Born to Be Good: The Science of a Meaningful Life

How then can we provide a coherent analysis of a category of behaviors—smiles—that includes my McD smile as well as the loving smiles of old friends and parents and children? At first glance, the empirical literature on the smile yields similarly paradoxical findings: People have been shown to smile when winning, losing, watching a film of an amputation, eating sweets, facing adversaries, experiencing pain, feeling affection toward loved ones. The answer is provided by Paul Ekman, and it involves looking away from the lip corners to that wellspring of the soul—the eyes.

A vocabulary of smiles comes sharply into focus when we consider the activity of the happiness muscle, the orbicularis oculi. This muscle surrounds the eyes and when contracted leads to the raising of the cheek, the pouching of the lower eyelid, and the appearance of those dreaded crow’s-feet—the most visible sign of happiness—which the Botox industry is trying to wipe out of the vocabulary of human expression. People may think they look prettier following Botox injections, but their partners will receive fewer clues to their joy, love, and devotion.

Ekman has called smiles that involve the activation of the zygomatic major muscle and the orbicularis oculi the Duchenne or D smile, in honor of the French neuroanatomist Guillaume Benjamin Amand Duchenne (1806–1875), who first discovered the visible traces of the activity of orbicularis oculi. Smiles that do not involve the activity of the happiness muscle, the orbicularis oculi, are sensibly known as non-Duchenne or non-D smiles. To try your hand at this subtle distinction between Duchenne and non-Duchenne smiles, see if you can detect which is which in the photographs below (answers provided on chapter 6).

Dozens of scientific studies speak to the importance of parsing the heterogeneous category of smiles according to the activity of the orbicularis oculi muscle. Duchenne smiles differ morphologically in many ways from the many other smiles that do not involve the action of the orbicularis oculi muscle. They tend to last between one and five seconds, and the lip corners tend to be raised to equal degrees on both sides of the face. Smiles missing the action of the orbicularis oculi and likely masking negative states can be on the face for very brief periods (250 milliseconds) or very long periods (a lifetime of polite smiling by oppressed airline stewardesses and fast-food servers). Non-D smiles are more likely to be asymmetrical in the intensity of muscle firing on the two sides of the face.

D smiles tend to be associated with activity in the left anterior portion of the frontal lobes, a region of the brain preferentially activated during positive emotional experiences. Non-D smiles, in contrast, are associated with activity in the right anterior portion of the brain—a region associated with the activation of negative emotion. When a ten-month-old is approached by his or her mother, the face lights up with the D smile; when a stranger approaches, the same infant greets the approaching adult with a wary non-D smile.

And importantly, several studies have found that Duchenne and non-Duchenne smiles, brief two-to three-second displays differing only in the activation of the orbicularis oculi muscle, map onto entirely different emotional experiences. For example, in a longstanding collaboration with my friend George Bonanno, a pioneer in the study of trauma (see chapter 7), we interviewed middle-aged adults six months after their deceased spouse had passed away. These individuals were asked to describe their relationship with their deceased spouse for six minutes. I spent a summer coding the occurrence of Duchenne and non-Duchenne smiles from videotapes of these narratives. We then related measures of bereaved participants’ D and non-D smiles to their reports of how much enjoyment, anger, distress, and fear they felt during the interview, which we gathered immediately after the participants had finished talking about their deceased spouse.

Portrayed in the table below are the correlations between how much participants showed these brief Duchenne and non-Duchenne smiles and their ensuing self-reports of emotion gathered moments later. Positive scores indicate that the more they showed the particular smile during the six-minute interview, the more they subsequently felt the particular emotion listed on the left. Negative correlation values reveal the opposite, that the more the participant smiled in Duchenne or non-Duchenne fashion, the less of the emotion they felt. Asterisks indicate that the observed correlation was statistically significant, and not likely produced by chance.

               

DUCHENNE SMILES        

NON-DUCHENNE SMILES        

ENJOYMENT        

.35*        

–.25*        

ANGER        

–.28*        

.09        

DISTRESS        

–.49*        

–.16        

FEAR        

–.31*        

.04