This paper has recently made a splash in the mainstream science media, including CNN and Jezebel. The stories left me more than a little skeptical. Do straight women want to sleep with a guy's immune system? I had to see the paper for myself.

And as I found out, it's an interesting paper, but I think it's interesting for different reasons than the ones the media picked up on. What's interesting to me is not that you want to hump a guy's T cells (though don't get me wrong guys, I'm sure you're T cells are very handsome), but that cortisol may play a role in this as well.

And as I was writing this post, I realized something even more interesting about this paper...the premise may be off. Or at least unproven.


Rantala et al. "Evidence for the stress-linked immunocompetence handicap hypothesis in humans" Nature Communications, 2012.

The hypothesis behind this paper starts with testosterone and something called "honest signaling". An honest signal is a sexual selection hypothesis regarding the signals used by males to signal to females (usually during mating season) that they are healthy, fit, and otherwise the best choice of potential baby-daddy. This involves signaling that is "honest" and usually costly to the males. The idea is that if you're doing something THIS costly (called the "handicap principle"), well you must be pretty fabulous to be able to support such a cost and still survive. The classic idea of this is the peacock's tail. The tail is huge, makes flying and survival difficult, and the energy that goes into it negatively impacts the immune system. A peacock with an awesome tail must be pretty great if he's able to support that thing and still get around.

The idea in humans is that high levels of testosterone may be an honest signal of male quality. Testosterone may suppress the immune system (though actual immunosuppressive effects are inconsistent so far), so the idea is that traits dependent on testosterone, such as facial shape in men, may serve as honest signals of high testosterone, and thus as proof of a man's quality. He must be pretty awesome to have all that testosterone suppressing his immune system.

But it appears that the reality is more complicated. And it may not just involve your immune system, but your stress responses as well.

The authors of this study wanted to look at correlations between immune function, testosterone, cortisol (that's a stress hormone), and facial attractiveness in men. They took 74 men, took measures of cortisol and testosterone, and then gave them an hepatitis B vaccine. The vaccine will produce an immune response, causing the body to produce antibodies, which will then attack if the body becomes infected with hep B. The amount of antibody produced and circulating in the blood in the first month after you receive the vaccine is a relatively good measure of someone's immune function. They also had the men's faces rated for attractiveness by 94 women.

This first graph shows a correlation between facial attractiveness and testosterone.

This shows a correlation between immune response (the antibody levels) and testosterone.

And this shows a correlation between facial attractiveness and the antibody response.

What they found is that cortisol levels modulated the testosterone relationship to antibody production. Lower levels of cortisol (which may mean lower levels of stress) went with the strongest correlation between testosterone and facial attractiveness as well as the correlation between testosterone and immune response, which suggests that cortisol levels may modulate both facial attractiveness and immune response (not to mention possible effect on testosterone). Sadly, they did not look at correlations with cortisol and antibody response alone, which I think would have been interesting, or the correlation between cortisol and testosterone, which would also have been interesting, and which I'm not sure has been clarified in humans.

So the conclusion is that high testosterone, low levels of cortisol, and a high immune response to a vaccine correlate with physical attractiveness ratings. But there are several things here which bring the potential impact of this study down a bit. First, the correlation between testosterone and vaccine response disappeared when they took out the people who had NO response to the vaccine. This was a large number of people. Secondly, the ratings of physical attractiveness varied widely. Very widely. So much so that they had to compute mean attractiveness ratings. This is not a bad thing, but it does mean that womens' ideas of what is physically attractive are very different (not to mention the potential social and cultural pressures which may shape what women determine to be "attractive"), and that this may render the actual sexual selectivity of choosing a handsome face for its testosterone content moot. Third, while testosterone ratings now are nice, the relationship between testosterone and facial shape should presumably be biggest when that facial shape is forming, during adolescence. Looking at testosterone now may be not be indicative.

Fourth, and most important to me, is the issue of the handicap hypothesis, the idea that a specific trait is damaging to a male, and that it is damaging as proof of their superiority, and that females are attracted to it. This hypothesis has come under some scrutiny from various sources. In particular, it appears that, for example, female peacocks don't care about the intricacies of male tails, as long as the tail is good ENOUGH. And there is no consensus on whether or not testosterone really IS an honest signal of male quality in humans. The results so far have been inconsistent with the exception of reptiles. So I'm not really sure this is the best place to take your hypothesis from. If we're not sure that testosterone is a real honest signal of male quality, and facial attractiveness varies widely...I'm not really sure what these data mean. I think it's very important and interesting to note the finding with blood cortisol (though those levels also vary widely in humans, as well as throughout the lifespan, so the effects of higher cortisol in development could be important, not to mention the effects of stress on cortisol levels). There's more and more evidence of how closely intertwined our immune responses and stress responses are, and I think that the correlations here with testosterone may also be interesting. But I'm not sure if they really relate to male attractiveness, and what that relationship means.

Rantala, M., Moore, F., Skrinda, I., Krama, T., Kivleniece, I., Kecko, S., & Krams, I. (2012). Evidence for the stress-linked immunocompetence handicap hypothesis in humans Nature Communications, 3 DOI: 10.1038/ncomms1696