December 02, 2010

Grumble Grumble Grumble

From AOL ParentDish: Nothing Personal, Dad: Your Daughter's Not Calling Because She's Fertile
When a woman is, ya know, fertile, she wants a guy like Brad Pitt or Johnny Depp standing by.
This article is an exercise in exquisitely confusing use of innuendo. The "ya know" (thanks, Ms. Palin) and the italics around "fertile" (in the original, not my addition) seem to be suggesting that what the author really means here is "when a woman is horny". (That, and the mention of two sexalicious Hollywood hunkinators, both of whom are enough to make anybody fertile. Hubba hubba.)

On the face of things, then, I get it. Except that if you read on it turns out the author actually does mean fertile. Like, as in "able to conceive". So why the innuendo? Is it some kind of double-bluff? Is this author trying to cack-handedly inject "personality" into his writing in a lowest-common-denominator sort of way? Or is he just yet another example of somebody getting paid to say something interesting even though he doesn't have anything all that interesting to say, thus giving grist to the web's great Inane-o-Matic Content Generator?

Based on this next paragraph, I'm going to go with the latter.
Psychologists have already established how women appreciate manly faces, masculine voices and a certain amount of macho swagger when their menstrual cycles are at their friskiest.
I'm sorry, when their menstrual cycles are at their friskiest? What exactly does a frisky menstrual cycle look like? ("Oh, gee, honey, better get the leash! My menstrual cycle's trying to hump the mailbox again!")

Now, to transition from my predictable I-hate-mindless-blogging rant into my equally predictable I-hate-evolutionary-psychology rant. (Hidden after the jump, for those who don't care.)
…women appreciate manly faces, masculine voices and a certain amount of macho swagger when their menstrual cycles are at their friskiest.

It makes sense from a Darwinian perspective.

Now, a study out of the University of Miami says that's all true, except when the manly face and masculine voice belongs to the woman's father.
Oh, what a multitude of sins that single-sentence paragraph disguises! In this case, from a writerly perspective anyway, I actually have to hand it to the author for managing to boil down so many decades of dubious, pseudo-scientific theory into such a succinct seven words. Where to begin?

1. If you take as a starting point for your empirical experiment a concept like "masculine" — a word that can be defined as both "[objectively] relating to men" and "[subjectively] associated with how men should be" — then, duh, your results are going to be just a tiny bit fucking useless. What does it mean that an "average" woman is turned on by a picture of what a researcher deems a "masculine" face? From the strictest empirical point of view, the only thing you can definitively call a masculine face is one that belongs to a man; if you start imputing culturally contingent values of masculinity to subsets of those pictures, all you are really proving is that your test subjects and your test administrators share the same cultural values about what it means to be "manly". (Presumably if they'd conducted similar experiments a couple of hundred years ago, they would have concluded that women have a genetic preference for powdered wigs.)

2. This particular study was conducted by examining the cell phone records of forty-eight women over one billing period — and because women called their fathers less during the portion of that (one!) billing period that overlapped with the peak of their fertility, the researchers concluded that all women avoid their fathers when they're fertile. Never mind that the sort of woman who can call either of her parents' cell phones from her own cell phone is at best representative of a particular class of twenty-first century American family, and not (a) any other kind of contemporary family arrangement, or (b) the kind of evolutionary environment where this trait supposedly developed; never mind that this same study found that women in general talk less to their fathers on the phone anyway; never mind that there are a gazillion other potential reasons why a person might talk to another person more or less, fertile or not: this is clearly evidence that women are afraid of bangin' their dads.

3. Um, hello, can't you see that you are projecting a social taboo onto a putatively biological behaviour, here? Even if you could convince me that American women really do avoid their fathers when they're fertile, and that there's value in flattening human behaviour into biological imperatives like this — and you're not, at the moment — isn't it possible that years of anti-incest social conditioning have made women just feel plain awkward around their dads at the time of the month when they've supposedly got sex on the brain?


And now, finally, back to my I-hate-mindless-blogging rant:
Fertile women apparently avoid their dads like the plague. Literally.
"I'm sorry, Dad: either you talk to me from quarantine this week or you don't talk to me at all."



Papa' said...

We might have some interesting discussions over Christmas. The findings under point (1) in your anti-evolutionary-psychology rant are actually pretty solid and well-replicated. I don't have any comment on points (2) and (3) - I agree this particular study seems a bit thin and a bit over-reported.

