January 29, 2007

"The plural of anecdote is not data"

Of course it is. What you can argue over is whether or not the data / anecdotes were properly collected, properly analyzed, properly interpreted, and so forth, but lots of anecdotes do constitute a dataset. Many well-collected, well-analyzed, and well-interpreted data began their lives as anecdotes that suggested a pattern to a curious person who then dug deeper into the issue. It's easy to generate lots of data, most of which will be worthless for a particular question. You can point any instrument at any part of the world and collect lots of data. However, what machines can't at present do is become intrigued by something that seems a bit odd and in need of explanation -- it's these initial, humble anecdote-seeds that the investigator, equipped with his tools, nourishes into a conclusion (either for or against his hunch).

Why is statistics the target of so many retardedly nihilistic sound-bites? Lemme think about that one...

January 24, 2007

Will Princeton *please* admit Jian Li and let the rest of us get on with our lives?

From the NYT, another story on anti-Asian Princeton with the by now obligatory griping of Jian Li, who claims he was rejected due to racial discrimination. Seriously, this guy got into CalTech, Cooper Union, Rutgers, and is currently attending Yale, so you'll pardon me if I don't commiserate. So he got a perfect SAT score and great grades -- unless he has also patented inventions, written Broadway plays, or something similar, his academic record doesn't make him a shoe-in at any top-tier university. Yet from the way he apparently expects every eminent school to roll out the red carpet for him, he would have us believe that he's an unrecognized genius. However, as noted by Francis Galton, who pioneered the study of genius, there's no such thing as an unrecognized genius, since we require proof in the form of accomplishment, and no one so accomplished will go unnoticed by those who matter. But again, who's to say that Li hasn't been recognized as an accomplished student -- for fuck's sake, he got into CalTech and Yale.

All I know is that this guy is going to be in for a rude awakening when he applies to grad school and discovers that his perfect 800 GRE Math score only puts him in the 94th percentile (or +1.56 SD) of GRE test-takers, compared to the 99th percentile (or +2.33 SD) an 800 SAT Math score represents. One of the better points that Charles Murray brought up in his series on intelligence and education was that smart kids need to be humbled. That doesn't mean they should wallow in self-pity, of course, but they need to know their limits in order to optimally guide their ship along its course.

Particularly among the groups of people Li is likely to associate with for the rest of his life -- academics, Google engineers, what-have-you -- an IQ of 145 is not going to be anything special. Such individuals are about 1 per 1000, and in a country where there are roughly 200 million people of working age (15-64), there are about 200,000 such people with a 145 IQ. In a high school representative of the population, whose student body is a bit over 1000, you'd expect to find 1 such person -- at every such high school across the country. I looked up info on Li's high school (Livingston High School), and the mean SAT Math score is 611, so I assume he's already experienced this phenomenon to some extent. He should get used to it.

Just remember, dude, next time you cry bloody murder because important people aren't paying enough attention to you -- having a high IQ, by itself, doesn't count for squat in getting important people's attention. You could always join the club that was invented so that high-IQ people who were bitter that their high IQ wasn't sufficient to earn them Galilean distinction would at least have something to brag about: "I'm a member of MENSA!"

Addendum: Since Steve commented, that reminded me of a recent blog entry of his on criteria and quotas. Should firefighters be required to lift 200 lbs or 250 lbs or just 150 lbs, for example? The answer to this question has implications for the percentage of females who will make the cut-off, since men and women differ in mean weight they can lift. The same is true of personality traits and interests in gaining admission to an elite college like Princeton: good grades and test scores are taken for granted, except in cases of affirmative action. Beyond that, different schools might emphasize different personality traits and interests -- Princeton may emphasize cocky, dominant son-of-a-guns who will use their brains as lawyers, politicians, and senior executives, while Berkeley may be much more tolerant of the personality traits and interests of geeks and nerds. If true, Northeast Asians will be overrepresented everywhere due to higher mean IQ, but even more so at Berkeley than at Princeton.

Independent evidence that this is likely a big factor in the different percentages of Northeast Asians at elite schools is that most of the rivalries among such schools center around the mean personality traits and interests of the other: Dartmouth is conservative and provincial, while Brown is liberal and cosmopolitan, and those affiliated with the schools use these differences to satirize the foibles of each other.

January 19, 2007

Murray on intelligence, education, and responsibility

There's a buzz in the blogosphere about three opinion articles that Charles Murray has written in the WSJ relating to intelligence, education, and responsibility. The first asks us to come to grips with the boring truth that for normally distributed traits like IQ, half of the population will be below-average, and that this will constrain our ability to have all students reach a decent level of academic achievement. The second moots a discussion on the merits of over-emphasizing a four-year college education for those who will continue on with education after high school. And the third focuses on how and why we should attend more carefully to cultivating the cognitive elite who will run society.

