Average Perversion
Jessica of Jezebel says, of a study that on the... um... face of it seems to small (sample of 52 families) to warrant anything but a "hmmm.... interesting..."
According to a new study, "men were more likely to pair up with women whose bone structure was similar to their own mothers, with a similar effect holding for womens' choice of men," the Guardian reports. Read all about it here.
This is another one of those studies where there's got to be more to it than that because I'm pretty sure that people have been making that observation for a *very* long time. Even *before* Freud. :-)
At any rate it's certainly true for me, although I always figured it was just familiarity. The women in my family, including my mom, my aunts, and my grandmothers all tended towards dark hair, medium height, and farmer/athletic/active builds rather than thick or thin who dressed simply and wore little or no makeup. And surprise, most of my partners, especially my long-term ones, have been... dark haired, medium height, and muscular rather than thick or thin, etc., etc. I've always sort of assumed people feel the same entanglement of familiarity in attraction.
I still could be wrong though, and it could just be me and I could just be a giant Freud-could-rest-his-case pervert. Even though the (small) study says maybe otherwise. :-)
The tricky part for me, though, is that being of, um, roughly medium height, build, hair color, and activity level my mom and all my other women relatives were actually just... pretty much like about 65-80% of the U.S. population at the time I was receiving any potential preference "imprinting..." and so *once again* I feel the prospect of perversion slipping from my grasp! :-)
Anyway, without further criticizing a study I've only read about second or third hand I'll just say I'll be impressed if the researchers were able to filter out the standard bell-shaped curves of bone structure with the sample size they used.
I tracked down the actual article. It's in Proceedings of the Royal Society September 2008, "Facialmetric similarities mediate mate choice: sexual imprinting on opposite-sex parents" by Tamas Bereczkei, Gabor Hegedus, and Gabor Hajnal. Not available online, but you can probably get it through your local library. Here are my thoughts:
First, the authors of the paper did not look at preferences in height, build, hair color, or anything like that; they just looked at preferences in face shape. "Bone structure" is not an especially helpful way of putting this, IMO.
Second, the authors took each individual, compared the facial structure of their partner to the facial structure of their parents, and then aggregated the results. They didn't just take aggregated data about facial structure of partner and compare it to aggregated data about facial structure of parents. Does that help with your worry about common types?
More generally, filtering out spurious correlations is a big and important problem in statistics. The typical solution to this problem, and the one the authors of this paper use, is p-values. A p-value is basically the probability of finding the observed result or a stronger one by random chance. Generally, a result is considered statistically significant if p < .05, or in other words, if random chance would give you the result less than 5% of the time. P-values are not perfect, and being a statistics nerd, I could probably write you a whole treatise about when and how they can be misleading, but they're pretty standard and they work decently in most cases. I don't think there's a problem with them in this paper.
Third, I don't think that this paper shows that we're perverts, or that Freud was not a quack. (This next bit is sort of my own spin on the idea of "imprinting", so somebody please call me out if I start to sound batty.) People seem strongly disposed to categorize each other as male or female, but exactly what goes into those categories is learned. Who better to learn from than your parents? They're around a lot, and they're among the first examples of male and female people that you encounter. One thing you might learn is what the face of sex you're going to reproduce with is "supposed" to look like, by observing your opposite-sex parent. The authors speculate that the behavior they found is adaptive, but they note that a lot of other people think it's just a byproduct. I think the byproduct explanation sounds much more plausible myself. People will rationalize anything if you give them the chance.
Fourth, I love (read: hate) how this was reported as "Freud was right", and how both Jezebel and the BBC illustrated the idea of "attractiveness" with hawt blonde chicks. Your hairbrush photo is extremely refreshing.
Well there, I wrote you a novel. Maybe I go should start a blog called "Real Adult Statistics". Oh, also, I've been enjoying your serious posts on sex work. I haven't commented because they seem about right, and I don't have enough experience with sex work to add much.
[Hey, like *I* have that much experience with sex work! You seem to actually know what you're talking about so why let it stop you? :-) Thanks for the good information, P. --fl]
And to make it even worse, that was only a partial novel, because the comment-o-bot mistook my "less than" sign for the beginning of an HTML tag.
A result is considered statistically significant if p is less than .05, in other words, if there's less than a 5% chance of getting your result or a stronger one by random chance. Lots of people smarter than me have written big treatises about the limitations of p-values, and they're clearly not perfect, but they're a nice simple test and my impression is that they mostly work OK.
[I fixed the less-than sign for you. Thanks, P! --fl]
P. Burke - I for one think your third point makes a lot of sense. If you're right, you'd probably expect men's preferences to resemble their mothers more than women's resemble their fathers, simply because up to now, mothers have been so much more present during children's early years.
Sungold, that's an excellent question, but I'm not sure how to answer it. Women and men seemed to single out different facial areas in their mates. Women mostly picked up on resemblances in the eye/nose area, while men mostly picked up on resemblances in the mouth/jaw area. Also, they took 14 measurements on every face and then compared ratios of various measurements, so there are a lot of parameters. The researchers thought the difference between face areas was evidence that facial imprinting was adaptive. They're probably right about there being some evidential relationship; I just think it's a weak evidential relationship and want to correct for the tendency to think that everything is an adaptation.
The authors did say that previous studies had found a link between a person's relationship with their opposite-sex parents and their attitude toward partners who resemble that parent. They say, "our former studies have
revealed that daughters who received more emotional support from their (adoptive) father were more likely to choose mates similar to the father than those whose father provided a less positive emotional atmosphere (Bereczkei
et al. 2004). Similarly, a negative correlation has been found between maternal rejection towards son and mother–partner resemblance (Bereczkei et al. 2002)." Maybe this is just a case of science telling you the obvious.