I recently did a study examining how your name affects how attractive you are perceived, and the resulting flurry of media attention has taken me very much by surprise. I think it's nice that so many people are interested in this topic, so I thought I'd write something up explaining the study in a little more depth than some of the articles that are out there.

First things first, though: I'm a bit nervous that some people seem to be drawing huge conclusions that aren't necessarily warranted by my data. This was just a preliminary study, and good science depends on replication of results. I think it would be great if someone else studied the same thing (perhaps in the lab, or on a different website); if they got the same results, that would make the finding stronger. One thing that may limit the generalizability is that the people who visit hotornot.com are mainly young men, and a few women... I don't know how this would generalize over a larger population including more older people. The other thing is that this is just a statistical effect over many people; you can't really draw conclusions about how attractive your specific name is on you. Indeed, there were exceptions to the rule in the photos I used in my study; though most showed the vowel effect, for some it was smaller and a few it even went in the other direction. Moreover, though I counterbalanced for the obvious things (e.g. masculinity/femininity of name), names in English come weighted with differences in connotation and frequency that I couldn't counterbalance for; it is possible that some of these differences might be responsible for this finding. Generally, when you present a poster at a conference (unlike, say, a paper in a refereed journal), it's often understood as "work in progress", which is what this was. I would not consider it a robust finding until and unless these caveats could be answered to my satisfaction.

What was the study? I took 24 photos of friends of mine (12 guys, 12 girls) and posted them on the website hotornot.com. I put each picture on the site multiple times, appearing with a different name each time. (Hotornot doesn't normally post names with the photos, but I photoshopped the names in and they allowed it). The names differed in how masculine or feminine they were, and also in the sound of the vowel (front or back).

What the heck are front and back vowels? Linguists classify vowels according to where in the mouth they are pronounced or articulated. Front vowels are like the i in beet or bit, the a sound in bait or bat, and the e sound in bet. Back vowels are pronounced in the back of the mouth, and include the o sound in food or hot, as well as the u sound in put.

What was the finding? There is an opposite effect for men and women. Men whose names' stressed vowel is a front vowel were rated statistically MORE attractive than men with names with a stressed back vowel. However, the reverse was true for women: women with names with stressed BACK vowels were statistically more attractive.

Give me some examples of sexy and unsexy names for men and women. I want to stress that this effect is statistical, meaning any one name or one individual might not see it happen. And in the study itself I didn't gather enough data per name to draw any conclusions about any specific name. Nevertheless, here are some example names, to show you how to classify by front and back vowel:

Men's names with stressed front vowel: Dave, Craig, Ben, Jake, Rick, Steve, Matt

Men's names with stressed back vowel: Lou, Paul, Luke, Tom, Charles, George, John

Women's names with stressed front vowel: Melanie, Jamie, Jess, Jill, Amy, Tracy, Ann, Liz

Women's names with stressed back vowel: Laura, Julie, Robin, Susan, Holly, Carmen

How big is this effect? Not very big. It's statistically significant, meaning it was unlikely to have come about by chance, but it's not large - less than half a point on a 10-point scale.¹ In other words, if you're Brad Pitt, you'd be more attractive than Joe Schmoe regardless of what your name is; but if Jud and Jim Schmoe are otherwise equally attractive, then Jim, who has the 'better' name, might be statistically more likely to be rated attractive than Jud. I wouldn't consider changing your name over this - my name is Amy, which is technically more unattractive, but I like it! And I think that factors like overall looks and personality are far more important.

Does this work for all names? Not all vowels can be classified as front or back. For instance, the i in Diane is what linguists call a diphthong, which means a vowel that actually has two or more vowel sounds in it. (Think about it: it starts off with the ah sound in hot and moves to the ee sound in feet. So it starts in the back and ends in the front). I didn't look at names with diphthongs, or more central vowels, in this study.

Why do you think this happens? I want to stress here that I'm JUST GUESSING... but there has been other work suggesting that cross-linguistically, people think that front vowels are 'smaller' and back vowels are 'larger.' Now, you'd think that that would make guys with front vowels do worse than guys with back vowels (and vice versa), but there are other studies suggesting that women actually aren't most attracted to the super hyper-masculine, macho guys (and men aren't most attracted to the super feminine girls). It seems we like people who are somewhat masculine or feminine, but not too much. The reasoning is that maybe women want guys who can be kind of sensitive and gentle, and good providers - while guys want women who have a bit of spunk. What does this have to do with names? Well, maybe a guy with a front-vowel name seems subconsciously gentler or more sensitive, hence more attractive (and vice-versa for women). Again, this is a complete chain of guesswork; it could be another explanation entirely. More research needs to be done. But it makes a bit of sense, when you think about it.

In this analysis, aren't you assuming that everyone using hotornot is heterosexual? Well, there's no reason that gay guys couldn't also like guys that aren't super masculine, and lesbians couldn't like women who aren't super feminine. But this question brings up a very good point: we don't know who rated these pictures. It seems reasonable to assume that about the same proportion of gay people visit hotornot as are in the regular population (which would probably be a small enough number not to change the results even if homosexuals acted differently for some reason), but that assumption might be totally wrong. And a number of straight people could have rated pictures of their own gender. This all supports my earlier point that replication of this study (preferably in the lab, where you know the gender and orientation of the raters) is a very good idea.

Any other findings? I also looked at how masculine or feminine a name was, and whether that affected attractiveness. Indeed, guys with feminine or androgynous names (like Jamie or Brittany) were rated significantly less attractive than guys with masculine names (like Dave or Steve). But for girls, it was slightly different: women with either very masculine or very feminine names (like Brittany or Dave) did better than women with androgynous names (like Jamie or Lee). I'm not sure why that is, but I bet it has something to do with our culture, in which it's okay for a girl to be a tomboy or have some more 'male' traits, but still unacceptable for a guy to have female traits - he'd be called a sissy.

Does this extend to other languages? I have no idea. Since presumably most people at hotornot were English speakers, I couldn't really predict if the same thing happens for other languages. Maybe; if so, that would be a lot stronger. But you'd have to do the research to find out.

Are you planning on following this up? Probably not. My main area of research is unrelated; I am working on developing computational models of language and conceptual representation and development. This project was inspired by a class I took on language and thought, taught by professor Lera Boroditsky. However, I would definitely be interested to see results of anyone else who does similar work - as I said, replication is the key to good science. I'd love to know how robust this result really is, and how much you can generalize to different groups of people or different situations.

1 - Because I got my data off of hotornot.com, I was able to gather thousands of data points, which is generally a good thing (the more you have, the more reliable the finding). But it also means that smaller differences are more likely to be found "statistically significant" than if the dataset had fewer points.

Here are some links to some of the better articles about this.

The original study. Perfors, A. What's in a name? The effect of sound symbolism on facial attractiveness. Presented at the 26th Annual Conference for Cognitive Science. Chicago, Illinois; Aug 7th, 2004.

Nature, UK. Name game increases sex appeal. By Michael Hopkin. Aug 10, 2004.

New Scientist, UK. Pleasing names make faces sexier. By Emily Singer. Aug 11, 2004

Guardian, UK. Why Matts are hotter than Pauls. By Ian Sample. Aug 12, 2004.

Boston Herald, Boston, MA. Study of the sexes shows lovers howl over vowels. By Thomas Caywood. Aug 12, 2004