unseeing and unsympathetic teachers, has sometimes led even to truancy on the part of gifted children [3, p. 258].

A second adjustment problem faced by all gifted persons is due to their uncommon versatility. Hollingworth says:

Another problem of development with reference to occupation grows out of the versatility of these children. So far from being one-sided in ability and interest, they are typically capable of so many different kinds of success that they may have difficulty in confining themselves to a reasonable number of enterprises. Some of them are lost to usefulness through spreading their available time and energy over such a wide array of projects that nothing can be finished or done perfectly. After all, time and space are limited for the gifted as for others, and the life-span is probably not much longer for them than for others. A choice must be made among the numerous possibilities, since modern life calls for specialization [3, p. 259].

A third problem faced by the gifted is learning to suffer fools gladly. Hollingworth notes:

A lesson which many gifted persons never learn as long as they live is that human beings in general are inherently very different from themselves in thought, in action, in general intention, and in interests. Many a reformer has died at the hands of a mob which he was trying to improve in the belief that other human beings can and should enjoy what he enjoys. This is one of the most painful and difficult lessons that each gifted child must learn, if personal development is to proceed successfully. It is more necessary that this be learned than that any school subject be mastered. Failure to learn how to tolerate in a reasonable fashion the foolishness of others leads to bitterness, disillusionment, and misanthropy [3, p. 259].

The single greatest adjustment problem faced by the gifted, however, is their tendency to become isolated from the rest of humanity. This problem is especially acute among the exceptionally gifted. Hollingworth says:

This tendency to become isolated is one of the most important factors to be considered in guiding the development of personality in highly intelligent children, but it does not become a serious problem except at the very extreme degrees of intelligence. The majority of children between 130 and 150 find fairly easy adjustment, because neighborhoods and schools are selective, so that like-minded children tend to be located in the same schools and districts. Furthermore, the gifted child, being large and strong for his age, is acceptable to playmates a year or two older. Great difficulty arises only when a young child is above 160 IQ. At the extremely high levels of 180 or 190 IQ, the problem of friendships is difficult indeed, and the younger the person the more difficult it is. The trouble decreases with age because as persons become adult, they naturally seek and find on their own initiative groups who are like-minded, such as learned societies [3, p. 264].

Hollingworth points out that the exceptionally gifted do not deliberately choose isolation, but are forced into it against their wills.

These superior children are not unfriendly or ungregarious by nature. Typically they strive to play with others but their efforts are defeated by the difficulties of the case… Other children do not share their interests, their vocabulary, or their desire to organize activities. They try to reform their contemporaries but finally give up the struggle and play alone, since older children regard them as “babies,” and adults seldom play during hours when children are awake. As a result, forms of solitary play develop, and these, becoming fixed as habits, may explain the fact that many highly intellectual adults are shy, ungregarious, and unmindful of human relationships, or even misanthropic and uncomfortable in ordinary social intercourse [3, p. 262].

But if the exceptionally gifted is isolated from his contemporaries, the gulf between him and the adult authorities in his life is even deeper.

The very gifted child or adolescent, perceiving the illogical conduct of those in charge of his affairs, may turn rebellious against all authority and fall into a condition of negative suggestibility–a most unfortunate trend of personality, since the person is then unable to take a cooperative attitude toward authority. A person who is highly suggestible in a negative direction is as much in bondage to others around him as is the person who is positively suggestible. The social value of the person is seriously impaired in either case. The gifted are not likely to fall victims to positive suggestion but many of them develop negativism to a conspicuous degree [3, p 260].

Anyone reading the super high IQ journals is aware of the truth of this statement. Negative individuals abound in every high IQ society.

Hollingworth distilled her observations into two ideas that are among the most important ever discovered for the understanding of gifted behavior. The first is the concept of an optimum adjustment range. She says:

All things considered, the psychologist who has observed the development of gifted children over a long period of time from early childhood to maturity, evolves the idea that there is a certain restricted portion of the total range of intelligence which is most favorable to the development of successful and well-rounded personality in the world as it now exists. This limited range appears to be somewhere between 125 and 155 IQ. Children and adolescents in this area are enough more intelligent than the average to win the confidence of large numbers of their fellows, which brings about leadership, and to manage their own lives with superior efficiency. Moreover, there are enough of them to afford mutual esteem and understanding. But those of 170 IQ and beyond are too intelligent to be understood by the general run of persons with whom they make contact. They are too infrequent to find congenial companions. They have to contend with loneliness and personal isolation from their contemporaries throughout the period of their immaturity. To what extent these patterns become fixed, we cannot yet tell [3, p. 264].

Hollingworth's second seminal idea is that of a “communication range.” She does not state this explicitly, but it can be inferred from some of her comments on leadership.

Observation shows that there is a direct ratio between the intelligence of the leader and that of the led. To be a leader of his contemporaries a child must be more intelligent but not too much more intelligent than those to be led… But generally speaking, a leadership pattern will not form–or it will break up–when a discrepancy of more than about 30 points of IQ comes to exist between leader and led [3, p. 287].

The implication is that there is a limit beyond which genuine communication between different levels of intelligence becomes impossible. To say that a child or an adult is intellectually isolated from his contemporaries is to say that everyone in his environment has an IQ at least 30 points different from his own. Knowing only a person's IQ, then, is not enough to tell how well he's likely to cope with his environment. Some knowledge of the intellectual level of his environment is also necessary.

If the optimum range of intelligence lies between 125 and 155 IQ, as Hollingworth suggests, then it follows that 155 can be thought of as a threshold separating an optimum adjustment zone below it from a suboptimum range above it. Other psychologists have also noticed how this score tends to divide people into two naturally occurring categories. Among these is one of the doyens of psychometrics, David Wechsler. He comments:

The topics of genius and degeneration are only special cases of the more general problem involved in the evaluation of human capacities, namely the quantitative versus qualitative. There are those who insist that all differences are qualitative, and those who with equal conviction maintain that they are exclusively quantitative. The true answer is that they are both. General intelligence, for example, is undoubtedly quantitative in the sense that it consists of varying amounts of the same basic stuff (e.g., mental energy) which can be expressed by continuous numerical measures like intelligence Quotients or Mental-Age scores, and these are as real as any physical measurements are. But it is equally certain that our description of the difference between a genius and an average person by a statement to the effect that he has an IQ greater by this or that amount, does not describe the difference between them as completely or in the same way as when we say that a mile is much longer than an inch. The genius (as regards intellectual ability) not only has an IQ of say 50 points more than the average person, but in virtue of this difference acquires seemingly new aspects (potentialities) or characteristics. These seemingly new aspects or characteristics, in their totality, are what go to make up the “qualitative” difference between them [9, p. 134].

Wechsler is saying quite plainly that those with IQs above 150 are different in kind from those below that level. He is saying that they are a different kind of mind, a different kind of human being.

This subjective impression of a difference in kind also appears to be fairly common among members of the super high IQ societies themselves. When Prometheus and Triple Nine members were asked if they perceived a categorical difference between those above this level and others, most said that they did, although they also said that they were reluctant to call the difference genius. When asked what it should be called, they produced a number of suggestions, sometimes esoteric, sometimes witty, and often remarkably vulgar. But one term was suggested independently again and again. Many thought that the most appropriate term for people like themselves was Outsider.