Anonymous said...

In point 1) typically masculine and feminine features are operationalized by taking many photos of men and women, measuring the dimensions of various facial features, then determining the likelihood that a feature with those dimensions comes from a picture of a male or a picture of a female. The images the subjects see are composite images with features taken from the set. The most 'masculine' face will have dimensions that occurred in many of the male pictures, and almost none of the female images. This kind of procedure has been widely used, and has good test-retest and cross cultural reliability. It's a pretty objective way of determining male vs female features.

Andrew said...

KC is alive!

Okay, you two Kool-Aid drinkers, I am prepared to admit that, given this new information, the way masculine and feminine are operationalised sounds more valid than I first thought. (Though I still wonder if there isn't room for bias creeping in here. Like, these men and women that the composites are based on — how are they selected, and identified as male/female in the first place? And do these composites attempt to account for variables like hairstyles, make-up, facial stubble, plastic surgery, etc.? Presumably not, since that would introduce a potential "cultural bias", but it seems like the *absence* of those features would also introduce a cultural bias in that people rarely make male/female distinctions based solely on the dimensions of facial features. How would Western subjects react to the "most masculine" face if it were slathered in lipstick and mascara? How would non-Western ones?)

In any case, I still have another objection to this kind of test, along the same lines as point (3) above: the possibility of people projecting cultural values onto a seemingly biological phenomena, i.e., of people "gaming" the test. Not in the sense that they're deliberately trying to achieve an outcome or distort the results, but in the sense that if you show people a bunch of pictures of different-gendered faces, they're going to react to them in whatever way they think is appropriate given their own gender and whatever expectations they have about what the test is trying to study. Do you really think that if you tell a (insecure) straight man to identify the faces he finds most attractive, he's going to sit there and tell you it's masculine ones?

Or is there some other ingenious piece of experimental design to counteract that kind of problem, too?

Papa' said...

Gaming the test: do you really think that, in the first experiments to demonstrate that women have subtly different preferences in male faces at different times in their cycle, the women who were experimental subjects had any reason to think that it was culturally appropriate to express different preferences at different times in their cycle? Maybe now that findings of this kind are more widely known, you might get some women trying to undermine the biologically-based interpretation, but I doubt even that. In my experience - which is admittedly in a less ideologically fraught area - experimental subjects genuinely try to do what you're explicitly asking them to do, not what they think you secretly really want them to do, and still less what they think would give findings they regard as ideologically acceptable.

Anonymous said...

As far as I know the male and female faces are based on self identified male and female college students. Some of the positive biases you mention are taken care of by taking the photos without makeup, only using clean shaven people, and all of their hair is cropped from the images. You're right that there is still a cultural expectancy for women to have unusually large seeming eyes because of makeup, but part of the point is to remove culture from the experiment as far as possible so that the results can generalize. It's a well understood caveat to the method, which I don't let bother me. As far as I know no trans people were included in these sets, and no testing was done to identify individuals whose reported gender differed from their biological sex. All I can say is that a large N solves most problems, if the frequency of people who don't fit into a self-reported Male / Female binary is sufficiently low, by recruiting enough people your average face even without those cases will be very very similar to your average face with that tiny population included.

Your Dad quite correctly points out the beauty of a within-subjects study design. Good job Andrew's Dad.

Further, while I know you're looking for snarky blog fodder, holding the whole field of evolutionary psychology to account based on the AOL parent dish rehash of a University of Miami press release probably doesn't do the actual paper justice. I haven't read it, and I know that you and I fundamentally disagree about the utility of EP, but I'm willing to go on a limb and presume that their conclusions in the paper were a whole lot more tempered than what made it to AOL.
Alive as always,

Andrew said...

I think I'm going to let this one stand as is, other than to point out that, yes KC, I am looking for snarky blog fodder, and AOL Parent Dish is a better place to find that than Nature -- but I've also studied EP in an academic setting and I'm often similarly dubious even when I'm reading this stuff in the original.

Besides, the biggest problem I have with EP is precisely that it always does sound so succulent that the media pick it up and spin it into something unrecognizable. My dad can also quite nicely support my curmudgeonliness on that point...

Oh, and speaking of large ns, did you read that thing in the New Yorker last week about the decline effect? Super-interesting stuff.

Post a Comment