Last June I wrote an entry on what makes a good teacher, which pretty much covered Murray's first two points, so I'll just make a few comments on them. He's surely exaggerating, or perhaps worded his statement poorly, when he says in the first article, "If you do not have a lot of g when you enter kindergarten, you are never going to have a lot of it." Most IQ tests aren't valid when tested on people under 16, let alone on 5 year-olds, and it's really by late adolescence / early adulthood that a person's IQ reaches its adult value. Five year-olds are also not used to testing procedures and strategies, and especially in the case of young boys, their disposition is to tell the teacher to go eat it. Thus, other measures should be used to make educated guesses about who will turn out by late adolescence to have a high IQ -- perhaps elementary schools could hold Tetris competitions. Boys love competing against each other in video games, and though a crude measure, skill at Tetris must surely involve a large amount of visuospatial intelligence. For girls, indicators of verbal intelligence would better discover the smarties -- perhaps a short-story writing contest. Math skill should not enter into the process in either case.

Also, again though not a pure measure of g, creativity does depend on intelligence, so perhaps the PTA could send out questionnaires asking parents if their kid is obsessed with creative endeavors. When I was about 8 or 9, I thought I was going to be a video game designer when I grew up, so I made up a video game loosely based on Bonk's Adventure: there was a rough plot of goals and obstacles, a cast of characters, fully mapped out levels, and an instruction manual that I stapled together. OK, so it wasn't like Mozart composing symphonies at the same age, but this streak of mine was clearly a better predictor of my adult GRE scores than the SAT score I got at age 12. I don't attribute the failure of the latter just to the fact that I couldn't have been expected to know high school math and vocabulary at that age, but also to the fact that I hadn't taken a standardized test before, and that my personality programs me to hate and rebel against such testing.

Now, if a 12 year-old does score highly on the SAT, that can't be due to chance, but we can't infer much from their failure to score highly on the SAT or other IQ tests that have a substantial loading on crystallized knowledge such as mathematics, vocabulary, and academic trivia (like world capitals). The Ravens Matrices, which are pure pattern-recognition tests, should be used instead, though if the kid hates taking pointless tests, again simply noting whether your son is a wiz at Tetris, or whether your daughter writes short stories, could be sufficient information to make a good enough of a guess that they'll be more suited to honors or gifted classes once they begin secondary schooling.

Turning to the second article, Murray is spot-on when he notes that many who attend four-year colleges would be better served by vocational schools. Incidentally, I was thinking not so much of the "paralegal" or "craftsman" type of vocational school, but of various Master's degree programs that are increasingly popular (the ones advertised in subway cars, for instance). Unlike doctoral and professional programs like those leading to a J.D., most of these Master's programs don't depend on four years of undergraduate study. Think of what a waste it is to major in education or communications: since these are not rigorous fields where undergrads learn lots of math in preparation for graduate studies, or where Classicists must memorize the history of civilizations and master dead languages, interested students should be able to skip straight to MA programs, study their subject intensively for a year or two, accrue far less debt, and enter the workforce four years earlier.

Cheerleaders of the four-year liberal arts education for the majority of post-secondary students would claim that, even if a graduate of such a program is ill-equipped to enter the workforce, at least he'll have learned the art of critical analysis and persuasive exposition. But they would be wrong. Remember, we're only talking about graduates from non-rigorous programs -- for example, English literature or political science -- where rational thinking and cogent presentation have been dethroned in favor of allegiance to theoretical fashion norms, obscurantist gobbledygook, mealy-mouthed "where is the thesis?" theses, and prescriptions founded on "assume a can-opener" premises. Such undergraduate programs -- among the most popular, mind you -- should be viewed not as an enrichment of the mind but rather as a clinically diagnosed form of brain trauma, and most students would be better off never setting foot in the classroom of the average English or PoliSci professor.

Lastly, Murray's third article advocates "a revival of the classical definition of a liberal education, serving its classic purpose: to prepare an elite to do its duty." I quite agree with him that we should emphasize to the cognitively gifted that they should consider themselves fortunate for having been blessed by the actions of their genes and good luck in development, which counteracts their inclination to believe they're smarter than average because they worked so hard, while others are just lazy or don't value education as much as they do. This is the secular version of a privileged person looking upon a beggar, thinking "There but for the grace of God go I," and feeling obliged to help them out of a sense of empathy. The more we obscure the central role of general intelligence in attaining high social status, and the central role of genes and lucky development in attaining high intelligence, the more likely we are to react callously to the financial hardships of the cognitively short-changed. We're likely to say, "They brought it upon themselves," or "Why don't they just go to college and make something of themselves like I did?"