The feeling of estrangement, or at least detachment, from society at large is not merely subjective illusion. Society is not geared to deal effectively with the exceptionally gifted adult because almost nothing objective is known about him. It is a commonplace observation that no psychometric instrument can be validly used to evaluate a person unless others like him were included in the test's norming sample. Yet those with IQs above 150 are so rare that few if any were ever included in the norming sample of any of the most commonly used tests, tests like the Strong-Campbell Interest Inventory, the Kuder Vocational Preference Record, the MMPI and so on. As a consequence, objective self-knowledge for the exceptionally gifted is nearly impossible to obtain. What he most needs to know is not how he differs from ordinary people–he is acutely aware of that–but how he is both like and unlike those of his own kind. The most commonly used tests can't provide that knowledge, so he is forced to find out in more roundabout ways. It is his attempts to find answers to these questions that may explain the emergence of the super high IQ societies. Where else can he find peers against which to measure himself?

There appear to be three sorts of childhoods and three sorts of adult social adaptations made by the gifted. The first of these may be called the committed strategy. These individuals were born into upper middle class families, with gifted and well educated parents, and often with gifted siblings. They sometimes even had famous relatives. They attended prestigious colleges, became doctors, lawyers, professors, or joined some other prestigious occupation, and have friends with similar histories. They are the optimally adjusted. They are also the ones most likely to disbelieve that the exceptionally gifted can have serious adjustment problems.

The second kind of social adaptation may be called the marginal strategy. These individuals were typically born into a lower socio-economic class, without gifted parents, gifted siblings, or gifted friends. Often they did not go to college at all, but instead went right to work immediately after high school, or even before. And although they may superficially appear to have made a good adjustment to their work and friends, neither work nor friends can completely engage their attention. They hunger for more intellectual challenge and more real companionship than their social environment can supply. So they resort to leading a double life. They compartmentalize their life into a public sphere and a private sphere. In public they go through the motions of fulfilling their social roles, whatever they are, but in private they pursue goals of their own. They are often omnivorous readers, and sometimes unusually expert amateurs in specialized subjects. The double life strategy might even be called the genius ploy, as many geniuses in history have worked at menial tasks in order to free themselves for more important work. Socrates, you will remember was a stone mason, Spinoza was a lens grinder, and even Jesus was a carpenter. The exceptionally gifted adult who works as a parking lot attendant while creating new mathematics has adopted an honored way of life and deserves respect for his courage, not criticism for failing to live up to his abilities. Those conformists who adopt the committed strategy may be pillars of their community and make the world go around, but historically, those with truly original minds have more often adopted the double life tactic. They are ones among the gifted who are most likely to make the world go forward.

And finally there are the dropouts. These sometimes bizarre individuals were often born into families in which one or more of the parents were not only exceptionally gifted, but exceptionally maladjusted themselves. This is the worst possible social environment that a gifted child can be thrust into. His parents, often driven by egocentric ambitions of their own, may use him to gratify their own needs for accomplishment. He is, to all intents and purposes, not a living human being to them, but a performing animal, or even an experiment. That is what happened to Sidis, and may be the explanation for all those gifted who “burn out” as he did. (Readers familiar with the Terman study will recognize the committed strategy and the marginal strategy as roughly similar to the adjustment patterns of Terman's A and C groups.)

If the exceptionally gifted adult with an IQ of 150, or 160, or 170 has problems in adapting to his world, what must it have been like for William James Sidis, whose IQ was 250 or more?

Aldous Huxley once wrote:

Perhaps men of genius are the only true men. In all the history of the race there have been only a few thousand real men. And the rest of us–what are we? Teachable animals. Without the help of the real man, we should have found out almost nothing at all. Almost all the ideas with which we are familiar could never have occurred to minds like ours. Plant the seeds there and they will grow; but our minds could never spontaneously have generated them [4, p. 2242].

And so we see that the explanation for the Sidis tragedy is simple. Sidis was a feral child; a true man born into a world filled with animals–a world filled with us.

* * * *

Some of those reading this paper may find the portrait painted here to be completely incredible. Their own experiences were nothing at all like those described, nor were those of most of their gifted friends. But the point of this article is not that there's some special hazard in having an exceptional IQ: There's not. The point is that the danger lies in having an exceptional IQ in an environment completely lacking in intellectual peers. It's the isolation that does the damage, not the IQ itself.

It is the belief of this author that the super high IQ societies were created primarily by those who have adopted the marginal strategy, and by rights ought to be aimed at fulfilling the needs of this subdivision of the exceptionally gifted. It's obvious from reading the journals that those who have followed the committed strategy rarely participate in society affairs, rarely write for the various journals, and indeed have little need to belong to such a group. They have far more productive outlets for their talents. It's the exceptionally gifted adult who feels stifled that stands most in need of a high IQ society. The tragedy is that none of the super high IQ societies created thus far have been able to meet those needs, and the reason for this is simple. None of these groups is willing to acknowledge or come to terms with the fact that much of their membership belong to the psychological walking wounded. This alone is enough to explain the constant schisms that develop, the frequent vendettas, and the mediocre level of their publications. But those are not immutable facts; they can be changed. And the first step in doing so is to see ourselves as we are.

Source : http://prometheussociety.org (From The Prometheus Society's Journal, Gift of Fire Issue No. 22, April 1987. This article was re-issued in Issue 72, March 95.)

In 1921, Lewis M. Terman, by means of teacher selection and multiple testing, identified 1,500 of the most intelligent children in the California school system. His aim was to follow this group of gifted people throughout their lives and observe their achievements in the widest possible sense. He was interested not only in their educational and vocational successes, but also their hobbies, interests, reading habits, health, and many other things.

As children, the Terman group was larger and heavier than the average California child; they matured earlier and their health was better. They averaged one full grade ahead of their contemporaries in grade placement, while at the same time more than half of them had mastered the curriculum a full two or more grades beyond their classmates. There was no evidence that the gifted group was cursed with any compensatory weaknesses, but because they did prefer older playmates and more solitary play, they were sometimes reported by others to be “queer” or “different”. Nevertheless, they were not thought to be any more unsociable or unpopular than other children.

As adults, most went on to college and many earned academic distinctions. Delinquency and crime were both below average in frequency, and suicide rates, marriage rates, and age of marriage were the same, or nearly the same as those of the general population. The group's war record, occupational success, and income was clearly superior. By age 35, they had published 90 books and 1,500 articles. By age 45, the group had produced 2,000 scientific papers, 230 patents, and 33 novels, as well as many shorter literary works. All in all, the general picture was one of high scholastic achievement and pronounced occupational success.

Terman entitled his longitudinal research project Genetic Studies of Genius, and quite clearly, this was exactly what Terman expected it to be. But soon the evidence began to mount that IQ was not the index of genius that Terman had first thought. There was no John F. Kennedy, no Robert Oppenheimer, no J.B.S. Haldane, and not even a Henry Ford. Clearly, IQ was the single most important variable ever discovered for the prediction of achievement, but it was just as clear that genius was too elusive to be captured by the IQ net alone.

Finally, in the 39th Yearbook of the National Society for the Study of Education Part I, pp. 83-84, Terman made a most astonishing statement. “Our conclusion is that for subjects brought up under present-day educational regimes, excess in IQ above 140 or 150 adds little to one's achievement in the early adult years.” A little farther on he says, “The data reviewed indicate that, above the IQ level of 140, adult success is largely determined by such factors as social adjustment, emotional stability, and drive to accomplishment.”

In other word, an extremely high IQ conveys no practical advantages at all.

