Do we remember a pretty face?

Are attractiveness and distinctiveness related? Are we more likely to remember a pretty face than an ordinary one? This data suggests not:

i-c7f9baade37c36803089788218db8a78-attractdev.gif

When people are asked to rate faces for attractiveness and deviation from an average face, there's a clear correlation: the more attractive the face, the less it deviates from average. Average faces are more beautiful, it seems. But now consider this data:

i-d94853f3ac16aa48bc24f074f7f34523-attracttrad.gif

Here, people were asked to rate faces on the same scale of attractiveness, but instead of rating whether a face deviated from the average, they were asked whether a face would stand out in a crowd. By this measure of distinctiveness, there is no correlation between attractiveness and distinctiveness. The best fit for the data is a quadratic curve, and even it doesn't fit very well. Some pretty faces are also distinctive, and some are not. So what's going on here?

These charts represent some of the results from a study by Lee Wickham of the University of Manchester and Peter Morris of Lancaster University ("Attractiveness, Distinctiveness, and Recognition of Faces: Attractive Faces Can Be Typical or Distinctive but Are Not Better Recognized," American Journal of Psychology, 2003).

Wickham and Morris had undergraduates rate photos of women from a different university on several different scales, including the attractiveness and distinctiveness ratings above, but also for memorability. In addition, they tested a separate group's actual memory for the faces both immediately after viewing the pictures and five weeks later.

The results for the memorability ratings were similar to the traditional distinctiveness ratings—how much a face would stand out in a crowd:

i-86a746c0f3f4648fa35b30ad9ce0851c-attractmem.gif

However, when participants' actual memory for the faces was tested, the results were curious: there was no correlation between attractiveness and memory, but there was a significant correlation between attractiveness and false positives. When a participant saw an attractive face that he or she hadn't seen before, he or she was more likely to falsely report having seen it than for less attractive faces.

The strongest correlations the study found were between traditional distinctiveness and memorability—how much a face stands out from a crowd is directly related to how well we think we'll remember it. In fact, we don't do as well as we think we will, but distinctiveness is still a very good predictor of how well we'll actually remember a face.

This study may explain why supermodels typically aren't flawless—Cindy Crawford's mole and Angelina Jolie's huge lips make it easier for us to remember them, even if they don't necessarily make them more beautiful.

More like this