However, I'm less optimistic than Murray seems with respect to how far this cultivation of responsibility will go. Restricting our attention only to those with IQs in the top 10% (for argument's sake) and the jobs they will fill as adults, as long as there is variation in personality traits among these smarties, as well as variation in the personality traits selected for in the different professions, I think it likely that those in positions of power will continue to be selfish bastards. The retiring types will shuffle themselves into a career in computer programming or accounting, while sons-of-bitches will beat a trampling path toward a career in social control, either of the direct variety (politics, executive-level management) or indirect variety (PR, advertising, and other "managers of public opinion"). Being sons-of-bitches, they will be nearly impossible to persuade to rule in the public interest. Most progress in this area of social life has been made by devising restraints on their desire to screw others over in order to enhance their power and wealth even more (checks and balances, limited terms served, election by a constituency, and so on).

It should be easier to instill a sense of responsibility in other groups of smarties -- say, engineers (I'm not an engineer). Most people will tell you that engineers and other MIT / CalTech geeks are basically amoral: they don't care much about others, but they're not power-hungry megalomaniacs. Moreover, they are in a much better position to positively change society -- who invented the wheel, air conditioning, the internet, hygiene products, nuclear & solar power, and perhaps before long genetic enhancement? These two facts are far more promising initial conditions for someone who wants to take up Murray's task. Consider just the last example I gave, that of genetic enhancement: strangely, this is the only method of exploiting our biology for the better that many object to. No one cries foul when we research which vitamins, minerals, and other foodstuffs are more likely to give us better health and longer lives. When we take prescription drugs to soothe arthritis or ease the strain on our hearts, again not a peep from the bioethicists. Ditto for administering vaccines or antibiotics to combat infectious disease. Or, for that matter, brushing our teeth and seeing the dentist twice a year.

Yet no other improvement of our biology has the power to render obsolete our worries about inequality than genetic enhancement. If the population were close enough to monomorphic for genes related to IQ, then Murray's entire first article would go out the window. Not being a geneticist, I don't know how close this is to becoming a reality for most people. The point remains, though: it is a hard, but not intractable problem to solve with genetics, while it is an impossible problem to solve with education, whose effects are limited by the pupil's underlying biology. At the same time, that day is not tomorrow, so until then we must face the issues that Murray has brought up. In my next post, I'll explore these questions in more depth for a particular case: that of math education.

January 17, 2007

Lying with statistics

I just hate that phrase. HATE. It makes it seem like the use of statistics is particularly suited to bamboozling people* -- this despite the fact that most people don't remember middle or high school math after they graduate (or don't, as the case may be). In a bastard's hands, sure, statistics can deceive (though it's hard to actually "lie" with them). But then, in a whistle-blower's hands, they can unmask unjustices, such as how an executive's golden parachute raised the average earnings (or whatever the proper finance term is) at his company for that year. We thus return to a boring truism: statistics don't deceive people -- people deceive people. We root out deception by cracking down on human mendacity, not by regarding "mere statistics" as the charlatan's art, as in the popular, smug dismissal that "you can prove anything with statistics."

In the end, though, most deception in the world doesn't involve the use of numbers -- again, if the audience can't comprehend the myth, they can't believe in it, and most people want to involve numbers in their daily lives like they want to sit in the dentist's chair every morning and evening. It will always be easier to bullshit using words rather than numbers, since our species' evolved method of communication is linguistic -- basic numeracy has not been widespread for more than probably 100 years (if that), so there is no evolved knack for lying with numbers, whereas there is a wealth of tricks wired into the inveterate huckster. Therefore, intellectual self-defense really should focus more on taking apart verbal arguments.

That said, you'll still come across bad statistical arguments more frequently than you'd like to, and so Mark Chu-Carroll from the ScienceBlog Good Math, Bad Math has started a nice "back to basics" series on statistics. So far, there are entries on mean, median, and mode and the normal distribution.

* A joke among topologists is their self-description as those who can't tell a coffee cup apart from a donut, and yet based on their wacky wisdom, we hardly conclude that "there are lies, damned lies, and topology."

January 10, 2007

Stacy's mom -- dream on!

OK, having criticized parts of the Psychology Today article in the previous post, maybe I should write about one of the better parts. An interesting feature (not a bug) of the male mind is that it tends to set a rather low threshold for inferring romantic or sexual interest from females. You drop something on the street, a girl picks it up for you, and boom -- she must have the hots for you, using this kind gesture as her "in." This results in males making many more sexually motivated advances than if their minds had a higher threshold for reacting, and so at least probabilistically this feature will result in more mating opportunities and thus more kids. It doesn't matter, therefore, how foolish it seems in a particular instance, nor if the numerous false positives leave the bitter taste of chronic rejection in their mouths.

We can examine when this phenomenon is at it's strongest -- when guys appear the most clueless and read the most into female "signals" -- to see during what age range males tend to fall in love, try to woo females, and start a pair-bond or sleep around. Further, we can examine when guys are most reasonable about who is out of their league, and thus not likely interested in them, to see during what age this phenomenon is not as important. I'm not suggesting that at one age guys think every girl wants them, and at another age they're perfectly realistic; I'm just comparing the "you've gotta be nuts" value across the post-pubescent lifespan.