For a man who had devoted most of his life to the study of gifted people, this was a painful admission for him to make. It is patently obvious that he would not have made it unless he sincerely believed it to be true. But was it true? Was Terman right?

I decided to find out.

The thing to do, it seemed to me, was to reverse Terman's procedure. Find as many intellectually demanding fields as possible, for which there were documented IQ records, and check them for measured intelligence. In other words, use demonstrable, real-world accomplishment as a check against the efficacy of the IQ as a predictor.

I began with a summary of IQs found for doctorates awarded in 1958 (Science, 1961, Vol 133, Jan-Jun, 679-688). The test used for these comparisons was the Army General Classification Test, which has a mean of 100 and a standard deviation of 20. A score of 162 on the AGCT is required for admission to TNS (Vidya #7, Nov 1979).

I also took the liberty of computing an estimate of LAIT equivalent scores as an aid in making comparisons.

Mean and standard deviation of the doctorates for 1958 on the AGCT and the LAIT

                                    AGCT scores              LAIT scores
   Field of doctorate               Mean    S.D.             Mean    S.D.    
   Math                             138.2   17.0             130.6   13.6
   Physics                          140.3   16.4             132.2   13.1
   Chemistry                        131.5   16.3             125.2   13.0
   Geology                          133.1   14.7             126.5   11.8
   Engineering                      134.8   16.2             127.8   13.0
      All physical sciences         134.7   16.6             127.8   13.3
      Biological sciences           126.1   16.4             120.8   13.1
      Social sciences               132.0   16.9             125.6   13.5
      Arts and humanities           132.1   16.4             125.7   13.1
      Education                     123.3   16.2             118.6   13.0
   Social sciences, arts, educ      129.8   17.0             123.8   13.6
   Natural sciences                 131.7   17.0             125.4   13.6

   Total                            130.8   17.1             124.6   13.7    

The following chart presents the same data in a somewhat different way. Using this format, it's a simple matter to compute percentages falling above or below any given IQ level. For example, more than 70 percent of all doctorates for 1958 had IQs below the 98th percentile. Even in the physical sciences, more than 60 percent fell below the 98th percentile, and only about 5 percent were above TNS admission standards.

Distribution of intelligence test scores for five general fields of the doctorate and for the total doctorate population:

                                        
Army Standard  Approx. gen   __________________Doctorates (N)___________________ 
 Scale dist    population      All     Phys     Biol     Social   Arts,   Educ-
               age 32, 1958  fields   sciences sciences sciences human.   cation 
170-and up           530       46        20        1       15       7        3
160-169            2,670      101        46       11       27      14        3
150-159           12,150      337       153       37       93      35       19
140-149           39,250      530       246       67      112      74       31
130-139          108,000      826       298      150      179     116       83
120-129          218,200      806       243      153      214      89      107
110-119          361,800      520       140      119      106      65       90
100-109          457,400      298        64       79       67      28       60
 90-99           457,400       81        19       19       12       8       23
 80-89           361,800       15         3        4        2                6
 70-79           218,200        7         1        5                1
Below 70         162,600
––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––
Total          2,400,000     3567      1233      645      827     437      425
100 (mean)                    130.8     134.7    126.1    132.3   132.1    123.3
No information               4220       931      676      924     771      918  

Terman's statement was based on data provided by the Stanford-Binet. As the remainder of the IQs in this paper will be based on the Wechsler-Bellevue (W-B), or the Wechsler Adult Intelligence Scale (WAIS), it's important to know that Stanford-Binet IQs are typically found to be 10 to 12 points higher than Wechsler-Bellevue scores. In fact, ”…the difference between the Revised Stanford-Binet and the Wechsler-Bellevue IQs become progressively greater the higher the IQs are above 100.” (Jour of psych XIV, 1942, 317-326. See also Jour of Consul Psych 25, 10 61, 390, and the Jour of Social Psych XXIII, 1946, 237-239) This should not be construed to mean that the Stanford-Binet is less valid than the Wechsler-Bellevue, the Wechsler Adult Intelligence Scale, or any other highly reliable test. In fact, the Stanford-Binet may be somewhat more valid at the very highest levels. All that it does mean is that the level of performance represented by a score on the W-B, or the WAIS, will be represented by a score 10 or 12 points higher on the Stanford-Binet. At the very highest levels this difference will be even greater.

One of the most outstanding groups of men that I was able to find was that of the faculty of the University of Cambridge. (Nature, 1967, 213, 442) These scores represent the Full Scale WAIS IQs of 148 faculty members in a variety of disciplines from one of the most distinguished Universities in the world.

The WAIS has a mean of 100 and a standard deviation of 15. Entrance requirements to TNS on the WAIS is a score of 148 (Vidya #7, Nov 1979)

Mean, range, variance, and standard error of the mean, for 148 Cambridge faculty

 
                                     Mean        Range of               Standard
Subject of                          deviation     scores    Variance    deviation
bachelor degree            n           IQs                             of the mean
----------------------------------------------------------------------------------  
Agricultural sciences     17          121.6       110-135     41.18       1.55
Biochemistry              10          130.0       122-141     41.33       2.034
Biological sciences       20          126.1       113-135     27.05       1.15
Chemistry                 12          129.6       121-138     30.45       1.59
Engineering sciences      16          125.0       111-138     57.46       1.9
Mathematics               16          130.4       124-136     16.0        1.0
Medical sciences          10          127.0       116-134     46.22       2.15
Physics                   20          127.7       112-136     39.05       1.4
Social sciences           10          121.8       112-132     41.55       2.04

“The scores range from 110 to 141 with a mean of 126.5 and a standard deviation of 6.3 points. All the scores fall within three standard deviations of the mean. The distribution of the scores in terms of Wechsler's classification show that all the scientists obtained scores above the seventieth percentile rank for the general population – 35.2 per cent are classified as “very superior”, 51.3 per cent as “superior”, and 13.5 per cent as “bright normal”. Approximate percentages of these three groups in a general population sample are 2.2, 6.7, and 16.1 respectively.”

The following chart is taken from Wechsler's Measurement and Appraisal of Adult Intelligence by Joseph D. Matarazzo, 5th edition, p 177.

                                      XX
                                    X XX  X     X
                                    X XX  X   X X
                                 X  X XX  X X X X
                                 XXXXXXX  X XXXXX   X
                                XXXXXXXX XXXXXXXXX XX
                         X     XXXXXXXXXXXXXXXXXXXXXX  X           X       
             |          |          |          |          |          |
            100        110        120        130        140        150

The Full Scale WAIS IQs for 80 medical students. Adapted from Kole, D. M. A study of intellectual and personality characteristics of medical students. (The Journal of Medical Education, 1965, 40, 1130-1143)

”…for individuals in a profession known to require high intellectual prowess there is, nevertheless, a wide variability around the median IQ of 125.5. The actual range of Full Scale IQ on the WAIS for these young physicians is from a low of 111 (77th percentile) to a high of 149 (99.9th percentile).

Matarazzo goes on to say, “Another point which is well understood among the few currently growing group of specialists in this country who deal primarily with the assessment of individuals applying for entry into medicine, law, graduate nursing, physics, economics, psychology, English literature, graduate business school, and a host of related scientific and humanistic disciplines is that, on the basis of a purely intellectual index, such applicants are so similar as to have come from the same population.

