Web Historical Disclaimer:

This is a historical page and is no longer maintained. Read our Web history statement for more information.


spacer
DISCOVERY MAGAZINE

spacer

Dr. Judith H. Langlois
is the Charles and Sarah Seay Regents Professor of Developmental Psychology at The University of Texas at Austin.


The Question of Beauty


When I took my Ph.D. examination all candidates were grilled by five professors about their dissertation to make sure they "know their stuff." My dissertation was an innocuous study of preschool children's peer relationships, and I knew the data and the relevant literature like the back of my hand. I answered all the questions quickly and articulately. Except one. A professor asked me if I had considered whether the children's facial attractiveness influenced their peer relationships and popularity and, therefore, my results. My immediate reaction was to roll my eyes back and say emphatically, "No, of course not!" But then flashbacks of college blind dates crossed my mind. I remembered well the first question asked by both males and females: "What does he or she look like?" I was uncertain how to answer my professor's question. On the one hand, I was taught "never judge a book by its cover," implying that attractiveness should not be important. On the other hand, there were all those blind dates....... I stumbled and stammered. My professor gently let me off the hook by telling me that perhaps I should think about the question before I did my next study.

That simple dissertation question became the impetus of a research effort originally conceived to demonstrate that the professor was wrong and facial attractiveness was not important for children and their relationships. Some twenty years and many research studies later have revealed that the professor was right and I was naive about the importance of attractiveness in child development. Facial attractiveness, I emphasize, is not the most important influence on children's relationships and development; intelligence and personality are both more important than attractiveness. Through my research, however, I now recognize that attractiveness makes a significant and meaningful contribution, more than previously believed.

Myths About Beauty. Many conventional truisms about attractiveness are really myths. Despite the old adage "never judge a book by its cover" and despite the prevailing belief that attractiveness does not matter once we know a person, even parents judge and treat their own children differently based on attractiveness, although they are not aware of it. In a study of more than 150 Caucasian, Mexican American, and African American newborn infants and mothers, we found that moms of attractive first-born infants were more attentive and affectionate than moms of less attractive first-borns. All the mothers denied that attractiveness should matter in parental treatment of children but their behavior belied their beliefs.

Many studies, in addition, demonstrate that facial attractiveness is a significant correlate of children's popularity in the classroom, where the children are among familiar peers. One study we conducted shows that attractiveness is significantly related to social acceptance and popularity for girls throughout the entire school year. For boys, low attractiveness is associated with rejection by peers. Moreover, the likelihood that unattractive boys would be rejected increased, not decreased, as the school year progressed and as the boys became better acquainted.

These studies on parent-infant interaction and children's popularity directly contradict the notion that attractiveness becomes less important as people become better acquainted. Many studies also contradict the conventional wisdom that "beauty is all in the eye of the beholder." A 1960 study by psychologist A. H. Iliffe establishes vast agreement in attractiveness judgments made by more than 4,000 men and women ranging in age from fifteen to fifty-five and from different regions of the United Kingdom. Michael Cunningham and his colleagues point out that attractiveness ratings of Asians, Hispanics, Blacks, and Whites made by Asians, Hispanics, Blacks, and Whites are highly related. Even the attractiveness of young babies can be reliably judged by college students, even though they initially complain that all babies look alike!

Even more surprising, we discovered that infants, as young as three- to six-months of age, agree with the attractiveness judgments of adults, suggesting that the standards and preferences for beauty are not learned gradually through exposure to the media but rather are in place early in life. How do we know that babies prefer attractive faces given that they cannot talk to us? To answer this question we've conducted two different types of studies to assess infants' preferences for attractive faces.

The first study is a visual preference experiment. In these experiments, we showed a group of babies several slides of faces reliably rated by adults to be more or less attractive. The babies saw the faces of either Caucasian male or female adults, African-American adult females, or the faces of other infants. In each experiment, eight attractive and eight unattractive faces, matched closely for hair length, style, color, and facial expression, were chosen from a large group of faces. Pairs of these faces were projected side-by-side so that each image was about the size of a real face. The amount of time each infant looked at each face was then recorded. The results of all these studies were straightforward and unambiguous. Babies look longer at adult-judged attractive faces than at unattractive faces, regardless of whether the face is male or female, white or black, adult or infant.

