I will write one more post of IQ before moving to other topics, but I feel that first I must discuss the research methodology, namely how can I create any contribution to this field without access neither to new empirical tests nor to all published articles for making a metastudy. IQ research uses IQ tests, twin studies, representative cohorts, skull measurements and brain scans, genome studies and so on, while a metastudy at least requires a subscription to relevant literature archives. I do not have these possibilities, nor will I invest enough to obtain them, and there seems to be a methodological fault in all these research methods: they produce contradictory results. In research methods of any field produce contradictory results, you already know that the field is not strictly speaking scientific. For instance, history produces contradictory results, which allows you to draw the unavoidable conclusion that it is not a field of science, it is a field of story telling, fiction, which can be used in propaganda. So is IQ research a field of science, or is it propaganda one way or another? I am not sure about that. Scientific results should converge to the same understanding, and this is not happening. Both sides seem to cherry pick results and avoid solving logical objections.
What I see as a logical objection to IQ research trying to show that there are racial differences in IQ? The main ones I think are that women have much smaller brains than men, about 120 g, but their average IQ is only 3.7 points (0.25 SD) below that of men. This is in contradiction with a genetic theory, which explains IQ differences between races with brain size differences. The second is that lots of effort has been spent on finding IQ genes, and 52 genes have indeed been found, but they explain only some 5% of IQ differences between European people. (As most of the IQ genome research has focused on Europeans, it is not yet possible to say much of other main races.) This leaves 95% of individual IQ differences unanswered, while 80% of adult IQ seems to be heritable by twin studies. This is clearly a logical problem in the explanation.
The opposite side, that the IQ differences are caused by the environment, has logical problems explaining the twin studies and other studies clearly demonstrating a large heritable component in IQ. Both camps seem to have a political motive.
So, what can I do? The problem of the research method is avoiding answering to these logical objections, a lack of logic, so to say. Let me digress to an IQ related test in order to illustrate what I mean. Finnish children have for many years obtained good results in PISA, while the results in TIMMS have not been so remarkable. Ever since the school system was revised, mathematics teachers in the Finnish universities have complained that freshmen in the university do not know basic mathematics and have to be taught things they should have learned in the school. As an example, one can mention a TIMMS question. It was a multiple choice question with four answering possibilities. I mention only two: the correct one is d) and the one Finnish students selected is a). The question was:
What is a correct method of calculating 1/3-1/4?
a) (1-1)/(4-3)
d) (4-3)/(4*3)
Why did the Finnish students give a) as the most common answer? Only 16% answered d). I read a discussion where one commentator, Timo, obviously a Finn, explained this phenomenon in a way that for me made lots of sense. He said that the Finnish students could calculate the answer 1/3-1/4=1/12 using the method they had been taught. It has two steps. The first step is to multiply both sides to have the same denominator, so 1/3-1/4 is changed to 4/12-3/12. The second step is to subtract the numerators, so (4-3)/12=1/12. None of the four alternatives gave this 2-step method. Indeed d) is not an intermediate step of this method as it still has 4*3 as implicit, while they have calculates it to 12. Then the students start wondering what the term correct method means, on the background that they have seldom had multiple choice questions and when they have had them, it is quite common that all answers are weird and you have to try to figure out what in the world the person posing this question was thinking. For most, the correct method means the method they were taught, not any mathematically sound method. This corresponds to an old Finnish joke of a fireman saying after a fire was put down: fire was put down in a wrong way (in Finnish, väärin sammutettu). This is the culture, correct means the way you were instructed. Probably many of them, if given the right to choose the method themselves, would calculate 1/3-1/4=0.333…-0.25=0.0833… and then try 1/10, 1/11, 1/12, that’s it, but this is a wrong method, even though it is mathematically sound. They have been told not to calculate it with decimals and guessing, though they have pocket calculators and that way they get correct results. Correct does not mean mathematically sound: it means do as the teacher wants.