As for when this value is at its greatest, consider the following music video for "Stacy's Mom" by Fountains of Wayne (the lyrics are here):

The guy is supposed to be 12 or 13, thus barely having completed puberty, and yet he's convinced that a red-hot, divorced businesswoman in her mid-30s is into him. Compare this with the scene in American Beauty where Kevin Spacey's character thinks the blonde high school cheerleader is putting on a private performance for him. Both show how eager males are to delude themselves, but I know which age-typical delusion I think is more ridiculously unlikely than the other one. Kevin Spacey's character is older, taller, higher-status, and athletic -- traits which are at least in the right direction, even if he's too old for the seduction to work in general. But the boy in the "Stacy's Mom" video is decades younger, probably at least a half-foot shorter, and remarkably lower in status than the female he's interested in, so he's doomed from the start. It would appear, then, that the low threshold for reacting to female "signals" is lowest during adolescence.

And to return to a parenthetical remark I made in the previous post, the low threshold of activation doesn't pertain only to the mind but to the groin as well. Boys covering that area with their notebooks or a jacket around the waist is a common trope in candid portrayals of adolescent male life. Of course, this uber-trigger-happy phase subsides somewhat during one's 20s, and toward the end of middle age a man's sense of ease is no longer held hostage by the defiant will of his junk. Indeed, roughly 30% of 50-80 year-olds are impotent. Again, the obvious inference is that falling in love, trying to woo females, and consummating a relationship belong mostly to 15-25 year-olds, to a lesser extent to those up through 30-35, and not very much to those over 40, at which time guys realize that things have gotten as good as they're going to get. Interventions can alter this timeline, as when women don't have children, buy wonder bras, and exercise so that they're less unattractive in their 40s than otherwise; or when men use Viagra and enter careers that result in peaking status-wise at 40 or 50. But it's still clear what age range we should ideally be studying to understand how the psychology of romantic love works.

You should be raising kids, not still dating

Via ScienceBlogs. There's a news article in Psychology Today on how wacky love appears, but how understandable these quirks are when we investigate it from the point-of-view of evolutionary biology. Overall it's a good article, but there are still some points that need to be cleared up if we really want to understand how love works.

To begin with, there are some rather bizarre phrases in the lead paragraph on the sense of nervousness during a first date:

Seated at a well-appointed table, you mull the choice between crab cakes and seared tuna, but truly you are sorting through a mental repertoire of wisecracks and war stories. If you are secure in your improvisational charms, you might use this moment to appraise the cleavage or cufflinks of the woman or man across the table. [...] And you need to know whether your dining companion is up to the task.

Anyone who has the option of seared tuna on a first date, whose dining companion might be expected to wear cufflinks -- and for that matter anyone who would refer to their date as a "dining companion" -- is, in the eyes of evolution, not very important as far as romantic love and reproduction are concerned. I imagine the people described above are at least in their 30s, and since human generation time is between 20 and 25 years, each should have already had a child so far, and that at least 5 years ago. My first thought from an evolutionary POV is: Why are 35 year-olds still going out on first dates? The male could continue to reproduce for a few decades, but the female above is nearing the end as far as reliably conceiving a child. This isn't nitpicking: studying life among well-to-do 30 and 40-somethings after the Demographic Transition is not the best way to infer what pressures gave rise to romantic love and its quirks during human evolution.

Since we're mostly adjusted to our post-pubertal existence by about age 15, and again bearing in mind the 20 to 25-year generation time, we would do best to study individuals from age 15 to 25. Just as it would be silly to study jumping ability among 40-somethings, so it would be to study romantic love and mating preferences among them. Handling relationship conflict and parent-offspring conflict, rearing of grandchildren, and so on -- sure, but it's clear that the group to study for love & mating is 15-25 year-olds. We get uneasy thinking about this fact in the context of the 15-17 year-olds because they're minors, but millennia of human evolution didn't anticipate 20th C. legal norms. It's not as if we'd be giving them the franchise, just studying what they think and how they act.

Not to beat a dead horse, but I am still baffled when adults forget what it was like being a teenager. Perhaps I'm immune to this process because I work with adolescents every day, and am thus (sometimes painfully) reminded of what middle and high school was like. Remember how it felt like your mind was under siege, commandeered by your romantic emotions for that one guy or girl? How you wouldn't get just nervous around the girl or guy you were stuck on, but got the worst case of butterflies in the stomach ever? Or how girls were so paralyzed by anxiety that they had their friends ask a boy out for them? (And guys, when was the last time you had to hold a notebook over your crotch to conceal an erection you got simply from hearing your crush call your name?) This irrationality of romantic love that renders its victims powerless stands in stark contrast to the pragmatic and calculated quality that adult-age relationships begin to assume.