We wondered whether these visual preferences of infants extended to other infant behaviors. To answer this question, we asked a professional mask maker to design and construct attractive and unattractive masks for a woman who would later interact with infants as a "stranger." The masks were realistic and life-like. They were thin enough to move the stranger's face so she could smile and talk freely. (Each mask, similar to those seen on the television program "Mission Impossible," took about 2.5 hours to apply to the stranger's face for an experimental session.) The stranger played with sixty one-year-olds, one at a time, using a strict, rehearsed script so that her behavior would be consistent for all infants. The stranger did not know whether the attractive or unattractive mask had been applied each day because the masks' interiors were identical. Thus her behavior toward the infants could not have been affected by her knowledge of whether she was "attractive" or "unattractive." The interactions between the stranger and the babies were recorded on videotape and scored by observers who could not see the stranger's face and who, therefore, could not be biased by the mask she wore.

The results showed that infants' preferences for attractive females do indeed extend beyond visual preferences to include actual behavioral differences. The infants more frequently avoided the stranger when she was unattractive than when she was attractive, and they showed more negative emotion and distress in the unattractive than in the attractive condition. Furthermore, boys (but not girls) approached the female stranger more often in the attractive than in the unattractive condition, perhaps foreshadowing the types of interactions that may later occur at parties and other social situations when the boys are older!

It appears that what our mothers and grandmothers said was true-that standards of attractiveness were gradually learned through exposure to the media-is actually wrong. Standards of and preferences for attractive faces are either innate or acquired much earlier than previously supposed.

Defining Beauty. What exactly IS facial beauty and why is it that infants, children, and adults all seem to prefer attractive faces? Defining beauty is a task scientists and philosophers have tackled for centuries, largely without success. Recent empirical work on the effects of facial attractiveness, nevertheless, has proceeded without any conceptual or scientific definition of attractiveness. Researchers simply have defined attractive faces as those that raters agree are attractive! Even Webster's Collegiate Dictionary (10th Edition) is not much help: "Beauty. The quality or aggregate of qualities in a person or thing that gives pleasure to the senses or pleasurably exalts the mind or spirit."

For our understanding of "what is facial beauty," we wanted and needed a more precise and scientific definition to help explain why infants, young children, and adults from various parts of the world could so easily agree on which faces are attractive and which are not. After searching all the existing research literature, we located a series of historic studies that showed that faces close to the average facial configuration of the population of faces should be attractive.

In the 1800s English anthropologist Sir Francis Galton and American psychologist George D. Stoddard created composite portraits by superimposing photographic exposures of faces. Essentially they constructed imprecise mathematical averages of faces by using this photographic technique. Galton's goal was to design facial types; he was especially fond of fabricating composites of criminals and vegetarians. Stoddard assembled composites of members of the National Academy of Sciences and of the 1883 and 1884 graduating classes of Smith College. Both Galton and Stoddard noticed and remarked that the composites were "better looking" than their individual components.

Why might "average" faces be preferred over non-average faces (not "average" in attractiveness, but average in facial configuration and proportion)? A cognitive theory emphasizes the role of prototypes in understanding how a person processes information. A prototype is an abstract, cognitive representation that reflects the best example of a category of objects or events, and it often is defined as the central tendency or the averaged members of the class of examples. For example, an average sized dog, neither very large nor very small, is usually represented as the prototypic dog in the human mind. So a dog that represents the average of all dogs would be the prototypic dog and a face that represents the average of all faces in the population would be the prototypic face. Many studies show that after seeing several examples of a category-for instance, schematic animals or schematic faces-one responds to an averaged representation of those category members as if it were special or familiar to us even if we have never seen it before. Studies even show that young infants can average incoming stimuli to form a prototype. Perhaps this is why infants in our research preferred attractive faces; if attractive faces are prototypical of the category of faces, they would be preferred because they seem more facelike or more familiar to the young infants.

A second theory also emphasizes the importance of "averageness." The field of evolutionary biology suggests that perhaps innate, built-in mechanisms account for these preferences for attractive faces. In the most common form of natural selection, called normalizing or stabilizing selection, evolutionary pressures operate against the extremes of the population and in favor of characteristics representing the average or central tendency of the population. Thus average values of characteristics shaped by normalizing selection are preferred in the population. According to this view, individuals with average population characteristics should be less likely to carry harmful genetic mutations than individuals with extreme population characteristics. In fact, evolutionary biologist Donald Symons has proposed an innate mechanism of perception that detects the population mean of anatomical features. For faces, Symons refers to this mechanism as a "beauty detector." The "beauty detector" averages observed faces and, because of stabilizing selection pressures, prefers these "average" faces over faces more distant from the mean.

