Friday, June 12, 2009

Spent, Miller and Kanazawa

I was wandering in the bookstore today and looked through a copy of Geoffrey Miller's book Spent: Sex, Evolution and Human Behavior. (I am a cheapskate, and I read pretty fast, so I often skim through entire books at the bookstore.) If you like evolutionary psychology, you will probably enjoy the book, which is well written and covers some novel topics. One point emphasized by Miller throughout his career is that the human brain evolved not just to survive, but to survive in competition against other human brains, and particularly in the context of sexual selection. Working out the implications of this observation seems to be one of his main interests. This book concentrates on marketing, consumer behavior, signaling and related subjects.

If you hate evolutionary psychology, perhaps because you feel it's unrigorous and consists of a collection of just-so stories, then you might not like the book as much. Nevertheless, much of what Miller writes remains interesting, particularly his discussion of what he calls the Central Six personality traits -- g (or IQ), Openness, Extraversion, Stability, Agreeableness, Conscientiousness. The "general factor" g is defined as the largest principal component that arises in analysis of the correlation between performance on cognitive tests. Similarly, the Big Five factors are the largest components that arise in the analysis of personality -- it is claimed that they capture most of the variation in personality (see here for more detailed discussion of Big Five). Miller argues that marketers don't make as much use of these principal components as they could.

You can take a brief Big Five test here. My scores were (percentiles): Openness 88, Conscientiousness 94, Extraversion 89, Agreeableness 74 and Neuroticism (opposite of Stability) 1.

Looking at Spent reminded me of a debate between Miller and LSE researcher Satoshi Kanazawa I read a few years ago. Each part was published separately in the journal Evolutionary Psychology (click title for PDF).

The Asian Future of Evolutionary Psychology, by Geoffrey Miller

Abstract: Asia’s population, wealth, cognitive capital, and scientific influence are growing quickly. Reasonable demographic, economic, and psychometric projections suggest that by the mid-21st century, most of the world’s psychology will be done in Asia, by Asians. Even if evolutionary psychology wins the battles for academic respectability in the United States and European Union, if it ignores the rise of Asian psychology, it will fail to have any serious, long-term, global influence in the behavioral sciences after the current generations of researchers are dead. ...

No, It Ain't Gonna Be Like That, by Satoshi Kanazawa

Abstract: For cultural, social, and institutional reasons, Asians cannot make original contributions to basic science. I therefore doubt Miller's prediction for the Asian future of evolutionary psychology. I believe that its future will continue to be in the United States and Europe.

Asian Creativity: A Response to Satoshi Kanazawa, by Geoffrey Miller

Abstract: This article responds to Satoshi Kanazawa’s thoughtful and entertaining comments about my article concerning the Asian future of evolutionary psychology. Contra Kanazawa’s argument that Asian cultural traditions and/or character inhibit Asian scientific creativity, I review historical evidence of high Asian creativity, and psychometric evidence of high Asian intelligence (a cognitive trait) and openness to experience (a personality trait) – two key components of creativity. ...

I find this debate amusing and thought provoking, although I am not at all convinced by most of the arguments presented. One wonders how careful Miller and Kanazawa are about deriving strongly held beliefs from limited data. To what extent do priors dominate their beliefs?

One interesting thing about Miller's first essay above is that he takes a quantitative stab at estimating the population of high-g (IQ > 130) individuals in different parts of the world. You can't get more un-PC than that :-) Of course, to make this estimate he needs to make assumptions not just about average IQs by population, but standard deviations as well!


Richard Sharpe said...

I work with lots of East Asians (mostly from China, a small number from Japan, mostly all born outside the US, although some are ABCs or ABJs) and cannot say that I find them un-inventive.

Of course, since they are all highly educated engineers, I am undoubtedly looking at a very range restricted population, and given that China has something like four times as many people as the US, that still leaves lots of room for there to be plenty of inventive people, even if their variance is a little lower.

Another aspect of this debate is that the average intelligence of a population is also very important, because you can have all the geniuses in the world, but if the average intelligence of your population group precludes making effective use of what they come up with, those geniuses are not much use.

Siddharth Sharma said...

average IQ in india at 81.

thats is probably not only politically incorrect, but there is almost certainly some methodological flaw somewhere.

indians ain't so dumb i tell ya. if my high school experience is anything to go by, IQ 130 is at the very very very minimum 1% in india, not 0.06% (sheesh) .

though i am pretty sure its 2%+.

Steve Hsu said...

Richard: the issue that Kanazawa touches on is quite complex and probably involves both cultural and heritable factors (personality is somewhat heritable, although not as much as g). It seems to me that there are noticeable differences between E. Asians raised in Asia and those born here that might afffect intellectual productivity. Not in g but in personality, values, creativity, etc.