Returning to the news article, the lead paragraph likens an adult first date to a job interview -- how exquisitely romantic! Now, in honesty, this is of course the best metaphor to choose, but that only shows how astray we will be lead if we analyze the thoughts and behaviors of 30 and 40-somethings to best understand romantic love. And although adolescent minors are routinely studied to see if they're doing drugs or committing crimes, they're rarely thought of as ideal subjects to study for normal behavior like falling head-over-heels in love and expressing preferences in a potential mate.

In the final section of the article, there is some real misunderstanding about the thoughts of 15-25 year-olds. First, though, there is a good point made about the effects of socializing adolescents around only other adolescents as we do by sending them to middle and high school, rather than raising them as part of the larger adult community. Evolutionary psychologist Geoffrey Miller believes this makes girls more insecure than they would otherwise be, since they're forming self-images based on their rank within a population that is much more young and healthy than the general population. If they spent most of their day instead around 30 and 40-somethings, they'd realize that they're actually in the right-tail of the fertility distribution. Conversely, he believes boys have overinflated egos because their self-image is formed from comparing themselves with other high schoolers, not with bankers and doctors.

The point about adolescent girls being more insecure than otherwise is well taken: we know that guys of all ages focus primarily on female attractiveness, which pretty much tracks health, youth, and well-maintained secondary sex characteristics. The point about boys is wrong, though. Here, Miller picks on jocks thusly:

"Young men who were captains of the football team graduate thinking they're God's gift to women, and women respond, 'I'm interested in corporate attorneys and well-cited professors. Who the hell are you?' "

Ha, nice try. The assumption is that adolescent jocks are the right-tail of a low-mean distribution, like the not-so-ugly girl in a trailer park. Hence, were they pooled with adult males who hold steady, sometimes high-powered jobs, they'd be put in their place, much as if the trailer park princess had grown up surrounded by Brazilian models. But for this to work, females must value the same qualities from adolesence through adulthood, which of course they don't. They do gauge a guy's rank in the status hierarchy, but what criteria determine a guy's rank shift from an emphasis on good looks, athletics, and whatever else makes you one of the popular boys during high school and college, to a greater emphasis on wealth during adulthood (at roughly 25, by which time her first kid should have been born). The natural inference is that girls first prefer to snag and mate with a top physical specimen while they're at their fertility peak in their early 20s, while afterwards they seek a long-term husband to invest in her and her children materially (preferably the same person, but that's difficult).

Again, how adults ever convinced themselves that adolescent girls dream of boys who exemplify intelligence, hard work, and high income, I'll never understand. I realize that most adult males, including me, were not popular dudes in high school, and so may have erased that part of their memory. But if the goal is to understand how things work, we'd better recover those memories or at least look at how current high school and college students behave. A quick reality check shows that East Asian males are the most likely to exemplify the above three traits during high school and college, yet they're the least sought-after group as mates, and ditto for their non-Asian counterparts. Looking at who the winners are (regardless of race), we again conclude that good looks, athleticism (as distinct from "good looking" traits such as a masculine jaw), and faux danger / excitement are better predictors of stud status in this age group.

The notion that younger girls yearn more for college professor types than jock types can be put to a simple test by looking at the behavior of college students. Having an IQ of at least 145, tireless ambition, and more wealth than any boy at the college won't make the professor a stud unless he's also good-looking, suave, and conveys a rule-breaker persona. But that just proves the point: this is just a jock or rockstar who happens to teach. Bear in mind that it's not simply the rules against student-teacher relationships that prevent this from happening: a girl only has four profs at one time, and many colleges are situated in college towns, where she could visit a neighboring institution to find an unforbidden professor. Another quick reality check is to look at where college girls prefer to go for Spring Break or summer vacation when they have no responsibilities and are on the prowl for mates. College towns, packed with profs though they are, always seem to empty out, while Cancun and Rio pullulate with coeds -- imagine that! Nor do they flock to Wall Street bars to hit on corporate attorneys and investment bankers. So the suggestion that adolescent jocks will become humbled by being pooled with professional adults is off the mark.

However, the article does end with some good observations on personality traits:

Teens are often equally clueless about the character strengths that make for a good partner. It takes a few years of experimental hookups and baffling breakups to learn the value of conscientiousness, trustworthiness, and emotional stability.

Still, no personality type makes for a superior mate. Context, not character, is destiny. The extroverted dervish may have an exhausting aversion to downtime. And chances are you're not the only person drawn to a woman with an operatic ability to connect. Because they're highly sought after, extroverts tend to have more affairs and end relationships more often, reports Daniel Nettle of the University of Newcastle. An agreeable man may be a helpmate, power-listener, and faithful husband, but he is also less likely to be an alpha earner than is his hard-charging, narcissistic brother.

