The cheerleader effect

by on December 4, 2013 at 5:23 am in Science, Sports, Uncategorized | Permalink

Whether you’re a casual user of social media sites like facebook and twitter or an avid online dater accessing eHarmony or Match.com, chances are you’ve created a personal online profile and been faced with a decision: What should you post for your profile picture? Many people post head shots or selfies, while others opt for pictures of their children, spouses, pets, or even favorite quotes or symbols. If your goal is to be perceived as attractive (and let’s be honest, whose isn’t?), then new research by Drew Walker and Edward Vul at the University of California, San Diego suggests your best bet is to opt for a group shot with friends.

A photo with friends conveys the fact that you are amiable and well-liked, but oddly enough that is not what makes you more appealing. Instead, the new research shows that individual faces appear more attractive when presented in a group than when presented alone — a perceptually driven phenomenon known as the cheerleader effect.

And why does this work?:

Walker and Vul posit that the cheerleader effect arises from the interplay of three different visuo-cognitive processes. First, whenever we view a set of objects like an array of dots or a group of faces, our visual system automatically computes general information about the entire set, including average size of group members, theiraverage location, and even the average emotional expression on faces. Thus although the group contains many individual items, we naturally perceive those items as a set, and form our impressions on the basis of the collective whole.

In addition, the impression that we have of the group as a whole influences our perception of any one individual item. We tend to view individual members as being more like the group than they actually are. Thus when we see a face in a crowd, we tend to perceive that face as similar to the average of all the faces in that crowd.

As it turns out, we find average faces very attractive. Composite faces, which are generated by averaging individual faces together, are rated as significantly more attractive than the individual faces used to create them. According to Walker and Vul, if presenting a face in a group causes us to perceive that face as more similar to the average, we are likely to find that face more attractive.

In one experiment, the researchers found that a group of four was large enough to create that effect.  Does this have implications for rock and roll?

That is from Cindi May, via Gareth Cook.

8 December 4, 2013 at 5:56 am

Post pic with hot friends; go to club with ugly friends.

Alan December 4, 2013 at 6:48 am

Hmmm. Rent models to take pics with or bone up on Photoshop?
@8: Average is over…

prior probability December 4, 2013 at 6:52 am

Phooey … This is the kind of “research” that gives “social science” a (deservedly?) bad name …

Eric December 4, 2013 at 7:16 am
ahduth December 4, 2013 at 7:21 am

I’ve just been using Rob Ford’s picture lately.

Rahul December 4, 2013 at 7:23 am

I think prior work shows that composite faces are more attractive than the average attractiveness of individual faces from the set used to create a composite. What the present reasoning seems to use is that a composite is more attractive than every individual member of the set. That does seem a far stronger conclusion.

I don’t think that holds but I could be wrong. Anyone know?

Thornton Mellon December 4, 2013 at 8:28 am

I think you have to be right. A picture of a beautiful face is fine on its own.

zbicyclist December 4, 2013 at 9:41 am

I have not read the original research, but it would seem hard to rule out the “I’m in a group so you can see I have friends” effect.

I remember some research by OKCupid a couple of years ago on the type of profile pictures that work. IIRC, posing with some object worked pretty well (e.g. holding a guitar). The OKCupid blogger theorized that this provided a start to the conversation. If you see an attractive person with a guitar, you can start by asking what type of music they like to play, which is perhaps a better start to a meaningful relationship than “nice rack”.

Nikki December 4, 2013 at 11:17 am

Depends on the sample. The ideal is the average of the entire population. If the features in your subset are not uniformly distributed on both sides of the features of that perfect face, then yes, some of the individual faces will be more attractive than the subset’s average (but still less attractive than the universal average).

Rahul December 4, 2013 at 11:28 am

In which case, shouldn’t the recommendation of this study be “Post group photos with people more attractive than you?”.

If I understand you correctly.

Nikki December 4, 2013 at 1:59 pm

Of this study, yes: it follows directly from “We tend to view individual members as being more like the group than they actually are.” But this contradicts the results of some other research, including Dan Ariely’s.

unibrow December 4, 2013 at 1:19 pm

Here’s the reason the composite looks better. We prize symmetry. In the composite the asymmetry of one person, with, say a larger right eye than left eye, will be balanced out with the asymmetry of another person whose left eye is larger than his right eye. Thus the composite picture will be more symmetric and hence, considered more attractive.
(I haven’t read the research but this seems reasonable, anyway.)

Sam December 4, 2013 at 7:24 am

FYI, the phrase “cheerleader effect” comes from How I Met Your Mother. And it’s totally real.

Dangerman December 4, 2013 at 10:38 am

AKA the Bridesmaid Paradox, Sorority Girl Syndrome, and… for a brief window in the mid-90s… the Spice Girls Conspiracy.

Thornton Mellon December 4, 2013 at 8:27 am

If you wanna look skinny, hang out with fat people!

Arturo December 4, 2013 at 8:50 am

I don’t dispute the experimental research finding, but as usual, the press is entirely incompetent when it comes to the policy implications. In the time I’ve spent looking at social media profile pictures, I always cringe whenever anyone posts a group shot because

– I recognize there’s self-selection here. Attractive people will be confident enough to post a shot of themselves, while less attractive and insecure people will rely more on this group effect.

– I’m often viewing these social media websites with limited resolution, perhaps on a smartphone, so a group shot will often limit my ability to see the person I want to see. I’m not sure to what extent this happened in the psych lab.

– Often times I’m viewing pictures of strangers and have absolutely no clue who the person in the shot is, and so I won’t bother clicking on the profile of someone in a group shot.

dead serious December 4, 2013 at 11:05 am

“– Often times I’m viewing pictures of strangers and have absolutely no clue who the person in the shot is, and so I won’t bother clicking on the profile of someone in a group shot.”

Back when I was dating I would assume the least attractive person in the group and was rarely wrong.

John B. Chilton December 4, 2013 at 9:56 am

Is this one reason a firm sends a team to make a Powerpoint presentation when one is sufficient?

Rahul December 4, 2013 at 10:09 am

“Billable hours”? :)

Renee December 4, 2013 at 3:35 pm

Interesting. Reminds me of a post the OK Cupid guys did using their data a few years back: http://blog.okcupid.com/index.php/the-4-big-myths-of-profile-pictures/

It covered whether to smile, look aloof, or do the duckface thing.
What angle is best (selfie? not selfie?)
Whether guys should keep their shirts on
Whether girls should show cleavage

and a bit more. So sad that they shut down their blog after the Match acquisition.

Christine December 4, 2013 at 8:47 pm

Everyone please stop saying “selfie.”

Sarasota December 15, 2013 at 7:28 pm

A person essentially lend a hand to make severely posts I’d state.
This is the very first time I frequented your web page and to this point?

I surprised with the research you made to create this actual put up
extraordinary. Great task!

Comments on this entry are closed.

Previous post:

Next post: