An interview with Alexander Todorov, author of Face Value: The Irresistible Influence of First Impressions
What inspired you to write this book?
I have been doing research on how people perceive faces for more than 10 years. Typically, we think of face perception as recognizing identity and emotional expressions, but we do much more than that. When we meet someone new, we immediately evaluate their face and these evaluations shape our decisions. This is what we informally call first impressions. First impressions pervade everyday life and often have detrimental consequences. Research on first impressions from facial appearance has been quite active during the last decade and we have made substantive progress in understanding these impressions. My book is about the nature of first impressions, why we cannot help but form impressions, and why these impressions will not disappear from our lives.
In your book, you argue that first impressions from facial appearance are irresistible. What is the evidence?
As I mentioned, the study of first impressions has been a particularly active area of research and the findings have been quite surprising. First, we form impressions after seeing a face for less than one-tenth of a second. We decide not only whether the person is attractive but also whether he or she is trustworthy, competent, extroverted, or dominant. Second, we agree on these impressions and this agreement emerges early in development. Children, just like adults, are prone to using face stereotypes. Third, these impressions are consequential. Unlucky people who appear “untrustworthy” are more likely to get harsher legal punishments. Those who appear “trustworthy” are more likely to get loans on better financial terms. Politicians who appear more “competent” are more likely to get elected. Military personnel who appear more “dominant” are more likely to achieve higher ranks. My book documents both the effortless nature of first impressions and their biasing effects on decisions.
The first part of your book is about the appeal of physiognomy—the pseudoscience of reading character from faces. Has not physiognomy been thoroughly discredited?
Yes and no. Most people today don’t believe in the great physiognomy myth that we can read the character of others from their faces, but the evidence suggests that we are all naïve physiognomists: forming instantaneous impressions and acting on these impressions. Moreover, fueled by recent research advances in visualizing the content of first impressions, physiognomy appears in many modern disguises: from research papers claiming that we can discern the political, religious, and sexual orientations of others from images of their faces to private ventures promising to profile people based on images of their faces and offering business services to companies and governments. This is nothing new. The early 20th century physiognomists, who called themselves “character analysts,” were involved in many business ventures. The modern physiognomists are relying on empirical and computer science methods to legitimize their claims. But as I try to make clear in the book, the modern claims are as far-stretched as the claims of the old physiognomists. First, different images of the same person can lead to completely different impressions. Second, often our decisions are more accurate if we completely ignore face information and rely on common knowledge.
You mentioned research advances that visualize the content of first impressions. What do you mean?
Faces are incredibly complex stimuli and we are inquisitively sensitive to minor variations in facial appearance. This makes the study of face perception both fascinating and difficult. In the last 10 years, we have developed methods that capture the variations in facial appearance that lead to specific impressions such as trustworthiness. The best way to illustrate the methods is by providing visual images, because it is impossible to describe all these variations in verbal terms. Accordingly, the book is richly illustrated. Here is a pair of faces that have been extremely exaggerated to show the variations in appearance that shape our impressions of trustworthiness.
Most people immediately see the face on the left as untrustworthy and the face on the right as trustworthy. But notice the large number of differences between the two faces: shape, color, texture, individual features, placement of individual features, and so on. Yet we can easily identify global characteristics that differentiate these faces. Positive expressions and feminine appearance make a face appear more trustworthy. In contrast, negative expressions and masculine appearance make a face appear less trustworthy. We can and have built models of many other impressions such as dominance, extroversion, competence, threat, and criminality. These models identify the contents of our facial stereotypes.
To the extent that we share face stereotypes that emerge early in development, isn’t it possible that these stereotypes are grounded in our evolutionary past and, hence, have a kernel of truth?
On the evolutionary scale, physiognomy has a very short history. If you imagine the evolution of humankind compressed within 24 hours, we have lived in small groups during the entire 24 hours except for the last 5 minutes. In such groups, there is abundant information about others coming from first-hand experiences (like observations of behavior and interactions) and from second-hand experiences (like testimonies of family, friends, and acquaintances). That is for most of human history, people did not have to rely on appearance information to infer the character of others. These inferences were based on much more reliable and easily accessible information. The emergence of large societies in the last few minutes of the day changed all that. The physiognomists’ promise was that we could handle the uncertainty of living with strangers by knowing them from their faces. It is no coincidence that the peaks of popularity of physiognomists’ ideas were during times of great migration. Unfortunately, the physiognomists’ promise is as appealing today as it was in the past.
Are there ways to minimize the effects of first impressions on our decisions?
We need to structure decisions so that we have access to valid information and minimize the access to appearance information. A good real life example is the increase of the number of women in prestigious philharmonic orchestras. Until recently, these orchestras were almost exclusively populated by men. What made the difference was the introduction of blind auditions. The judges could hear the candidates’ performance but their judgments could not be swayed by appearance, because they could not see the candidates.
So why are faces important?
Faces play an extremely important role in our mental life, though not the role the physiognomists imagined. Newborns with virtually no visual experience prefer to look at faces than at other objects. After all, without caregivers we will not survive. In the first few months of life, faces are one of the most looked upon objects. This intensive experience with faces develops into an intricate network of brain regions dedicated to the processing of faces. This network supports our extraordinary face skills: recognizing others and detecting changes in their emotional and mental states. There are likely evolutionary adaptations in the human face—our bare skin, elongated eyes with white sclera, and prominent eyebrows—but these adaptations are about facilitating the reading of other minds, about communicating and coordinating our actions, not about inferring character.