The first paragraph above needs some clarification: clearly a lack of conscientiousness, trustworthiness, or emotional stability are not bad things as far as passing along your genes goes. If they were, they would've been weeded out long ago, and we'd all be model citizens. Of course, this doesn't mean that pairing up with someone who's opportunistic rather than deliberate, or scheming rather than trustworthy, will make us feel good. So this falls under the category of learning how to avoid the pitfalls of going with your gut. I'm not so sure about emotional stability (or low-Neuroticism) being such a bad thing, though. Combined with high Extraversion, this results in a hotheaded personality, and I know I'm not the only guy who has a soft spot for cocky, mock-confrontational girls. Just look at any "lad mag" like Maxim, or this commercial for the Xyience energy drink (TV-safe, but perhaps not work-safe). How are you going to engage in flirtatious repartee if she's difficult to excite and provoke? Plus, when I rant about someone cutting me off (or what have you), I want her to understand and assist me in cursing the guy out, rather than think my mind was outta whack for getting worked up over such a "trifling" event.

Also, the remark about Extraverts having more affairs and ending relationships earlier conflicts with other studies I've read which suggest that it's really Psychoticism (or in Big Five terms, the low-Agreeableness by low-Conscientiousness interaction) that correlates with relationship infidelities of various sorts. But in any event, the really important point is that there's not one value of a particular personality trait to rule them all. We can see that by the moderate heritabilities of personality traits -- if being only extraverted paid off, directional selection would use up the existing genetic variance in the trait, and there would be no introverts (or, they would only pop up due to random mutations). I referred to this point in a recent post at GNXP regarding male height.

Unfortunately for those promoting the sexiness of intelligence, the same reasoning applies here as well, since intelligence has a moderate heritability. Moving the goalposts, we might say that that's true, but what brought general intelligence distributions to their current state, with means much higher than when humans first appeared on the scene, is sexual selection. There is a hint of this too in the news article, to the effect that we may have evolved higher levels of g in order to manage stormy, infidelity-prone relationships. Better pattern-discrimination would certainly help do this: Wait a minute, the soap bar is on the left side of the sink -- and we're both right-handed! However, that makes the prediction that populations where relationship chaos is most rampant should have higher mean IQ than those where it is less prevalent. Concretely, sub-Saharan Africans should show higher mean IQ than Europeans, and Asians should show the lowest mean IQ. That's the opposite pattern of reality. The conclusion is that, even if this variable accounts for some variance, it's apparently tiny enough not to worry much about.

A much stronger relationship, at least at first blush, is with level of societal complexity -- especially after the transition to agriculture, a more complicated existence could easily require higher IQ just to break even. Indeed, a recent paper on the evolution of higher IQ among Ashkenazi Jews posits as selection pressure the greater complexity of the managerial niche that they filled for over a millennium. This may not be the only variable accounting for non-trivial variance in group IQ differences, but it at least fits the data properly, and the ultimate cause is reasonable enough.

How can it be that a simple news article that has many good points has managed to get me so worked up? Well, first, remember you're talking to someone with a pretty on-edge personality. But more seriously, there are enough off-the-mark aspects that are annoyingly common in evolutionary psychology, especially the popular presentation of it, that it's worth addressing them at length.

January 8, 2007

Vindication of my Euro-centric girl strategy

Razib's "hot sci-fi geek" thread is still getting comments, the most recent of which is from a European guy noting that a girl who American guys offered as an example of a good-looking chess geek (here) would not qualify as that good-looking in her continent of origin.

This is confirmation of my previous advice to look for a girlfriend among Europeans raised in Europe. They're better looking in most countries (Swedes for blonde-lovers, Greeks for callipygiophiles, etc.). Because the females have grown up surrounded by better looking people than in the US, a female who is objectively a 7 will subjectively view herself as less than 7, and so will be less discriminating in her tastes / standards than a 7 in the US. And since the Euro males have grown accustomed to a higher standard of female looks, an objective 7 female will get less attention there than if she were raised in the US. Thus, regardless of the guy's objective score, he'll find a more receptive audience among 7s in Europe than in the US. This feature of human psychology -- that we are programmed to adjust our preferences to accord with our percentile rank in the local population, rather than our absolute score -- is obviously adaptive, but it can be exploited in the modern age where global travel isn't difficult or prohibitively expensive.

The worst group to approach, strategy-wise, is those who are objectively high-scoring but who grew up surrounded by lots of unattractive people. For example, those in the right-tail of a distribution with a low mean (Queen of the Trailer Park / Band Geeks), and particularly those in the right-most cluster of a distribution that approaches a bi-modality of haves and have-nots as far as sexiness goes (see next paragraph). This makes the contrast stand out more, boosting their ego. The same is true of intellectual snobbery: kids who went to secondary school in a high-scoring, unimodal environment should think less of themselves than their intellectual counterparts in a public school magnet program, an oasis surrounded by a desert of mediocrity, as they would describe it.