So, none of the alternatives give any method (a method is do this and that), none even give an intermediate step of the correct method, and in multiple choice questions you have to guess what some idiot intended when posing the question. What an idiot might have thought is that if you replace – with * in (1-1)/(4-3) you get (1*1)/(4*3)=1/12. Is this a method? Is this illogical and mathematically unsound procedure a research method? Why not (4-3)/(4*3), which also gives 1/12? Both are correct in the sense that they give the correct result, but that is actually not the meaning of correct. And here is the most significant insight I learned from Timo. He said that the students have been taught to give the result in the simplest form. So, which one is the most simple? Of these two, and of all four possibilities, the simplest form is (1-1)/(4-3), and this explains why Finnish students chose this answer.
Was the explanation convincing? I tried to explain the logic to my wife, who is a mathematician by background. She did not buy this explanation. I also have mathematical background, but I am also Finnish. I thought it was perfectly logical. Why do I mention this? One reason is that questions in IQ tests are often of this multiple choice type and the way do decide what answer is correct is often not mathematically sound. For instance, one may ask you to make an equation from 14 = 62 without using +,-,* or /. The answer, 42=16, is not derived by sound mathematical logic. It is solved by pattern recognition without a clear rule what is the pattern to be applied. The pattern is unique to this particular problem and has nothing to do with a mathematically sound transformation of an equation to another. It is similar to, transform (1-1)/(4-3) to 1/12 with a minimal change.
Going now from arithmetic to research methods in non-mathematical field, I have the feeling of déjà vu. There is mathematical mumbo-jumbo with calculation of correlations and so on, all correctly done, so you assume the author follows mathematical logic, and then you notice that there is a well-known contradiction to the result and the author does not address it. You also know that he did not address the contradiction because he cannot, because it is an old contradiction. How does this differ from changing – to *? It is worse. If a student did not explain why he at some point changed – to *, I would consider it an error, but if he explained and the explanation showed that he knows (4-3)/(4*3)=1/12 and that (1-1)/(4-3)=0/1 is not the correct answer to 1/3-1/4 but he had a logical explanation why in this case the correct answer was d) and not a), I probably would accept it. But I would not accept a paper, which argues that brain sizes explain IQ differences, but it does not address the issue why women are almost as intelligent as men, or proposes without a good argument that the reason is that brains of the two sexes are very different. I could not accept that special explanation, because it begs the question that if there is a way to pack the brain into a smaller volume and have the same intelligence, then why men, who share almost all genes with women, do not save energy by having compressed brains,
This brings me to the conclusion that a research method I actually can use in the field of IQ studies is simply logic. That’s what seems to be missing.
Let us start from the fact that a woman brain is 120 g lighter than that of a man and there is a 3.7 IQ point difference in averages for man’s favor. That is, roughly 30 g of brain is needed for one IQ point. There is 187 g difference between the brain weights of aboriginal Australians, 1199 g, and Australians of Caucasian descent, 1386 g. The IQs are 61 and 100 respectively. This gives 4.8 g of brain per one IQ point. There is a major difference between 30 g/point and 4.8 g/point and I do not think it can be explained by differently structured brains of men and women. There are small structural differences, but for logical reasons there should not be a superior packing way, which is not used by both sexes. We have to conclude that the brain size only explains a small part of IQ and it is on the range of 30g per IQ point. The IQ difference between aboriginals and Europeans is largely genetic, but not directly caused by IQ genes, which increase the brain. Let us assume these genes only explain 6 IQ points (187/30) of the 39 point IQ difference. About 15 point can easily be environmental, leaving 18 IQ points to be explained by genes, which are not IQ genes.
What can it mean that there are genes, which raise IQ by 18 points, yet they are not IQ genes? I mean genes, which cause the individual to develop his IQ in a favorable environment. Such genes do not affect the brain size. They influence personality. Men and women of the same population have mostly same genes. Only genes in the Y chromosome are missing from women and men have a single X gene, which has effect e.g. in illnesses caused by a fragile X chromosome. Because of sex differences, men have larger brains and consequently slightly higher IQ. Most of the genes affecting personality are inherited by both sexes, while there are personality differences between sexes and they influence academic achievement.
There are several models for personality and they select different personality traits, but I will use the Big Five personality model as the five personality traits in that model have been found to be about 50% heritable. It has been shown that there are differences in the Big Five personality traits between populations [1], [5], though some differences may feel counter-intuitive.
Openness to ideas is the only one of the Big Five personality traits, which correlates with IQ also in the average. Openness to ideas has a 0.33 positive correlation with general intelligence and 0.3 with crystallized intelligence. There is a slight gender difference in openness in men’s favor while there is a similar difference in Openness to feelings favoring women. Neuroticism is more closely associated with some psychological disorders than with IQ. There is also a gender difference in neuroticism, women are more neurotic. A high level of agreeableness hinders the person from competing against more aggressive individuals and it can be a negative trait for high achievers. Women are in average more agreeable than men, which may partially explain their lower intellectual achievement on very high levels.
Two Big Five personality traits seem intuitively closely connected with intellectual achievement: introversion/extraversion and conscientiousness. There seems to be no gender differences in introversion and conscientiousness, which means that both are suitable as independent variables in the model for IQ differences. The model has now three genetic components: brain size influencing IQ genes, genes influencing introversion and genes influencing conscientiousness. It is interesting to notice that the Big Five personality traits may not be universally valid, but introversion and conscientiousness may be. In [2] Gorven et al. study Bolivian aboriginal people and find only two Big five personality traits: pro-sociality and industriousness, these are naturally introversion and conscientiousness.
IQ tests apparently have not found any average IQ difference between introverts and extraverts, while very high-IQ individuals are much more often introverts. This in understandable, since IQ develops from training and high-IQ introverts spend more time in activities that increase IQ, while normal-IQ introverts spend the time in other ways. I do not know of a similar study concerning conscientiousness, but it is obviously so that diligent students get better grades than lazy ones. It is possible to be lazy as a student and diligent in something else, so an example of a lazy but intelligent student is not a contradiction to the proposal that high-IQ individuals score higher in conscientiousness, though this difference is not seen in average IQ people. As is the case with introversion, we may assume that a higher IQ conscientious student studies diligently and increases the IQ, while a normal-IQ conscientious person does the job well, which does not increase IQ. If in the culture there is no emphasis of getting a higher education, a conscientious introvert will do something else, while if there is pressure to go to the university, the student will do so and develop the IQ.
These two personality traits can in certain cultures increase IQ. The effect does not need to be visible in the averages inside the culture, but if we measure IQ averages between different cultures, these traits may explain much of the IQ differences. Personality traits are heritable, thus the IQ differences will turn out to be heritable in IQ studies, while they can not appear without favorable environmental circumstances. We may assume that the optimal personality traits reflect the environment. Thus, in order to explain IQ differences between individuals in a population, we can use the brain size and personality traits, while to explain IQ differences between populations we can use the brain size, personality traits and the cultural differences. Let us see how it could work out.
The 39 point IQ difference between aboriginal and European Australians is then composed of several parts. Using the men-women brain size to IQ relation, the smaller brain size of aboriginals explains 6 points of the difference. Based on data from other cultures, we can estimate that improvements of education would increase the IQ by 15 point. There is 18 points to explain by effects of introversion and other personality traits, but can this be done?
First, what do we know of introversion among aborigines? [3] tells that the aborigines were actually not hunter-gatherers but rather fire-stick farmers and they practiced certain kind of animal husbandry with kangaroos and wallabies by creating feeding areas for these wild animals by burning. The aboriginal society was peaceful and introverted. If the society was introvert, we may assume the people were introverts, and this seems to be the most consistent finding of cross-cultural differences in Big Five personality types: Europeans and Americans are more extraverted than Asians and Africans. This result is mentioned in [1] and a large survey in [5] found slightly higher ratings for Western Europe and America in extraversion, agreeableness, conscientiousness, neuroticism and openness than in the rest of the world. Africans were least extraverted. Aboriginal Australians were not studied, but we may assume they are more introverted than Europeans.
