We do it automatically. As soon as we observe another person, we try to read his or her face for signs of happiness, sorrow, anxiety, anger. Sometimes we are right, sometimes we are wrong, and errors can create some sticky personal situations. Yet Paul Ekman is almost always right. The psychology professor emeritus at the University of California, San Francisco, has spent 40 years studying human facial expressions. He has catalogued more than 10,000 possible combinations of facial muscle movements that reveal what a person is feeling inside. And he has taught himself how to catch the fleeting involuntary changes, called microexpressions, that flit across even the best liar's face, exposing the truth behind what he or she is trying to hide.
Ekman, 72, lives in Oakland, Calif., in a bright and airy house near the bay. As I talked with him there, he studied me, his eyes peering out from under bushy brows as if they were registering each brief facial tic I unknowingly exhibited. Does his talent make him a mind reader? "No," he says candidly. "The most I can do is tell how you are feeling at the moment but not what you are thinking." He is not being modest or coy; he is simply addressing the psychological bottom line behind facial expressions: "Anxiety always looks like anxiety," he explains, "regardless of whether a person fears that I'm seeing through their lie or that I don't believe them when they're telling the truth."
The professor calls the ever present risk we all take of misreading a person's visage "Othello's error." In Shakespeare's drama, Othello misinterprets the fear in his wife Desdemona's face as a sign of her supposed infidelity. In truth, the poor woman is genuinely alarmed at her husband's unjust, jealous rage. Othello's subsequent decision to kill Desdemona is a fatal error, and Ekman wants to make sure that police, security personnel and secret service agents do not make the same mistake. "Arresting the guilty is a good thing," he acknowledges, "but decreasing the number of innocent people who are falsely accused is just as important." His system for understanding the emotions that faces portray, and his expertise in applying it, could help all kinds of law-enforcement and legal personnel in their work. It could also help the rest of us better negotiate how our family members, friends and colleagues really feel.
On supporting science journalism
If you're enjoying this article, consider supporting our award-winning journalism by subscribing. By purchasing a subscription you are helping to ensure the future of impactful stories about the discoveries and ideas shaping our world today.
Face Code Deciphered
The very fact that psychologists are studying the emotion of facial expressions at all is due in large measure to Ekman's work. When he began studying psychology at the University of Chicago in the 1950s, emotions were a neglected subject at the periphery of the discipline. Many researchers believed that an individual's emotional world was inaccessible to scientific scrutiny--or at least was less interesting than, say, the mechanisms of learning and thinking or the motivations behind human actions.
Ekman, however, was fascinated by the mystery of nonverbal communication. He wanted to understand why some people had little trouble decoding the feelings of others, almost as if they were reading an open book, whereas others fell for one con artist after another. His motto was: trust your eyes, not conventional wisdom. The widespread belief then was that facial expressions arose simply from cultural learning: a child in a given culture learned the faces that accompanied particular emotions by observing people, and over time different cultures developed different expressions. Even renowned researchers such as anthropologist Margaret Mead were unconvinced of the existence of a universal repertoire of expressions, as Charles Darwin had proposed in his book The Expression of the Emotions in Man and Animals, published in 1872 but subsequently ignored.
To test his own hunch, Ekman headed for Brazil with a stack of photographs in his suitcase. The portraits showed sad, angry, happy or disgusted faces of white Americans, yet Brazilian college students had no trouble identifying the feelings depicted. Expeditions to Chile, Argentina and Japan generated the same results; regardless of where he went, local people seemed to understand, and use, the same facial expressions as the North Americans.
Concerned that perhaps inhabitants of "modern" societies had somehow cross-pollinated their facial movements, Ekman in 1967 visited extremely isolated tribes living in the jungles on the island of New Guinea. There again, though, he found that the basic emotions he had postulated, such as happiness, sadness, anger, fear, surprise and disgust, were associated with universal facial expressions. The excursion sealed it for him: the language of the face has biological origins, and culture has no significant effect on it.
This recognition raised a whole new set of questions. How many different facial expressions are human beings capable of? What precisely does a particular expression signify? Is it possible to learn how to read emotions? Ekman decided to create a sort of common dictionary of facial expressions, and he set about doing so with a mixture of meticulousness and daring.
"If I had known how long it would take to set up such a system, I might never have begun," he says now, with a slight sigh. "At the time, we didn't even know whether a person can make the same expression twice--whether his expressions always differ, even if only in minor ways."
Perceiving Microexpressions
Ekman and his U.C.S.F. colleague psychologist Wallace Friesen spent six years formulating their Facial Action Coding System (FACS), which they published in 1978. The system makes it possible to describe and classify any facial expression based on a combination of 43 facial-action units [see box on next page]. The 43 elements yield more than 10,000 possible combinations. Ekman and Friesen catalogued each combination by a FACS number, the Latin names for the muscles involved and the associated emotion, if any. For example: "1; inner brow raiser; frontalis, pars medialis, is one element of sadness."
One interesting aspect of this inventory is that many muscle combinations signify absolutely nothing. Ekman discovered another interesting phenomenon after spending the day in his laboratory trying to reproduce a convincing look of sadness: that evening he realized that he was feeling depressed. He then found that if he spent time engaged in imitating the components that make up a smile, his mood lifted. "That was like an epiphany," he recalls. It contradicted the naive notion that feelings originate solely in the psyche and that the body merely communicates them outwardly.
