Article - study on name perception

I ran across an interesting article: "Do you look like your name? People can match names to faces of strangers with surprising accuracy." https://www.sciencedaily.com/releases/2017/02/170227100804.htm The study discussed in the article controlled for "ethnicity, age and other socioeconomic variables." One of the comments made by the lead author: "Prior research has shown there are cultural stereotypes attached to names, including how someone should look. For instance, people are more likely to imagine a person named Bob to have a rounder face than a person named Tim. We believe these stereotypes can, over time, affect people's facial appearance."

Replies

1
March 20, 2017 1:35 PM

I was thinking of writing about that study! The mechanism the researchers propose as an explanation -- that we subconsciously transform our very faces to conform to stereotypes about our names -- strikes me as rather a stretch.

If you look at the sample image in their paper, it's a full headshot packed with cultural signifiers: hairstyle, facial hair, clothing. Similarly, every name is packed with social signifiers. IMO it's likelier that we're matching that meta-information than that we're, say, gaining weight to conform to our "round" names.

(FWIW, the authors heavily reference this study I wrote about some years ago: http://www.babynamewizard.com/archives/2007/6/funny-he-doesnt-look-like-a-methuselah )

2
By EVie
March 21, 2017 10:26 AM

Really interesting! I asked my husband about this (he's a research psychologist), and his reaction was the same as yours, Laura, except rather less diplomatic ;) He verified that the journal it was published in, the Journal of Personality and Social Psychology, is the top journal in social psychology, so totally legit; however, he says that in recent years, they have been leaning heavily toward "cutesy" studies that make big, splashy claims over less sexy but more substantive research, and that encourages researchers to make their claims bigger and splashier in order to pique the interest of the editors. (He may also have a slight chip on his shoulder about this issue... don't tell him I said that!)

The Bob/Tim thing strikes me as a variation on the Bouba/Kiki effect--still fascinating, but not new. 

3
March 22, 2017 10:05 AM

That's interesting about JPSP -- I guess nobody's immune to the desire to make headlines.

The good news is that the two proposed mechanisms to explain the result should be testable. Scrub the faces of hair, clothes etc. and try again? Or how about this: repeat the experiment using faces of military personnel who are issued regulation clothing and haircuts. That would let you retain natural photos as stimuli but eliminate the confounding variables. 

 

 

4
March 22, 2017 11:13 AM

I tracked down the original study, and the authors state that they intentionally included hairstyles and such:

"because research has demonstrated that the external features (e.g., hairstyle) of unfamiliar faces are essential for recognition (Ellis, Shepherd, & Davis, 1979), we kept targets’ original hairstyles and focused on real faces with their everyday appearance (except in Study 6, where we isolated both the hairstyle and the facial features)."

They tried to control for ethnicity and age with their sample photographs, but I'm really not convinced that these are the only factors that come "before" naming rather than after. Just to take an extreme example, if men named Whisper are less likely to have a stereotypically "masculine" appearance than boys named Nitro, it seems less likely to me to be because each is trying to "look like his name" than because of the kinds of values their parents had, which manifest themselves in their children both socially (the kinds of hairstyles and activities they favor) and perhaps even genetically (if the mothers are selecting both names and mates based on how well they fit a particular image, those traits may be passed on to their sons).

I would love to have someone who is familiar with the Israeli or French naming landscape to take a look at the name results--some names were apparently much easier to match to the correct individuals than others (e.g. Netanel vs Ofri), but I don't know enough about the names themselves to guess what might be different about them.

5
By EVie
March 22, 2017 11:52 AM

I totally agree that the parents' values are more likely a third variable influencing both name and appearance than the causation running name --> appearance. 

When I was in college, I participated in a study that used photos from congressional races as stimuli--not quite as standardized as military, but similar while still allowing for some personality. They were shown in pairs, and I had to rate which face looked more "competent." They were looking for how people's snap judgments on appearance predicted election results--I think they found it was about 60% of the time, though I'd have to look up the paper (I think it was an Amos Tversky study, though I'm not sure). 

6
March 22, 2017 12:55 PM

Definitely -- there are lots of ways that a person's appearance can reflect their culture and background without subconscious self-transformation is required.

I'm thinking more and more that I should write this column!