Currently practicing physicians are also similar to these young medical students in Full Scale WAIS IQs. (Matarazzo and Goldstein, Journal of Medical Education, 1972, 47, 102-111)

Holt and Luborsky in Personality Patterns of Psychiatrists, 1958, p. 40, reported that the “Average Wechsler-Bellevue Scale of Adult Intelligence was 128” … “Scores ranged from 110 to 145. IQ was not decisive in gaining admittance; the average of those accepted was only slightly higher than those who were rejected. The average verbal subscore was 131 with a range of 115-145.”

Balinsky and Shaw (Personnel Psychology, 1956, 9, 207-209) reported on an appraisal of top level executives a Full Scale IQ of 124.09, S.D. 7.90; Verbal Scale IQ 125.1, S.D. 8.09; and Performance IQ 117.09, S.D. 9.81. They also reported that only the Verbal IQ was significantly correlated with performance ratings (r = .32, p<.05). However, the subscale of arithmetical reasoning proved to be the best predictor of management performance (r = .42, p<.01).

Up to this point I have tried to avoid using data from less demanding vocations. But because the article by L.M. Simon and E.A. Levitt (Occupations, 1950, 29, 23-25) is the most extensive set of norms I've ever seen in one place for the W-B, I've decided to include their full table.

        SCORES ON THE WECHSLER BELLEVUE SCALE IN RELATION TO OCCUPATION             
     Group             N         Range     10%   Q-1   Median    Q-3     90%        
Engineers             52
    Full                        116-148    121   127     133     135     140
    Verbal                      110-144    120   125     130     136     140
    Performance                 114-145    119   124     129     133     137
Professionals I       52
    Full                        113-141    120   126     132     134     137
    Verbal                      109-144    121   126     134     136     141
    Performance                  99-137    115   119     124     129     132
Educators             45
    Full                        104-141    118   123     129     134     137
    Verbal                      106-143    112   123     129     134     137
    Performance                  95-139    112   117     125     132     135
Professionals II      61
    Full                        106-143    117   123     128     133     138
    Verbal                      112-143    117   123     128     132     137
    Performance                  94-141    111   118     124     129     136
Teachers             421
    Full                         94-152    114   120     126     132     137
    Verbal                       94-145    113   120     126     131     137
    Performance                  83-161    108   115     123     130     134
Social Service        66
    Full                        108-145    117   121     125     132     135
    Verbal                      105-145    114   118     124     133     136
    Performance                  95-146    106   113     122     129     134
Managers             134
    Full                         92-146    113   120     125     130     136
    Verbal                       81-140    110   118     124     131     135
    Performance                  91-154    109   115     122     129     135
Nurses               191
    Full                         93-141    113   118     124     128     132
    Verbal                       99-143    110   115     121     127     131
    Performance                  78-145    102   114     125     130     136
Arts                  62
    Full                         74-147    109   117     124     131     138
    Verbal                       73-142    107   115     120     127     136
    Performance                  78-145    102   114     125     130     136
Sales                153
    Full                         95-142    105   112     122     128     132
    Verbal                       93-143    101   111     120     128     133
    Performance                  93-144    103   111     118     125     130
Secretaries          107
    Full                         92-135    108   114     121     125     130
    Verbal                       94-134    107   114     120     125     130
    Performance                  88-136    102   108     118     126     129
Bookkeepers           55
    Full                         99-137    105   109     117     125     129
    Verbal                       98-141    102   107     115     125     129
    Performance                  94-141    101   107     114     120     125
Clerks               128
    Full                         74-140     97   105     116     122     131
    Verbal                       76-145     99   106     117     122     127
    Performance                  69-140     94   103     114     121     129
Office Workers        62
    Full                         88-139     99   108     116     123     128
    Verbal                       89-135     98   104     113     122     127
    Performance                  81-141     97   106     115     124     128
Skilled Labor        107
    Full                         87-139    103   110     115     123     127
    Verbal                       82-136     98   106     114     120     127
    Performance                  93-136    102   108     117     122     126
Personal Service      57
    Full                         54-130     79    92     106     113     122
    Verbal                       59-132     80    90     104     115     121
    Performance                  54-129     84    99     105     114     119        

Professionals I -- physicians, dentists, lawyers
Educators -- college deans and instructors, high school and grammar school principals
Professionals II -- pharmacists, accountants
Social Service -- social workers, clergy
Managers -- small business owners, retail store managers, office managers, foremen
   business executives, small manufacturers.
Arts -- singers, dancers, musicians, actors, artists, designers, commercial artists
Sales -- all inside and outside sales people, wholesale and retail
Office Workers -- stenographers, comptometer operators, typists, receptionists,
   telephone switchboard operators
Skilled Labor -- machinists, automobile and aircraft mechanics, radio and television
   repairmen, plumbers, electricians
Personal Service -- barbers, beauticians, waiters, food handlers, soda fountain
   attendants, domestics

I included this particular reference because there were enough high level professions included in the table to make it of interest to my topic. Readers who are interested in similar scores for a very much larger range of occupations are referred to the Army General Classification Test scores from World War II. most of that data is not based on intellectually demanding occupations and is not included in this paper for that reason.

Just after World War II, Dr. G.M. Gilbert, a psychologist-psychiatrist, was given access to the prisoners who were to be put on trial for war crimes at Nuremberg. Included in his examination was a German version of the American Wechsler-Bellevue Adult Intelligence Test. The results of those tests, presented in the table below, may be found in Dr. Gilbert's book, Nuremberg Diary, and also in The Reich Manual, p. 363, by Leonard Mosley.

                    Name                          IQ
                    1. Hjalmar Schacht            143
                    2. Arthur Seyss-Inquart       141
                    3. Hermann Goering            138
                    4. Karl Doenitz               138
                    5. Franz von Papen            134
                    6. Erich Raeder               134
                    7. Dr. Hans Frank             130
                    8. Hans Fritsche              130
                    9. Baldur von Schirach        130
                   10. Joachim von Ribbentrop     129
                   11. Wilhelm Keitel             129
                   12. Albert Speer               128
                   13. Alfred Jodl                127
                   14. Alfred Rosenberg           127
                   15. Constantin von Neurath     125
                   16. Walther Funk               124
                   17. Wilhelm Frick              124
                   18. Rudolf Hess                120
                   19. Fritz Sauckel              118
                   20. Ernst Kaltenbrunner        113
                   21. Julius Streicher           106

It's interesting to speculate on what Adolf Hitler's IQ might have been. I think we're on pretty safe ground in saying that he must have been at least as bright as the top members on this list. If he hadn't been, these wolves would surely have torn him apart.

TNS accepts a WAIS score of 148 for admission, but does not accept Wechsler-Bellevue scores (Vidya #7, Nov 1979). However, if TNS did accept such scores, the admission requirement would be about 143. The W-B has a standard deviation of about 14 points. This places the median score (129) of the Nazi leaders just above the 98th percentile, and the highest scoring of the group just below TNS admission requirements.

It's also interesting to not that Dr. Gilbert was one of Terman's research subjects. He can be easily identified from the clues offered on page 365 of The Gifted Child Grows Up (5th edition, 1976).

The evidence presented so far supports two conclusions. First, that practical success in even the most intellectually demanding professions, such as scientist, professor, physician, or high level executive requires an intelligence quotient no higher than the 96th percentile (125 WAIS). Second, that even in these demanding professions, there is a broad range of ability. Some doctorates were even awarded to individuals with IQs below 100, although this was rare.