Re: India, Miller is not a dumb guy. I think he gives results for 3 different assumptions for India: 81, 90, 100. For China he uses 100 and 105.

The actual situation may be more complicated for India as there seem to be diverse subgroups. It could be the case that for some groups the average is very high, like +.5 SD whereas it is lower for other groups. It's also not clear to what extent these results are determined by environment, genetics, etc. The future of India depends quite a bit on where the reality lies.

See my posts on IITs and human capital in India.

Siddharth Sharma said...

The other 2 results are possible flynn effect projections for 2050. he does take 81 as the current norm, and i have serious doubts about 81 being the national average.

I've read your posts on how many truly 130+ performing people we have in india given the economic and literacy conditions and your analysis is generally correct.

However, i'd really have to think twice before accepting a national mean of 81 for india. Thats placing a significant chunk of the nation as near retarded. I am not buying that.

anon said...

I'd hire a European lawyer over a Mongoloid lawyer. The Mongoloids have no philosophy. "Chinese philosophy" is an oxymoron. Chinese is not a language. Mongoloids always lose on Jeopardy.

Explantion: the SD of Mongoloid verbal intelligence is half that of Europeans.

Steve Hsu said...

Anon, can you do your trolling somewhere else? BTW, you can actually look up the east asian verbal SD in the PISA or SAT data and what you wrote is just incorrect.

Carson C. Chow said...

Hey Steve,

Have you read Andrew Gelman's critque's of some of Kanazawa's recent papers? I was pretty convinced that it was mostly noise due to small sample size.


Steve Hsu said...

I don't remember if it was Gelman, but I've seen criticisms of Kanazawa's use of statistics. BTW, Kanazawa has an entertaining blog. He writes well, but never seems in doubt about the evidence supporting his theories ;-)

A. said...

Hey Steve,

I know you and I can't believe that you got a 1 out of 100 on the neuroticism scale! Either that's a typo or you lied on the test. How could you have fit into the theoretical physics community?


Steve Hsu said...

Now that you mention it, I must be more neurotic than the test says.

I think I answered the questions honestly, although perhaps I was channeling Marcus Aurelius at the time!

Try the test yourself and see what you think. Possibly the questions they use don't get at my particular neuroses :-)

OneSTDV said...

I didn't read Kanazawa's article, so I don't know if he mentioned this. But Asians tend to exhibit higher temerity relative to other racial groups. One notes that many scientific breakthroughs are iconoclastic, in that they explicitly counter traditional beliefs (epitome being Einstein's relativity countering Newtonian physics). It takes balls to go against years of traditional thought and it's possible some Asian genuises are uncomfortable with doing so. Thus, this becomes a depressive aspect to their achievement.

I saw another great paper from Kanazawa comparing average income across racial groups and found blacks actually had higher incomes at a given NSLY level.

freelikebeer said...

Openness: 47
Conscientiousness: 52
Extraversion: 31
Agreeableness: 57
Neuroticism: 27
IQ: Somewhere in the vicinity of the sum of three of those numbers

Apparently, I am pretty vanilla

Bee said...

Openness: 80%
Conscientiousness: 89%
Extroversion: 7%
Agreeableness: 57%
Neuroticism: 22%

Very proud to have such a low Neuroticism score ;-)

Carson C. Chow said...

On these types of quizzes, I think it is difficult to separate what you really feel from what you would like to feel. I've scored all over the place on them. It's also reasonably clear how you could game the quiz, especially the neuroticism part. It might be more informative if otheres scored you.

Steve Hsu said...

One of my colleagues, after reading this post, mentioned to me that I have the wrong personality for physics! (Too extroverted? Too stable? Too agreeable?)

Carson: Yes, your score could reflect how you *want* to be rather than how you really are.

It's been pointed out that the test for g is the one that is hard to game ;-)

Ben017 said...

"If you like evolutionary psychology, you will probably enjoy the book, which is well written and covers some novel topics."

I actually found out about evolutionary psychology from a friend who had downloaded an e-book by David DeAngelo (dating guru). DeAngelo recommended books like 'the Mating Mind' by Miller & 'The Red Queen' by Matt Ridley. I think other pick up artists like Mystery also draw on material from evolutionary psychology.

I'll keep an eye out for Miller's latest book.

Kanazawa's blog is very entertaining, he has some quite provocative ideas. He has some good essays on the need to pursue truth and slams those who commit the moralistic fallacy.

anon said...

Now "trolling" is disagreeing with Steve even when his opinions are ridiculous. Everything I've said is correct. But I suppose if I were Chinese I'd convince myself the truth was not the truth.

Also Steve you must explain why despite their superior mean IQ Mongoloids are conspicuously absent from the ranks of America's richest people.

BTW, Miller is "a dumb guy". Every psychologist is.

Blog Archive