Ekman and Friesen were able to demonstrate that the coordinated tightening of certain facial muscles not only affected blood pressure and pulse rate but also could trigger the corresponding emotion. It seemed clear that a feedback mechanism was at work between the facial muscles and the brain's emotion centers.
Such a linkage caught the attention of psychologists, and by the early 1980s FACS started to be applied to real-world situations. Practitioners wanted to know how they might ascertain whether their patients were telling the truth by watching their faces. Such a talent could be critical, as an old videotape Ekman had made proved. The tape showed a psychiatric patient named Mary, who, apparently recovered from a severe bout of depression, begged her treating physician to allow her to spend the weekend at home. The doctor approved her request, but luckily before leaving, Mary admitted that she had planned to commit suicide.
Ekman had already studied the tape; he told viewers that if facial expressions indeed unveil a person's true feelings, they should be able to read Mary's intentions. Most viewers did not see the telltale sign at first, so Ekman pointed it out. He originally had watched the video over and over, often in slow motion so as not to miss even the slightest detail. And suddenly he saw it: for the briefest moment, a look of sheer desperation could be seen flitting across Mary's face. Such microexpressions--which often last no more than a fifth of a second--were the key. Regardless of how stoic we try to look or how heartily we laugh off a situation, the control that we can exert over our own facial features has its limits. Our true feelings always leak out, even if only for an instant.
When he discovered microexpressions, Ekman was teaching at U.C.S.F., and he spent several years putting together a self-teaching program that enabled people to decode faces according to the FACS system. By paying close attention to microexpressions, people can learn to read signals that previously would have been perceptible only in slow motion. And here Ekman hit on another interesting phenomenon: most people--including law students, police officers, judges and prosecuting attorneys--find it difficult to expose fakers, but a small number of people seem to be able to correctly interpret microexpressions intuitively. Apparently, some of us are born with handy lie detectors.
This capability should make more than a few fibbers--kids, criminals and politicians among them--very uneasy. On the other hand, certain individuals can learn to be convincing prevaricators. "Think of a chess player who controls his emotions and sets his facial expression so the other player will interpret it in a certain way," Ekman suggests. In addition, the more an individual believes his or her own fabrications, and the more often he or she serves them up successfully, the more difficult it will be for others to see through the deception. "Lies that are being told for the first time and that have an emotional component are the easiest to expose," Ekman says. That is why he recommends that interrogators ask their questions quickly and with an element of surprise. For example, instead of asking, "Were you in the parking lot of the Wal-Mart yesterday evening at six o'clock?" it is better to ask, "Where do you usually buy household items?"
Hard Truth
Although it is possible to learn to recognize microexpressions, Ekman warns against using them as a clear-cut indicator of a lie. Whenever he trains security personnel, he emphasizes the importance of asking a suspect how what the person just said made him or her feel. The response will make it less likely that an investigator will commit Othello's error. In addition, questioners need to pay attention to details other than facial expression, such as small shifts in posture, speech or hand gestures, all of which could indicate a fabrication [see "Gestures Offer Insight," by Ipke Wachsmuth, on page 20]. Unless the suspect is Pinocchio, there is no unambiguous proof of a lie.
As to why so many people find it difficult to recognize deception, Ekman says, "Many people simply want to believe what they are being told, even if they really know better. Who wants to find out that your spouse is being unfaithful with your best friend? Or that your kids are using hard drugs? You should want to, but it's terrible when you discover it. And if you knew this, you'd have to do something about it; most of us are pretty avoidant."
From an evolutionary perspective, it would not necessarily have been advantageous for humans to be perfect lie detectors. In small, close-knit groups, little falsehoods are frequent and help group members gloss over unimportant mishaps or inequities. If every lie was singled out, the resulting confrontations would almost certainly do more harm than good. In the end, the smooth talkers would probably be expelled from the group, weakening its number if nothing else, and none of those remaining would have gained any benefit from the expulsion.
When it comes to hunting down terrorists, however, the ability to unmask has real survival implications. Ekman spends a great deal of time training antiterror specialists, even though he has been retired for over two years. Nevertheless, he is well aware of the narrow scope of his methods. "The tools I have to offer are pretty modest," he notes.
He also sees parallels between his work and that of the Dalai Lama, whom he has met several times at conferences. Like the spiritual leader, Ekman wants to help people understand their own feelings and master their impulses. "The only area where I differ with the Dalai Lama is on the issue of reincarnation," he muses.
Before we end our discussion at his home, I ask him about his own relationship to the truth. Ekman considers the question for a moment. "I have a golden rule," he responds, "according to which I decide when a lie is permitted. I ask myself how the other person would feel if he found out that he had been lied to." If the person would feel betrayed or taken advantage of, then the lie would be damaging.
Vigilantly offering the truth at every daily interaction and social occasion, though, may be of little value. "Would you tell your hosts that you were bored out of your mind at their party?" Ekman asks. "You see, no one would expect that--not even from an expert on lying."