It may have occurred to the reader that the surprisingly low showings may be the result of insufficient ceiling on the tests cited. Unfortunately, this is not the case. Neither the WAIS nor the W-B were designed to measure into the stratosphere of human ability, it's true, but the means and medians cited were all more than a standard deviation below test ceilings. The AGCT has an even higher ceiling than the WAIS or the W-B. Consequently, the average IQ of 125 can't be interpreted as an artifact of the measurement process.

It may also have occurred to the reader that a moderately low minimum IQ for the professions still doesn't imply that a 160 IQ isn't a better indicator of potential success than a 125 IQ. Unfortunately, the evidence indicates that, while IQ does count for something above 125, it doesn't count for much in terms of practical success.

The strongest and most easily understood evidence for this conclusion comes from the Terman study itself. Terman divided 750 of his subjects into three groups which he called A, B, and C. The A subjects were the most successful members of his group, and the C subjects were his least successful. These evaluations were made by three judges whose primary criterion was the extent to which a subject made use of his superior intellectual ability. Professional recognition counted heavily, but earned income was discounted even in evaluating businessmen. There were 150 subjects in the A and C groups, and the remainder were in the B group.

The A group included all those subjects listed in Who's Who or American Men of Science, teachers above the rank of university instructor, those outstandingly successful in law, medicine, engineering, business, and some of those successful in literature, art, and motion pictures.

The C group was not composed exclusively of failures by any means. In fact, by ordinary standards most of these men could not be considered failures at all. Nevertheless, they were subjects that had failed to live up to their potential. Some were classed as professionals or semi-professionals, but a large number were in skilled trades, clerical and minor business positions, and some were policemen or firemen. Although there were more of the mentally and socially maladjusted among the Cs, there were also many who were well adjusted and contented. Often the difference between a C and an A was nothing more than level of aspiration.

The most important fact for our purpose is that the childhood IQs of these two groups differed by only five points. The adult differences in Concept Mastery Test score (form A) was 18.3 raw score points. Both differences are highly significant statistically, but are too small in either case to be of any practical importance. In short, there was no important difference in intelligence between Terman's most successful subjects and his least successful ones.

The ineffectiveness of IQs above 140 can also be corroborated by a host of correlation studies. Unfortunately, this data is not easily tabulated, and a summary of the results depends upon the reader's understanding of the statistical phenomena of range restriction. Rather than present a tedious mathematical analysis, it seems better to me to sketch an easily understood analogy, and then to draw the obvious parallels.

Suppose that you wished to predict how well potential basketball players would perform. Among the things that you would look for would be players' physical heights. One could say with near certainty that a player who was seven feet tall would perform better than a potential player who was only five feet tall. We would expect the correlation between players' heights and measures of player's effectiveness to be quite high, perhaps .8 or .9. This correlation is, of course, computed from the full range of potential players' heights.

Now, suppose we selected players from only the very tallest candidates. Suppose we selected only those that were at least six foot six. We would find that the seven foot player's performance, when compared with a player who was only six foot six, would not necessarily be superior. Other factors now begin to have overriding importance: who is better motivated, faster, more accurate, etc. The six foot six player could easily turn out to be the better player. Moreover, the correlation between height and player effectiveness has now shrunk to, say, .3 or .4. This shrinkage of the correlation coefficient is the mathematical consequence of shrinkage in range; this is the phenomenon of range restriction.

In a similar way, when only those at the very top of the IQ spectrum are selected out for some intellectually demanding task, the predictive power of the test will be less. If two young men both graduate from medical school at the same time, one with an IQ of 125 and an undistinguished academic record, and the other with an IQ of 160 and a brilliant record, we would have no way of knowing who would be the more successful physician ten years later. Other factors would have assumed overriding importance: things like good bedside manner, good health, connections, specialty, ambition, etc. The more gifted physician would have an edge in the comparison, but it would be a very slight edge.

In almost every investigation of IQ with real world performance, two conclusions almost always stand out. First, the correlation of IQ with training criteria (grades, credentials, teacher evaluations, etc.) is almost always higher than with performance on the job. This is another example of range restriction. Training often depends heavily on what IQ tests measure, but performance on the job also depends on personality traits, and special abilities, with IQ becoming less important.

The second finding is that IQ does correlate with performance on the job; correlations being near zero in menial jobs, and about .3 to .5 in the most intellectually demanding jobs. But even here, the strength of the correlation coefficients aren't very impressive, and other, more personal traits can be expected to be of more practical importance than measured intelligence. In fact, E.E. Ghiselli says, “The correlation between IQ and job success in a given occupation is only about .20; this should be compared with the correlation of .50 typically found between IQ and occupational attainment – taking into account different occupations.”

In short, after job training or formal education, IQs become relatively ineffective predictors of success.

For the reader who is interested in the subject of predicting job performance by means of IQ and other kinds of tests, I highly recommend E.E. Ghiselli's book, The Validity of Occupational Aptitude Tests, New York; Wiley, 1966.

The strongest, clearest, and most comprehensive summary of all these research finding can probably be found in A.R. Jensen's book, Bias in Mental Testing, 1980, p. 113.

“Although IQs are an interval scale, the practical, social, economic and career implications of different IQs most certainly do not represent equal intervals. Again, this is not a fault of the IQ scale, but is the result of personal and societal values and demands. The implications and consequences of, say, a 30-point IQ difference is more significant between IQs of 70 and 100 than between IQs of 130 and 160. The importance of a given difference depends not only on its magnitude, but on whether or not it crosses over any of the social, educational, and occupational thresholds of IQ. To be sure, these thresholds are statistical and represent only differing probabilities for individuals' falling on either side of the threshold. But the differential probabilities are not negligible. Such probabilistic thresholds of this type occur in different regions of the IQ scale, not by arbitrary convention or definition, but because of the structure of the educational and occupational systems of modern industrial societies and their correlated demands on the kind of cognitive ability measured by IQ tests.

The four socially and personally most important most important threshold regions on the IQ scale are those that differentiate with high probability between persons who, because of their level of general mental ability, can or cannot attend a regular school (about IQ 50), can or cannot master the traditional subject matter of elementary school (about IQ 75), can or cannot succeed in the academic or college preparatory curriculum through high school (about IQ 105), can or cannot graduate from an accredited four-year college with grades that would qualify for admission to a professional or graduate school (about IQ 115). beyond this, the IQ level becomes relatively unimportant in terms of ordinary occupational aspirations and criteria of success. That is not to say that there are not real differences between the intellectual capabilities represented by IQs of 115 and 150 or even between IQs of 150 and 180. But IQ differences in this upper part of the scale have far less personal implications than the thresholds just described and are generally of lesser importance for success in the popular sense than are certain traits of personality and character.

The social implications of exceptionally high ability and its interaction with the other factors that make for unusual achievements are considerably greater than the personal implications. The quality of a society's culture is highly determined by the very small fraction of its population that is most exceptionally endowed. The growth of civilization, the development of written language and of mathematics, the great religious and philosophic insights, scientific discoveries, practical inventions, industrial developments, advancements in legal and political systems, and the world's masterpieces of literature, architecture, music and painting, it seems safe to say, are attributable to a rare small proportion of the human population throughout history who undoubtedly possessed, in addition to other important qualities of talent, energy, and imagination, a high level of the essential mental ability measured by tests of intelligence.”

Terman was right; an IQ above 140 has very little significance in terms of personal success. But if Jensen's assessment is correct, and I am confident that it is, where were the historical personages that should have emerged from Terman's study?