For example, females from the better-looking ethinic groups who are reared in the US often have a special name: X-American Princess, for X = Jewish, Italian, Persian, Indian. Italian-American girls grow up recognizing that their mean attractiveness is beyond that of girls of other ethnic groups, such as English-Americans. They find this out not only through their own self-other comparisons, but through the amount of attention they get. Back in Italy, there's not so much hetereogeneity, so they wouldn't get as much attention from boys growing up. This creates a huge split between the attitudes of Italian-Italian and Italian-American girls, which anyone who's visited anywhere in Italy for more than a few days will have noticed. Admittedly, another component of the "princess" phenomenon is social status, but Persians and Italians are happy to boast as well about how good-looking their co-ethnics are.

South Asia, though, is heavily stratified by class, and thus by looks as well, as the higher-status males choose the better-looking females for their wives. (Over time, this cross-assortative mating results in there being a greater fraction of females who are both brainy and beautiful as well as a greater fraction who are unattractive and slow-witted.) So, the prediction is that there would be less of a chasm between Indian-Indians and Indian-Americans compared to the Italian case. This doesn't mean there would be no difference between Indians in India vs the US -- in my personal experience, those raised in South Asia have better manners and are more publicly agreeable (of course, I don't know how they behave or what they think in private). Now, these were international students I met during my undergraduate years, and so were drawn from the super-elite in South Asia, as no financial aid was given to international students, but South Asian-Americans who I run into also tend to be high-status. Attractive females in both locations are predicted to be highly conscious of how much better-looking they are compared to other females, just perhaps a little more so in the US. But since I've never actually been to India, the heterogeneity could be far greater than I imagine, and attractive Indians in India could have more inflated egos -- I welcome any clarification in this matter.

In any event, though it may sound like you're preying on the insecurities of another, my advice is not to create further insecurities or manipulate others -- it is simply to take advantage of auspicious circumstances. In this case, it means approaching girls who were raised in a population whose attractiveness distribution is more homogenous and higher in the mean than your own.

January 2, 2007

More on adolescents gone wild

Via Half-Sigma. There's an opinion letter in the NYT that expresses the writer's shock that middle school girls in a talent show had performed so sluttily (not his words, but that's what he meant), though it's not that shocking. He also voices concern that such performances make girls into "one-dimensional" sex objects, which is not true. This is somewhat of a continuation of the previous news article about high school principals cracking down on sultry moves at school dances.

Starting first with the worry about turning girls solely into sex objects, he says that the involved parents "allow the culture of boy-toy sexuality to bore unchecked into their little ones' ears and eyeballs, displacing their nimble and growing brains and impoverishing the sense of wider possibilities in life." This despite his admission earlier in the paragraph that these parents "dote on and hover over their children, micromanaging their appointments and shielding them in helmets, kneepads and thick layers of S.U.V. steel." Suburban super-parents are more likely than any other group to encourage their daughters to play sports and to bully them into doing all of their homework so they can go to college and get set in a comfortable career. So he's being disingenuous when he says, "it's a cramped vision of girlhood that enshrines sexual allure as the best or only form of power and esteem." This is a non-starter.

And as for his shock that middle schoolers could dance so provocatively, coupled with his wish that the innocent period of their lives didn't come to an abrupt end in secondary school, recall from my previous post on this topic that, whatever the increase in exhibitionism, this doesn't translate into increased sexual activity. Thus the time at which innocence ends has stayed pretty constant over the last generation or so, not to speak of most generations during our species' existence. We've recently stopped living in societies where young adolescent males would have learned how to kill one another in battle, or perhaps join their parents in the drudgery of farm maintenance, so it's easy to slip into thinking that human nature is designed to remain innocent far beyond age 12. Like all facets of mature existence, we grow into these traits gradually -- a boy apprentices in fashioning spearheads and practices "killing" trees before actually becoming a member of a war party. Similarly, 12 year-old girls practice dolling themselves up so that they've got the basics down when it comes time to find a boyfriend in a few years. Therefore, one can't really be that shocked to see middle schoolers performing provocative dances. Now, this is separate from the matter of whether we should allow it or not -- just because human nature is designed to enter a transition from naivete at around age 12 doesn't imply that we should or should not allow this to happen.

Also, it needs emphasizing that the writer's rosy retrospection, like most cognitive biases, gets in the way of clear thinking. He laments that the talent show "was an official function at a public school, a milieu that in another time or universe might have seen children singing folk ballads, say, or reciting the Gettysburg Address." Admittedly, I didn't grow up in one of these alternate universes, so I don't know if they exist or not, but if they do, again they are surely very recent in the history of homo sapiens and would represent a temporary deviation from the norm. If this were true, then his lament would be that our fleeting experiment in suppressing adolescent sexuality had ended, and we had returned to the crude way that things were before.

