- Penton-Voak & Chen (2004) found that more masculine-looking men had higher circulating T levels.
- Peters et al (2008) did not find such a connection
- von Bokhoven et al (2006) notes that there is no link between testosterone levels and facial appearance in adolescence
A recent anonymous comment on the February 10th post offered two more possible citations for this debate:
- Roney et al (2006) found that men's testosterone levels predicted women's ratings of their facial masculinity
- Hönekopp et al (2007) found no correlation between facial masculinity and testosterone levels.
- Pound et al (2009) (with Penton-Voak) found that men with more masculine faces had a stronger testosterone response to competition.
- Campbell et al (2009) found a link between facial masculinity and sexually unrestricted behavior in women (which is hypothesized to be a marker of higher testosterone)
My main problem with the facial-masculinity = higher-testosterone argument was that testosterone is so reactive to environmental factors, so I find Pound's argument (that masculine-faced men have more reactive endocrine systems) fairly compelling. I can't argue that increased testosterone doesn't contribute to facial masculinity. With my n = 1 experiment of myself, I can see distinct changes in my face pre-and-post hormones. But my hormones are externally regulated in a way that almost never happens to normatively gendered people. So I feel really awkward trying to compare my experience to any non-trans person's.
Find out the day's topic before you read: follow diffblog on Twitter! Diffblog also available on LiveJournal.
Difference Blog Reader Poll
Posts
8 comments:
what does facial masculinity mean?
@Penny
Ooh, good question. In most of the studies, they didn't define it, leaving the judgments of "what is masculine" to the college-aged-female-straight-raters. This is true for Penton-Voak, Peters, Roney, & Honekopp.
It looks like most used a 1 to 7 scale from "far less masculine than average" to "far more masculine than average".
von Bokhoven, as it turns out, used self-report ratings of facial changes reported by adolescent boys, if I'm reading it correctly, but that would be ... well, really weird to me. I have a hard time believing it.
Pound used what I think is the most objective measure:
"Five facial dimensions (ratios) previously shown to be sexually dimorphic (Penton-Voak et al. 2001) were measured. These ratios were: (i) eye size (horizontal inter-exocanthial distance/inter-endocanthial distance), (ii) lower face/face height (vertical distance from mean pupil height to gnathion approximation/vertical distance from trichion to gnathion approximation), (iii) cheekbone prominence (horizontal distance between most outward projecting points on the face at or below the eyes/horizontal distance between left and right gonion approximations), (iv) face width/lower face height (horizontal distance between most outward projecting points on the face at or below the eyes/vertical distance from mean pupil height to gnathion approximation), and (v) mean eyebrow height (mean vertical distance from pupil to inferior aspect of brow/vertical distance from trichion to gnathion approximation)."
Campbell used both measurements and ratings from male and female raters.
Apicella et. al found it was related to behavior, but not circulating or prenatal testosterone levels.
http://www.people.fas.harvard.edu/~apicella/CV_files/EHB2008.pdf
Another study relevant is Neave et. al 2003, which was discussed in both studies.
One important thing to keep in mind is that Pound et. al's findings were around 47 men. Almost every study with at least 75 failed to find a relationship.
Yay for random anon poster.
Ohh and very random, but a recent study actually challenged the widely publicized findings that women's preferences for masculinity in faces vary cyclically:
Peters et. al PLOS 2009.
For what it's worth, both measured masculinity and rated masculinity are associated with improved health to a similar degree.
Also, one thing about Pound et. al: no relationships were evident for perceptual masculinity, just measured masculinity.
This is an interesting thread. A few comments to add to the discussion:
If there was no relationship between facial masculinity and T in men, then studies would be as likely to report a negative correlation and a positive one. But they're not; every single one of the studies that report a correlation between T and men's facial masculinity reports a positive correlation (although in many cases it's not significant). I feel that this is a difficult point to ignore; while considering individual studies suggests evidence for a positive relationship between T and facial masculinity is rather mixed, stepping back and looking at the bigger picture suggests that there is a weak positive relationship between T and facial masculinity. Pound's suggestion that masculine men may be more sensitive to testosterone surges (coupled with the fact that measuring baseline levels of T is notoriously difficult) might explain why the relationship is relatively weak.
Although an earlier poster claimed that the Pound study found 'no relationship' between T and ratings of facial dominance, that's not quite true. The relationship wasn't signficant, yes, but it was around r=0.25. That's clearly not 'no relationship', nor is it significantly weaker than the correlation for measured masculinity.
I'm very wary of those studies like Neave et al that didn't minimize the effects of hairstyle of masculinity ratings. It's interesting to me that those studies that did minimize the effects of hairstyle on ratings have typically been those that have reported stronger correlations between T and facial masculinity (e.g. Roney et al; Penton-Voak). I think this an important point to bear in mind when considering these studies.
Finally, an earlier post mentioned the recent Peters study that found no cyclic shift in women's preferences for masculine men. They may not have found an effect, but I don't see how it challenges the view that women's preferences for masculine men are strongest during the fertile phase of the cycle. Since this effect has been reported in well over 10 different studies (several demonstrating the effect in multiple independent samples), it'll take more than one null finding to detract from the many positive effects.
apologies. the first sentence of my previous post should have read "If there was no relationship between facial masculinity and T in men, then studies would be as likely to report a negative correlation AS a positive one."
An important thing to note... The critical period when testosterone impacts men's facial features most likely happens during puberty. After growth is finished testosterone levels may change resulting in different correlations during the testing period.
Interesting study, that found that baby-faced men, who are less dominant-looking are actually more intelligent, better educated, more assertive and win more combat medals than "dominant" men.
http://www.marginalrevolution.com/marginalrevolution/2005/06/should_only_the.html
Post a Comment