Terman suffered badly from halo effect He not only exaggerated the importance of IQ, he also inflated test scores. All of his subjects were chosen as children of approximately 11 years of age. Their childhood IQs were reported to be 152, but Quinn McNemar reported their adult IQs to be only 134, 18 points less than their childhood scores (The Gifted Child Grows Up, p. 146). Yet adult IQs were avoided as often as possible in reporting adult achievements; childhood IQs were preferred in every case. Other investigators making use of Terman's data almost always follow this practice as well. Consequently, a myth has grown up reflecting a superiority that Terman's group did not in fact possess.

Terman also selected members of his group with IQs above 170 (average 177.7 men and 177.6 women) for special study. If we subtract the same 18 point difference from 170 that we found between childhood IQs and adult IQs in the main group, we will arrive at an estimate of 152: an estimate less than the TNS average of 156. But even 152 is an overestimate for two reasons. First, the farther a score is from the mean, the greater the regression to the mean will be: 18 points is an underestimate. Second, Terman says specifically that an IQ of 170 is found about 3 times in 10,000 of the general population (The Gifted Child Grows Up, p. 282). That's 3.43 sigma according to my tables, or about 155 IQ, before the 18 or more points are subtracted to allow for adult regression. The average TNS member is at least the equal of Terman's high group, and may very well be their superior.

Nevertheless, Terman's “Subjects of IQ 170 or Above” is the only group that I'm aware of that comes close to furnishing a picture of what TNS members are probably like.

Jensen's reference to the social implications of exceptionally high IQs suggests that intellectual achievements on an historical scale will be made only by those of very rare ability indeed. Can we find such achievements and measure the IQs of those who made them? Indeed we can. In fact, it's already been done. In the early 1950s, Doctor Anne Roe wrote a book titled The Making of a Scientist (Greenwood Press, Westport Conn, 1973) in which she investigated the biographies, personalities, and intellectual abilities of 64 of America's most eminent living scientists, some of whom were Nobel Prize winners. She found the average verbal IQ of this very eminent group to be 166 (p. 164). These scientists comprise the only research group I've come across whose intellectual abilities resemble those of the Four Sigma Society.

CODA

If IQs above 140 have little importance for personal achievement, what then is left for us? Jensen says, “The evidence is overwhelming that scholastic achievement increases linearly as a function of IQ throughout the entire range of the IQ scale…” (Bias in Mental Testing, p. 319). In other words, what's left for us is knowledge. We are the ones who learn, who understand, who discover, and who invent. That should be ample challenge for any of us.

The mathematical technique called factor analysis was invented by psychologists specifically to answer the age old question, “Is there more than one kind of intelligence?” We now know that there are two: one called fluid g, measured by culture fair tests such as the Raven Progressive Matrices or the LAIT, and another called crystallized g, measured by culture loaded tests like the Concept Mastery Test or the Miller Analogies Test. What we call g has been defined as the ability to “educe relations and correlates,” or in more everyday terms, the abilities for inductive (“relations”) and deductive (“correlates”) reasoning. Culture fair tests measure the ability to educe relations and correlates using abstract diagrams, and other material that requires only a minimum of formal learning. Culture loaded tests measure the ability to educe relations and correlates using learned and over learned material, such as vocabulary, algorithms for arithmetic or multiplication, recognition of common objects and their uses, etc. Ordinary IQ tests measure both kinds of intelligence, but not necessarily to the same degree; they are generally biased in favor of crystallized g.

The currently accepted relationship between these two kinds of ability is called the investment theory of intelligence. It says, in effect, that we are all born with a certain raw ability, or the eduction of relations and correlates, which can be measured with culture fair tests. As we get older, we “invest” this fluid g in certain kinds of judgment skills, such as those involved in doing a mathematical word problem, or parsing a sentence. When we are young, the theory goes, our formal educations are so much alike that we all invest our fluid g in much the same kinds of judgment skills. That means that our fluid intelligence and our crystallized intelligence are so similar at an early age that it's almost impossible to tell them apart. After we leave school, however, we all begin to invest our fluid g abilities in different things. Measures of fluid g and crystallized g begin to draw apart. Those that invest their fluid g in school-like activities, such as accounting or law, continue to show intellectual growth on conventional (crystallized) IQ tests. Those that put their intelligence to work in other ways, such as becoming ranchers or artists, will not show the same intellectual growth, and may even show a decline in IQ on conventional measures of intelligence.

Many years ago, Mensa was faced by a serious policy decision about the kind of intelligence that it wanted to select for. It turned out that three out of four prospective members who were selected using a culture fair test could not pass a culture loaded test. At the same time, it also turned out that three out of four prospective members who could pass a culture loaded test could not pass a culture fair test. In the end, Mensa chose to use culture loaded tests exclusively in selecting its members. Almost all other high IQ societies, with the exception of Four Sigma and Triple Nine, have followed suit. As a consequence, there are now three qualitatively different kinds of high IQ societies extant. One kind, represented by Four-Sigma and most of the membership of the Triple Nine Society, was recruited with the LAIT–a culture fair test–and is made up mostly of people gifted with fluid intelligence. A second kind, represented by Mensa, Intertel and ISPE, was recruited by more conventional tests, and is made up of those gifted primarily with crystallized intelligence. Some individuals, however, have joined or qualified for membership in both kinds of societies, and are about equally gifted with both kinds of ability. A large minority of Triple Nine members, as well as a majority of those in Prometheus, appear to belong in this category.

None of this would matter except that each kind of ability brings with it its own kind of cognitive style, its own kind of personality, and its own set of values. In fact, the contrast between persons gifted with fluid g and those gifted with crystallized g is so sharp that, with a little practice, most people find that they can learn to tell them apart at a glance. Those gifted with fluid g (LAIT) tend to be socially retiring, independent of the good opinion of others, analytical, interested in theoretical and scientific problems, and to dislike rigid systematization and routine. Those gifted primarily with crystallized g (conventional tests) tend to be sociable, quick in reactions, artistic, and to dislike logical and theoretical problems. And then there are those who are equally gifted with both kinds of ability, and tend to be mixtures of all these qualities–sometimes paradoxically so: Prometheans tend to be paradoxical.

The discovery that there are really two kinds of intelligence was made by Raymond B. Cattell in 1940, and was repeatedly confirmed in the following years. The issue of multiple intelligence, consequently, should have been considered resolved decades ago. Nevertheless, there has been a recent resurgence of interest in these theories among some younger psychometricians who either do not understand factor analysis, or simply refuse to accept its results. One such theorist is Howard Gardner.