I realize that I harp on this point a lot in my posts, but "recent fall from perfection" stories only lead us down ill-conceived utopian paths -- if only we tinkered with this or that facet of society, we'd regain perfection! That assumes that our nature is pretty much perfect, and so "wants" to be in a state of perfection; once we reached that level, we would come to equilibrium. The reality is not entirely the opposite, but opposite enough that it matters for the decisions we make about how to structure society. If we led enjoyable existences for so long and have only recently wandered in the wrong direction, then it will be easy to carom back onto our original course. If, on the contrary, our steering wheel is rigged to deviate from a particular path, then it will require much more ingenuity and vigilance to go where we want to go. That's why priests struggle to resist worldly vices, why teachers and parents must badger most teenagers into doing all of their homework, and why we've established the rule of law.

Then the writer describes the present trend as "society's march toward eroticized adolescence." Ah c'mon, what other period of our lives will end up "eroticized?" Surely not pre-puberty, nor old age. But even by age 35 or 40, you're pretty much done with the stormy, passionate portion of your life. No more falling uncontrollably, obsessively in love. No more sexual abandon; instead, you get the sex lives of married people that are the perennial favorites of stand-up comics. Sexual passion becomes so vanishing that those few who care to attempt resuscitating their expired patient do so by means of sex toys, KY warming personal lubricants, dress-up / role-playing, and so forth. You'll never see any of these catching on among adolescents because irrational passion comes naturally at that age.

Nature's intended timeline of sexual behavior goes something like this: 12-14, becoming comfortable and adjusted to hormonal changes; 15-19, explore the opposite sex; 20-24, pick best mate encountered so far as parent of your children; 25-death, raise children & grandchildren. Obviously, individuals in the 15-24 range will "eroticize" their lives whether anyone else wants them to or not, in the sense that they'll fall madly in love with members of the opposite sex, try to join them in a relationship, and become physically intimate to some undetermined degree. Someone may argue that teenage behavior shouldn't start any earlier, such as at age 12, but that's a quibble. It would only be a real objection if the person thought that the erotic, passionate period of one's life began at age 25 and lasted well into middle age.

That's why, however shocking a group of hip-gyrating 12 year-olds is, it's even more odd to see a group of 35-40 year-old women wearing push-up bras and jeans that trace the curves of their bodies. And hence the term "MILF" -- it has its own special term because it's unusual to encounter a woman past a certain age who still possesses youthful sexuality, passion, and first-rate looks. I'm not saying women should start dressing frumpily at 35; I'm simply noting that, given human life history, it's odd that women whose reproductive potential is pretty much spent are trying to attract sexual attention as if they were still 23. On that note (also via Half-Sigma), there is a recent NYT article on the increase of female porno stars who are over age 40. Granted, this is not on the same scale as middle school provocativeness, but it does reflect the wider, increasing acceptance of middle-age sexuality (e.g., Dr. Ruth talking to married couples about anal sex on The Tonight Show). Conditioning our expectation on the entire evolutionary history of our species -- hell, even just the last 200 years -- we would surely find middle-age hypersexuality more shocking than exhibitionism among middle schoolers.

The writer argues that "There is no reason adulthood should be a low plateau we all clamber onto around age 10;" recall, though, that the average age of middle schoolers is 12-13. He's also mistaking adolescence for adulthood, and must be trying to recapture his youth by suggesting (with "we all") that adolescents are adults, as if his current existence were as full of sturm und drang as that of a teenager or college student. The middle school dancers he saw weren't interested in bemoaning the bland romantic life that attends marriage, in working a dead-end job 50 hours a week, or in resolving a scheduling conflict between their appointments at the podiatrist's and chiropractor's offices. Rather, they wanted to join the ranks of adolescents: the older kids at their middle school, high schoolers, and college kids. Exhibitionism, social drama, lack of most responsibilities -- that's what they sought, not adulthood.

So, the writer's simply mistaken to fret over the girls' future as multi-faceted beings just because they strutted their stuff in a talent show, and his shock is an overstatement. Girls performing provocative dances at age 12 instead of 15 may be slightly odd from a biological view, but the truly bizarre social pattern is females whose age suggests they would be rearing children but who are instead remaining unattached, dressing in form-fitting jeans and thongs, and dithering away their precious few remaining days of sex appeal, as if they were eternal teenagers. That one would "eroticize" life after 35 is what's really shocking -- especially considering all the sordid little devices like toys and lubricants that are now required -- since humans tend to be in child-rearing mode at this age.

This is a variation on a point Steven Pinker made in his answer to the 2007 Edge Question. On the causes of the decline of violence over historical time: "What went right? No one knows, possibly because we have been asking the wrong question -- 'Why is there war?' instead of 'Why is there peace?' " Ignorance of evolutionary and recorded history leads us to ask stupid questions. Similarly, it's not puzzling to observe that adolescence, including the preparatory years of 12-14, is "eroticized" while life past 35 is not. What does appear enigmatic is not only that adolescent sexuality is discouraged, but that at the same time middle-aged people so sexualize their lives that ads for boner pills on daytime public TV are as unremarkable as those for toothpaste. And remember, the issue here isn't what should or shouldn't be allowed, but what should shock a rational Martian scientist.