There is some evidence that the more average a face looks, the more attractive it is perceived to be. In 1994, however, David Perrett and colleagues showed that exaggerating certain traits away from average makes a face even more attractive.
We looked at how preferences for these traits differ when the starting attractiveness of the faces or the difference between the faces is changed.
We found that the face with higher levels of shapes associated with attractiveness are almost always preferred.
Preference for the left column versus the top row
 -200% |
 -150% |
 -100% |
 -50% |
 0% |
 50% |
 100% |
 150% |
||
|---|---|---|---|---|---|---|---|---|---|
| 200% |  |
93% | 90% | 83% | 77% | 67% | 58% | 48% | 47% |
| 150% | |
95% | 94% | 91% | 83% | 73% | 62% | 56% | |
| 100% | |
96% | 96% | 91% | 87% | 78% | 65% | ||
| 50% | |
97% | 94% | 94% | 88% | 75% | |||
| 0% | |
96% | 95% | 90% | 76% | ||||
| -50% | |
95% | 92% | 82% | |||||
| -100% | |
94% | 86% | ||||||
| -150% | |
83% |
The graph above shows what percent of people thought the face in the left column was more attractive than the face in the top row.
The more people that preferred the face with higher levels of shapes associated with attractiveness (i.e. faces in the left column), the lighter the background.
-
(1994).
Facial shape and judgments of female attractiveness.
Nature,
368:
239–242.