Gardner postulates the existence of seven different intelligences: linguistic, musical, logical-mathematical, spatial, body-kinesthetic, interpersonal, and intrapersonal. He says that he is ”…convinced of the existence of an intelligence to the extent that it can be found in relative isolation in special populations (or absent in isolation in otherwise normal populations); to the extent that it may become highly developed in specific individuals or in specific cultures; and to the extent that psychometricians, experimental researchers, and/or experts in particular disciplines can posit core abilities that, in effect, define the intelligence. (Frames of Mind, p. 9.) In defense of his seven intelligences, Gardner offers evidence drawn from studies of ”…prodigies, gifted individuals, brain-damaged patients, idiot savants, normal children, normal adults, experts in different lines of work, and individuals from diverse cultures.” (ibid.) In short, he offers virtually no statistical or psychometric support for his thesis, but relies instead almost completely on a patchwork of anecdotes and idiosyncratic impressions. The most troubling aspect of Gardner's work is that his theory is at least partially testable with currently available psychological instruments, and yet he makes no effort to obtain the necessary proof. It's true that we have no test for intrapersonal intelligence or body-kinesthetic intelligence, and the only test of interpersonal intelligence available was developed for the mentally retarded (the Vineland Social Maturity Scale), but tests do exist for all the other “intelligences” Gardner postulates. Why doesn't he use them to obtain the appropriate correlations, factor analyze them, and, then show that these abilities are in fact co-equal intelligences? The obvious answer is that Gardner already knows that they aren't. He says on page 284 of Frames of Mind:

“And what of my use of the loaded term 'intelligence'? As hinted at earlier, part of the motivation for using this term is my desire to put forth a more viable model of intelligence: I seek to replace the current, largely discredited notion of intelligence as a single inherited trait (or set of traits) which can be reliably assessed through an hour-long interview or a paper and pencil test. But it should be said here as well that nothing much hangs on the particular use of this term, and I would be satisfied to substitute such phrases as 'intellectual competences,' 'thought processes,' 'cognitive capacities,' 'cognitive skills,' 'forms of knowledge,' or other cognate mentalistic terminology. What is crucial is not the label but, rather, the conception: that individuals have a number of domains of potential intellectual competence which they are in the position to develop, if they are normal and if the appropriate stimulating factors are available.”

No competent psychometrician has ever claimed that an intelligence test measured all mental abilities. No competent psychometrician has ever claimed that some of the abilities left out of intelligence tests aren't valuable. What he would claim is that an ability must meet certain other requirements before it merits being called intelligence. In the first place, it must be a mental ability, which leaves out Gardner's body-kinesthetic intelligence. In the second place, it must be an ability. This means that it must be objectively observable under standardized conditions, and that there must be objective criterion of better-worse performance. This seems to leave out Gardner's intrapersonal intelligence. How can one measure a person's capacity for self-understanding? How could you tell the difference between self-understanding and self-deception? And aren't these attributes of personality, in any event?

The most important objection that a psychometrician would offer, however, is that Gardner is attempting to jettison the criterion of “the indifference of the indicator”. This principle was enunciated by Charles E. Spearman in 1923, and says, in effect, that the specific content of an item in an intelligence test is unimportant, so long as all persons taking the test understand it. No item can be without content, of course, but the principle emphasizes that the content of an item or a test is merely the vehicle for measuring g, and is unimportant in itself. That's why a test of verbal analogies can be used to estimate an individual's mathematical ability. Or why a test of number series can be used to predict a person's ability to write poetry or solve anagrams. That’s why intelligence is conceived to be a general ability, and why it's given the symbol g. Most of Gardner's “intelligences” are content specific, and not general abilities at all. (It may seem at first glance that the existence of fluid g and crystallized g are violations of the same principle, but this is a misunderstanding. The distinction between culture fair tests and culture loaded tests is often mistakenly thought to be the same as the distinction between nonverbal tests and verbal tests. This, however, is simply not the case. Verbal items (or any other kind of item) can be used, in principle, to measure either fluid g or crystallized g, depending on how much prior knowledge is necessary to understand the item. The verbal items on the LAIT, for example, are very nearly pure measures of fluid g. They make little demand on a person's verbal knowledge, but large demands on his ability to “educe relations.”) The fact is that Gardner is little more than an IQ basher. His research on computational modules has merit and promises to be an important contribution to cognitive science, but it in no way disproves the existence of a general cognitive ability, nor does it justify his assertion that IO tests have been largely discredited. Nothing could be further from the truth. As with many other IQ bashers, he deliberately attempts to minimize the scope of what intelligence tests can do. He tries to present the picture that IQ tests can only predict school-like performance, and that none too well. The reality is that a score obtained from a conventional IQ test can be used to predict performance in a profusion of activities outside the classroom, many of them bearing only the slightest resemblance to bookish or puzzle solving behavior. As evidence for this, here is a partial list of activities (and other qualities) that are positively or negatively associated with IQ.

POSITIVE CORRELATES:

Achievement motivation

Altruism

Analytic style

Anorexia nervosa

Aptitudes: cognitive abilities; 'abstractness' of integrative complexity

Artistic preferences and abilities

Craftwork

Creativity, fluency

Dietary preferences (low-sugar, low-fat)

Educational attainment

Eminence, genius

Emotional sensitivity

Extra-curricular attainments

Field-Independence

Health, fitness, longevity

Height

Humor, sense of

Income

Interests, breadth and depth

Involvement in school activities

Leadership

Learning ability

Linguistic abilities (including spelling)

Logical abilities

Marital partner, choice of

Media preferences (newspapers, TV channels)

Memory

Migration (voluntary)

Military rank

Moral reasoning and development

Motor skills

Musical preferences and abilities

Myopia

Occupational status

Occupational success

Perceptual abilities (for briefly-presented material)

Piaget-type abilities

Practical knowledge

Psychotherapy, response to

Reading ability

Regional differences

Social skills

Socio-economic status of origin (parental)

Socio-economic status (achieved)

Sports participation

Supermarket shopping ability

Talking speed

Values, attitudes

NEGATIVE CORRELATES:

Accident-proneness

Acquiescence

Aging

Alcoholism

Authoritarianism

Conservatism (of social views)

Crime

Delinquency

Dogmatism

Hysteria vs other neurosis

Impulsivity

Infant mortality

Psychoticism

Racial prejudice

Reaction times

Smoking

Truancy

Weight height ratio, obesity

The data in these tables were obtained from studies using conventional (crystallized) intelligence tests. Comparable data for culture fair (fluid) intelligence tests is more meager, partly because culture fair tests haven't existed as long, and partly because much less practical use has been made of them. One useful goal that the Four Sigma and Triple Nine Societies might adopt would be to provide the same kind of validation studies for culture fair tests like the LAIT that already exist for conventional tests. Perhaps tests of fluid intelligence would be utilized more if we knew more about them. Four Sigma and Triple Nine are in a unique position to help provide that knowledge.

Another multiple intelligence theorist is Robert J. Sternberg of Yale University (Beyond IQ; Intelligence Applied; Conceptions of Giftedness; Practical Intelligence: Nature and Origins of Competence in the Everyday World). Like Howard Gardner, Sternberg wants to break intelligence down into its component parts; unlike Gardner, however, Sternberg may actually have discovered one or more new kinds of intelligence. He calls his new theory the triarchic theory of intelligence because, as the term suggests, he believes that he has identified three kinds of intelligence: componential, experiential and contextual.

Componential: This is intelligence as conventional IQ tests measure it. It's called componential intelligence because Sternberg found a way to analyze the thought processes involved in solving IQ test items into components and metacomponents. He not only studied how a person solves an item, but also how a person chooses the strategy he does when attempting an item. People who are good at these things have high IQs, and are especially acute at analyzing arguments, or in situations calling for critical thinking. They are the typical members of the high IQ societies.

Experiential: This is the ability to have new insights. Traditional methods of studying intelligence concentrate on what's going on inside a person's head. Sternberg's approach to insight ability focused on finding out how experience mediated one's internal, mental world, and how one's internal world changed one's experiences. When he and his graduate student Janet E. Davidson began studying insight ability, they found that nobody knew what it was because every one had assumed that it was only one thing. Sternberg and Davidson soon discovered that there are three insight abilities, which they called selective encoding, selective combination and selective comparison.

