Animals want their children to have good genes. However, most lack the laboratory equipment to test potential partners for genetic quality. So instead they find genetic "proxies" attractive.
Proxies like the peacocks tail. It shows how badass his genes are; as he can survive despite lugging a giant weight around. Clearly he must be worthy to have all the peahen babies.
It was thought that many aspects of the human face are such proxies. They correlate with good genes, making them attractive to potential mates.
However, a review of research on the topic has found the support for this hypothesis just isn't there.
The claims
Evolution is all about reproducing successfully. As such, animals will invest an awful lot of time, energy, and effort into doing it. Hence the reliance on these proxies.
By mating with individuals with good genes (as identified through these proxies) they can ensure the investment in reproduction is worth it; from an evolutionary point of view. Previous research had identified a link between fitness and many parts of the human face. This might suggest that these bits of the face were such proxies. They were a handy way of identifying that the individual was fit and thus worthy of mating with.
For example, in men the height-to-width ratio of the face appears to be correlated with testosterone levels. The higher that ratio, the more testosterone a guy had. Additionally, they were also found more attractive. Might this be because the ratio is being used as a proxy for good testosterone-y genes by potential mates? Meanwhile better quarterbacks (which I'm told is some key role in an American sport) are often perceived as better looking. Seemingly further support for the face as a proxy.
Now, that's not to say that these traits are attractive on a conscious level. People aren't sitting on a date going " well his chin structure indicates he has genes that will help our child run from a lion ". Rather, individuals attracted to these traits are more likely to have children with the good genes they correlate with. Thus, their children are more likely to survive and evolution favours being attracted to those traits. Before long everyone is naturally drawn to them.
An attractive review
The "face as a proxy" idea seemed pretty good. It was supported by the evidence and evolutionary plausible. However, a new meta-analysis has found that this evidence isn't actually that attractive (hahaha).
Five studies were reviewed, all purporting to show evidence for the face being a genetic proxy for fitness. They found a link between facial characteristics, attractiveness, and a bunch of fitness variables. From general fitness to success at sports. So what's wrong with all this evidence?
Well, these studies did find a link between all those variables. And that this correlation was statistically significant. The issue this meta-analysis found is that that the correlation just wasn't very strong. On average, only 3% of the difference between faces was explained by fitness. This is a very, very small correlation. In fact, anything below 35% is typically looked at with suspicion.
This is because statistical significance isn't necessarily the same thing as biological significance. Sure, there's a real correlation here. But is it large enough to influence evolution? To influence the decisions about mates? To be significant in any way, from a biological point of view? None of the studies examined demonstrate that it is significant enough.
Could evolution (and you) be perceptive enough to spot a 3% improvement?
A cautionary tale
The worrying thing is that this study doesn't only raise issues with hypotheses about facial attractiveness. That's just one case study. There may well be other biological theories based upon such weak correlations.
And it's not exactly implausible that there are. The authors of this meta-analysis note that statistical analysis is often viewed as the "gold standard" to determine the validity of an idea. In reality, this may not be the case. They also identify many other problems they found with the research they examined. Like my old pet-peeve: the fact they were all conducted on WEIRD samples. In fact, it's even worse than that as many of them only examined WEIRD activities (like being a good quarterback). Is that really a behaviour with evolutionary significance?
So when examining new (and old) hypotheses; it's important to ask not just is there a real link, but how strong is that link? If it's very weak, like this whole face thing, then it might not be that biologically significant.
tl;dr
There is a statistically significant correlation between an attractive face and fitness; but it's so small it probably doesn't mean anything.
References
Lefevre, C. E., Lewis, G. J., Perrett, D. I., & Penke, L. (2013). Telling facial metrics: facial width is associated with testosterone levels in men. Evolution and Human Behavior, 34(4), 273-279.
Smoliga, J. M., & Zavorsky, G. S. (2015). Faces and fitness: attractive evolutionary relationship or ugly hypothesis?. Biology letters, 11(11), 20150839.
Williams, K. M., Park, J. H., & Wieling, M. B. (2010). The face reveals athletic flair: Better National Football League quarterbacks are better looking. Personality and Individual Differences, 48(2), 112-116.