Development of Preferences


Although many people still assume that beauty is in the eye of the beholder, 40 years of research has conclusively shown that people agree both within and between cultures who is attractive and who is not. Because facial attractiveness is thought to be a universal preference, many evolutionary psychologists argue that humans inherit an innate beauty-detecting template to identify healthy and fertile mates who will be more likely to pass on their genes.

The idea that facial attractiveness is a mate-value cue is still widely accepted; however, this viewpoint does not adequately account for two key pieces of evidence. 1) Although facial attractiveness is highly correlated with the perception of health, research has demonstrated that there is no link between attractiveness and actual health outcomes (i.e., attractive people have the same diseases [or not] as unattractive people). 2) Human beings perceive attractiveness in stimuli without mate-value. For instance, people are able to perceive beauty in same-sex faces, animals (i.e., robins are considered more attractive than ostriches), and even manmade objects.

cutefish uglyfish
Examples of attractive and unattractive fish

Rather than an innate beauty-detector, perhaps what has evolved is the ability to perceive the beauty of averageness. The one thing that attractive faces, birds, and buildings all have in common is that they are good fits of their category. Categories themselves are constructed via experience. For example, we prefer robins because we see far more robins and robin-like birds than we see ostriches and ostrich-like birds. Thus, attractive faces are “face-like” faces. We tested the idea that experience shapes facial preferences by exposing participants to novel faces created by morphing the faces of humans with those of chimpanzees. A separate group of participants agreed that these “chumans” were extremely unattractive. However, after participants had viewed examples of chumans repeatedly over 30 seconds, we found that they rated faces that were 80% human and 20% chimp as more attractive than faces that were 100% human. Participants who did not see any chumans preferred the 100% human faces.  Thus, we concluded that people prefer attractive faces because they best reflect experience and not a hardwired vision of beauty.


Yes. We have found that infants, ranging in age from two to six months, prefer to look longer at attractive faces (as rated by adults) than at unattractive faces. We have also found that 12-month-olds prefer to approach and play with a stranger with an attractive face compared to a stranger with an unattractive face.

To determine why infants have preferences for attractive faces so early in development, we once again tested the hypothesis that preferences for attractive faces may be due to categorical fit, thus making attractive faces appear more “face-like”. We found that infants categorized attractive and unattractive faces into separate groups. Like the “chuman” results, this finding suggests that the development of facial categorization may underlie the development of attractiveness preferences.



Given that 2-month-olds are already looking longer at attractive compared to unattractive faces, the precursors of a full-blown stereotype (defined here as differential attitudes and behaviors based on appearance) are present not long after birth. This finding strongly suggests that this stereotype is not the result of a media that celebrates attractiveness and derides unattractiveness. Those who believe that Cinderella creates the belief that beautiful chambermaids are inherently good while ugly stepsisters are necessarily evil may be advised to wonder where the Brothers Grimm generated their idea. We supposed that infants may bootstrap their own stereotypes based on valence. Put another way, attractive faces may generate positive feelings, devofprefwhereas unattractive faces may generate negative feelings. The same may be said of other categories that range from positive to negative.  In fact, we found that infants look longer at attractive faces when they are paired with smiling schematic faces, pleasant-sounding voices, and simple helping actions, whereas they look longer to unattractive faces when they are paired with frowning schematic faces, unpleasant-sounding voices, and simple hindering actions.

Infants’ own beliefs may be due to, maintained, or reinforced by social observation. We observed mothers who interacted with their firstborn babies and found that mothers are more affectionate toward and play more with attractive infants (as rated by college students) compared with mothers of less attractive infants. In addition, mothers of less attractive infants perceive them as interfering more in their lives than do mothers of more attractive infants. Note that these results do not suggest that mothers treat their unattractive infants badly–all the mothers in this study were excellent moms; however, the attractiveness of an infant seems to influence maternal behavior even among excellent moms.


As children grow older, attractiveness continues to influence their interactions with others. Physical attractiveness affects whom children choose as peers and what traits they attribute to other children.emg1

Attractive children are liked more, are perceived as being smarter, and are rated higher on sharing and friendliness and lower on meanness and hitting than less attractive children. Furthermore, this finding is just as true for children who know each other as it is for children who do not.

When children are paired with a peer of the same level of attractiveness, they tend to show more friendly behaviors than children paired with a peer of a different level of attractiveness.

Children’s attractiveness stereotypes may be strengthened and maintained as a result of schematic information processing. We presented young children (ages 3 to 7) with a picture portraying two story characters who varied in attractiveness. The characters displayed positive or negative traits that were either consistent or inconsistent with the “beauty is good” stereotype. When the children were asked questions about the characters, they made more errors identifying female characters with stereotype inconsistent traits than female characters with stereotype consistent traits. This shows that level of attractiveness affects information children remember about females.