Selective encoding is the ability to focus on the really critical information in a problem. When one of Sir Alexander Fleming’s bacterial experiments was spoiled by a mold, he recognized that the mold's ability to kill the bacteria was more important than his ruined experiment. His ability to see the implications of the accident eventually lead to the development of penicillin.

Selective comparison is the ability to see an old thing in a new way, or a new thing in an old way. When the tyrant of Syracuse suspected that his goldsmith had cheated him when making a gold crown, he asked Archimedes to find out if the crown really was made of pure gold, but forbade him to destroy it in the process. Archimedes solved the problem when he suddenly realized that the water overflowing from his bathtub when he stepped into it demonstrated a method of measuring the volume, and thereby the density, of any irregularly shaped object. He saw that if the density of the crown was different from that of an equal weight of gold, then the crown had to be an alloy. He immediately leaped from his bathtub, and ran through the streets naked, yelling, “Eureka!, I've found it!” (Having insights tends to do that to people.)

Contextual: This is the ability more commonly called street smart. It's learning how to play the game, and learning how to manipulate the environment. Most definitions of intelligence include environmental adaptability in them, but ordinary IQ tests don't measure this very well. Sternberg calls this kind of ability contextual because it involves tacit learning. This is knowledge that is not explicitly expressed or taught, but is only implied or indicated. It has to be learned directly from one's environmental context. People who are good at this tend to come out on top in almost any real world situation, even if they are not especially intelligent in terms of IQ or insight. The head of General Motors or the President of the United States are good examples of people with this kind of ability. The Psychological Corporation in San Antonio, Texas, is now developing the Sternberg Multidimensional Abilities Test. It will be based completely on Sternberg's triarchic theory, and will provide measurements of all three intellectual abilities. Once published, studies using this test will quickly tell us if Sternberg's experiential and contextual abilities genuinely qualify as new intelligences. On the face of it, there's good reason to believe that his experiential (insight) ability has a good chance. It is a mental ability, it is a mental ability, and it appears to meet the principle of the indifference of the indicator; all good signs. The status of Sternberg's contextual ability is harder to evaluate, but in any event, we will soon know; factor analysis will tell the tale.

DISCUSSION

Interpreting the results of a factor analysis is a bit like attempting to read the entrails of a chicken, as the ancient Roman priests once did to discover the will of the gods. It is more difficult than actually carrying out the mathematical procedures, which are quite difficult in themselves. It takes a lot of practice, and even a skilled interpreter can easily go wrong. The trickiest part of the problem, but also the most fun, is naming the factors that the procedure reveals. Sometimes factors can't be characterized verbally at all. The safest procedure, and one often followed in the investigation of intelligence, is to assign letters to the factors discovered instead of just names. This is why the general factor is called g, and why special factors such as verbal comprehension is called v, verbal fluency called w, spatial ability called k, and so on. How does an investigator tell if he has discovered a g factor? The rule of thumb is that he has found a g when one of his factors accounts for at least twice as much variance as any other factor in the same analysis. In the case of intelligence tests, it usually turns out that one factor alone accounts for more of the variance than all the other factors combined. What is often misunderstood by laymen, and sometimes forgotten even by experts, is that all a factor analysis can do is cut up the data in a mathematically parsimonious way. In order to detect a factor, at least two of the tests in the battery must load on that factor. If there aren't two tests in a battery that load on verbal ability, for example, no verbal factor will be uncovered. That's why it took so long to discover that there were two g factors, fluid and crystallized. Conventional IQ tests measure both kinds of intelligence, but the loadings on fluid g are so small that at first it took a special trick to identify it. Once Cattell suspected its existence, he made up new tests that loaded heavily on fluid g, and used them to prove that there definitely was another form of intelligence than that measured by conventional IQ tests. A similar situation presently exists with reference to the new theories of multiple intelligence. It may be that there really is a third form of intelligence not yet confirmed simply because no test has yet been invented to measure it. As things stand now, only fluid intelligence and crystallized intelligence are definitely known to exist. I joined the high IQ societies looking for people with strong insight abilities. Instead, I found an army of logical analysts who wanted to nitpick everything to death. I really shouldn't have been surprised at this, as this was the very quality they were originally selected for. Nevertheless, I not only felt disappointed with the high IQ societies, I also felt I didn't belong in them despite my IQ. The fact is, I don't enjoy arguments of any kind, and logical puzzles bore me. What I do enjoy, more than I can say, are insight puzzles like this one:

A hunter went hunting for bear. He walked five miles east of camp, but couldn't find any game. So he walked five miles north, where he saw a bear and shot it. Then he walked five miles directly back to camp. What color was the bear?

It is precisely items of this kind that Sternberg is using to construct his test of experiential (insight) intelligence. I don’t know if his test will turn out to be a measure of a genuinely new kind of intelligence, or whether it will turn out to be a special factor like verbal fluency, and frankly I don't care. What I know for certain is that whichever way it turns out, it's of immense personal importance to me. You see, it's the source of almost all of the essays I write for the high IQ societies. (But not this one, however.) I know from personal experience that the three kinds of insights identified by Sternberg and Davidson really do exist, because I use them all the time. I can even point to specific essays I've written and tell you which kind of insight sparked it. I don't claim that my insights are profound, only that I seem to have a lot of them, and that most of my readers seem to find them interesting. I am not, of course, the only individual in the high IQ societies who writes this kind of essay, but we do seem to be spread exceedingly thin. So thin, in fact, that I really don't believe it. I think there are many more people with this “knack” in the high IQ societies than have ever appeared in the journals. I think we see so few of them because most of them realize what kind of harsh treatment new ideas receive in the journals, and don't want to run that gauntlet themselves. And to be quite candid, I can't say I blame them very much.

Source : http://www.prometheussociety.org

Did you know that Homo erectus had an IQ of about 45? This is not just a guess, but the result of an experiment carried out by Thomas Wynn, an anthropologist at the University of Colorado.

Homo erectus produced a category of stone tool that anthropologists call the Acheulean tool industry. Dr. Wynn discovered that it takes a mental age of at least seven years to learn how to reproduce an Acheulean tool. Using the classic formula

IQ = 100 x (Mental Age) / (Chronological Age)

and using 16 for chronological age, we get an IQ estimate of 44. Using 15 for chronological age, we get an IQ estimate of about 47. An IQ of 45 seems to be a good compromise.

I used to be an anthropologist, and lived for several years with an indian tribe, which gives me a perspective on IQ that no one else in the super-high IQ societies is likely to share. So let me take this opportunity to expand your vision of human intelligence.

IQ tests measure something real and something terribly important, but they do not assess all of what is called intelligence. Many important mental abilities are left out. Abilities responsible for art, music, dance, cooking, mechanical invention, clerical exactness, foreign languages, caring for a baby, defeating an enemy in war, and so on, have little connection with IQ. They have little connection because literacy and numeracy have little to do with excellence in these fields.

IQ tests are powerful predictors only in the fields in which literacy and numeracy are of central importance. These are the core abilities responsible for the creation, maintenance and progress of civilization. Without them there could be no literature, law, religion, philosophy. There could also be no mathematics, science, technology, market economy, computer science, etc. No one could have more respect for these qualities than I have, but I don't mistake them for an index of human worth. There are other mental qualities of equal worth not assessed by IQ tests.