This should be on the list for this year's Ig Nobels

Is there an Annals of Just Plain Silly Research?

From the What Sorts of People blog, here's a lovely article from the Daily Mail (complete with file photo of a temperamental toddler --- 'cause that's what autism is, y'know, all autistics are children and we are always throwing tantrums) describing a study comparing the waist-to-hip ratios of women judged attractive by fathers of autistic children and by other men.

Waist-to-hip ratio (WHR) has come into vogue in evolutionary-psychology circles as some kind of numerical proxy for Ideal Womanhood, being an indicator of higher estrogen levels and (the theory goes) higher fertility and greater ease of childbirth. (And, of course, everything men find attractive in women is firmly rooted in reproductive biology *snort*). Preference for low WHR is also claimed, by these same evolutionary psychologists, to persist across cultures, though Echidne casts some doubt on that:

[T]he 0.7 waist-to-hips ratio was initially established in a study done in the U.S. Its supposed international applicability was based on the fact that some of the students used in the study had roots in various other countries. It's possible that later studies have been done to prove this one, but but Anne Innis Dagg (in "Love of Shopping" Is Not a Gene) reports on several studies on nomadic and semi-nomadic tribes in Tanzania and Peru which found that the men in those tribes preferred women shaped big and tubular, not like an hour-glass, because the former shape was associated with a fatter woman and hence a better worker. Once some of these men got into contact with Western lifestyles their preferences changed. --- This suggests that it is very difficult to disentangle "evolutionary" explanations from the cultural hegemony the Western nations currently have.

Why would anybody expect that to have anything to do with autism? Well, if you recall Simon Baron-Cohen's theory that autism is an extreme case of normal male brain wiring, and thus related to higher levels of testosterone exposure in utero, you might go on to suppose (as Baron-Cohen has done) that autism's persistence in the gene pool results from assortative mating of strongly "systematizing" --- though not necessarily themselves autistic --- types of people. Since he has equated that cognitive style with high testosterone levels, it would make sense in the context of his theory that the women fathers of autistic children are attracted to would show signs of having high testosterone levels (i.e., higher waist-to-hip ratio).

In this study, researchers the men's (52 fathers of autistic children and 65 fathers of NT children) preferences by showing them photographs of five women, each woman's image shown both as is and digitally altered to have a noticeably lower or higher WHR. The men were shown images in pairs and asked which was more attractive. Pairings could include either two different women or two versions of the same woman, differing only in WHR. There was no difference between the two groups of men in which of the five women they rated most attractive, but when comparing the Photoshopped images of any given woman, the fathers of autistic children tended to prefer versions of her with a higher waist-to-hip ratio than did the men in the control group.

They also found, in the course of asking demographic questions to the study participants, that the fathers of autistic children had more sons (the ratio of male children: all children was 0.68 for that group and 0.47 for the control group). That's somewhat interesting, since it takes into account non-autistic children of the same parents, and does suggest there might be something more to the gender imbalance than mere diagnostic invisibility of female autistics. However, this was not an epidemiological study, and a sample of fifty-two families is hardly adequate material for one.