Both cognitive and evolutionary theory suggest that faces representing the average of the population will be perceived as attractive as did Galton's work in the 1800s. We examined the hypothesis that averaged faces would be attractive by performing a "high tech" version of Galton's technique. We photographed a substantial number of UT Austin male and female undergraduates using a standard background, lighting, and distance. We randomly selected ninety-six male faces and ninety-six female faces and randomly put them into three sets of thirty-two faces for each sex with no overlapping faces in the sets. We then created three "composite" or averaged faces for each sex on a computer in a two-step process.

In the first step, we "digitized" the individual faces (Figure 2) by converting the light and dark shades that comprise each face to an array of numbers, or chromatic values, that represent each face and can be manipulated just like any other set of numbers. Each color value represents a different shade present in the picture of the face.

In the second step, we mathematically averaged the numbers representing the different individual faces in each of the six sets (Figure 3). We created composite faces of two, four, eight, sixteen, and thirty-two different faces averaged together for each set of randomly selected faces. These averaged faces were photographed and rated for attractiveness by 300 judges along with the photographically equivalent (in other words, slightly blurred) slides of the individuals who went into them.

We found that averaged faces made of sixteen-and thirty-two-faces were judged to be significantly more attractive than the average attractiveness level of all the individual faces (Figure 3). Additional analyses indicated that, of the ninety-six individual male faces, only three were judged to be significantly more attractive than their thirty-two-face composite-about what is expected by chance. Of the female faces, only four were rated as significantly more attractive than the composites, again only the number one would expect by chance (Figure 4).

By using advanced computer technology, we demonstrated that "averaged" faces are perceived as attractive; we replicated this finding in two populations, male and female, and in three samples from each population. Although we do not think that "averageness" is the only aspect of facial beauty (expression and age are important as well), we do believe that "averageness" is a necessary and critical element of attractiveness. Without "averageness" even the most youthful, smiling face will not be judged as attractive.

Although evolutionary and cognitive theory are generally considered quite different theoretical accounts of human behavior, they both posit similar mechanisms in the case of preferences for attractive faces by suggesting that prototypic or averaged faces underlies the tendency of infants and adults from diverse cultures to notice and prefer attractive faces. At this point, we can't choose between evolutionary theory, which suggests that preferences for attractive faces are innate, and cognitive theory, which suggests that preferences for attractive faces are acquired early in life through exposure to category members. Indeed one of the most exciting aspects of this work is that we will eventually be able to contribute some answers to the age-old nature vs. nurture debate: what capabilities are we born with and what capabilities are developed due to experience?

Caveat. I end with a cautionary note. After hearing of this work and about the "myths" of attractiveness, one might wonder if this research on the nature of facial beauty is in some way an advocacy of the importance of beauty, an approval of the emphasis on beauty in the media, or a suggestion that because preferences for attractiveness are evident so early in life, they are an immutable aspect of human nature. The answer to all three questions is no! It is true that even the youngest of us fall victim to the "beauty-is-good" bias automatically, and that often we are not aware that we have such biases in favor of attractive individuals. As cognitive humans, however, we are capable of controlling many aspects of our thoughts and behaviors, an ability that distinguishes us from lower animal species and allows us to change undesirable aspects of our behavior. Studying and examining these unconscious influences, such as biases toward facial beauty, help us become aware of how and when they operate and thus allow us to learn to oppose them. The same research that identifies these unconscious processes can rob them of their mystique and influence and can lead us to behave more consciously and humanely.

Acknowledgments: I thank all the superb graduate and undergraduate students who have worked as my colleagues in this research. I am grateful to all the parents who so generously gave their time to bring their infants to the Children's Research Lab to participate in our research. I also thank the National Institute of Child Health and Human Development, the National Science Foundation, the National Institute of Mental Health, The University of Texas Research Institute and the Charles and Sarah Seay endowment for providing the funding for this research program.


HOME | | | NEWS | FACTS | ON CAMPUS | DISCOVERY | ABOUT O P A

August 6, 1997
Comments or suggestions to utopa@www.utexas.edu
Copyright © 1997 Office of Public Affairs
at The University of Texas at Austin