The IQ in East Asia is higher than in Europe, but this can be explained by genetic mutations increasing the brain. The IQ is higher in Europe than in South Asian and Africa, and there is a brain size difference in favor of Europeans, but it explains only a small part of the IQ gap. A vast majority of the personality differences (95%) are within a culture, only 5% are between the cultures [1]. Thus, it seems difficult to explain major IQ differences between cultures by introversion. However, this is not the goal either. The goal is to explain the differences by using introversion and culture.
Traditional cultures are indeed more introverted than modern societies, but they also do not enhance IQ as much as modern societies. Introverted and conscientious people in traditional societies do the work belonging to their role well, but they do not develop their IQ any higher. It is the same situation in modern societies: introverted and conscientious people with average IQs do not develop their IQ more than others. The modern society favors extraversion, thus also introverts must behave in a more pro-social way. But when it comes to high IQ people, they have in a modern society, possibilities to develop their IQ still higher and introversion and conscientiousness becomes an asset. This leads to larger differences in IQ in the population and explains a large part of IQ differences inside the culture.
In a traditional society this cannot happen, but if a child from a traditional society get s a chance to study and go to the university, will he develop his intelligence and excel in high-IQ tasks? If he comes from East Asia, then yes. If he comes from certain other areas, the answer is usually no. There is confusion with the Big Five personality traits. Introversion does not mean the same in every culture. The questionnaires show this clearly. In [5] a person is an introvert if he is shy and talks little. That is not the correct definition. Introversion is if being in company, which is not composed of close friends, wares out the energy or gives more. One introvert reads books and thinks of intellectually demanding issues, another just sits. A survey of personality traits must be wrong if, as [5], it concludes that Italians are more introverted than Finns. They may be more fed up with people, as there are more people, but they feel alienated while spending only one week alone in a summer cottage, most Finns do not. Why these surveys are wrongly made? Partially probably because the fans of the Big Five model want to show that the model is universal, partially because they do not want to find differences between peoples. It is the same Boasian fallacy. I conclude that aboriginal Australians are in a way more introverted than Europeans, but not in the same way, and the model can describe IQ differences between populations.
The reason I conclude so is that the failure to find IQ genes after so long trying and the small difference in IQ between men and women shows, in my opinion undeniably, that IQ genes mainly influence the brain size and they explain only a small part of IQ differences in a population. Twin studies fix the heritable part of IQ firmly to 80% or so. Then there must be genes, which do not show up in genome studies of IQ since they do not actually and alone increase IQ. I have no other solution to this dilemma than that there are genes, which increase the ability to increase the IQ. Looking for traits that can lead to increase of IQ, I found only such commonly known personality traits as diligence, persistence, intellectual curiosity, reading books, solving problems. Mapping these to Big Five personality traits, which at least have 50% heritability, there pops up two: introversion and conscientiousness. I cannot see a positive contribution to IQ from agreeableness or neuroticism, while openness is just a restatement of intelligence. We cannot explain intelligence by stating that it is caused by intelligence. This leaves just the two, and the other traits that have not been well studied.
References:
[1] Antonio Terracciano, “Cross-Cultural Studies of Personality Traits and their Relevance to Psychiatry”, Epidemion Psichiatr Soc. 2006, https://www.ncbi.nlm.nih.gov/pmc/articles/PMC2756039/
[2] Michael Gurven et al., How Universal Is the Big Five? Testing the Five-Factor Model of Personality Variation Among Forager–Farmers in the Bolivian Amazon”, J Pers Soc Psych, 2014,https://www.ncbi.nlm.nih.gov/pmc/articles/PMC4104167/
[3] Francis Gordon Clarke, The History of Australia, p. 15. 1996, https://books.google.pl/books
[4] David P. Schmitt et al., “THE GEOGRAPHIC DISTRIBUTION OF BIG FIVE PERSONALITY TRAITS, Patterns and Profiles of Human Self-Description Across 56 Nations,” 2007, http://biculturalism.ucr.edu/pdfs/Schmitt%20et%20al_JCCP2007.pdf