Do People See Emotions In Your Face That Aren't There?


New research suggests that women have a tendency to attribute more hostility to other women’s faces than is actually present or intended. The notion of “resting bitch face,” or the tendency for a woman’s neutral facial expression to appear angry or annoyed, has caught on in the media, including a recent piece in the New York Times. A series of new studies by Jaimie Arona Krems and colleagues at the University of Arizona suggests that whether or not we see anger in a woman’s neutral face may be partially due to a tendency to see them as our sexual competition.

Krems and colleagues investigated how women perceive anger in other women. According to the researchers, women face challenges in understanding if other women are threats for two reasons:
Women are more likely than men to eschew physical aggression in favor of indirect aggression —gossiping, excluding others, and back-stabbing (just think of the movie Mean Girls). This makes women’s aggression more subtle and difficult to detect.

Women may be more likely than men to mask their anger, making it especially tough to spot.

To determine if women really do mask their expressions of anger, particularly toward other women, the authors surveyed 218 adults about how likely they would be to put on various facial expressions in a given situation. (For example, participants were asked to describe what sort of facial expression they would put on if they were at a funeral, but found something to be funny.) Of particular interest to the researchers was the expression they would put on when angry with a stranger. Some participants were asked to imagine they were angry at a male stranger, others at a female stranger. In most cases, participants reported they’d be more likely to show an angry face than a neutral one when mad at a stranger. The exception to this pattern: Female participants reporting how they would react when angry at another woman. Women imagining how they’d react when angry at a female stranger claimed they were more likely to put on a neutral than an angry expression.

Given that women tend to mask their anger at other women, Krems and colleagues propose that women have developed a defense against aggression from other women. Specifically, the authors claim, women may be biased toward perceiving anger in other women’s faces, even if those women’s expressions are emotionally neutral. The authors argue that it’s more adaptive to have a few false alarms (perceiving non-threatening people as threatening) than to fail to notice a real danger. So women will err on the side of caution and over-perceive anger on the faces of other women.

To test this hypothesis, the authors conducted an additional experiment, in which 88 participants viewed a series of photographs of people making neutral facial expressions. They were told that each person had just relived a particular emotion (anger, fear, pride) and then had tried to hide the emotion with a neutral expression. They were also told that traces of the emotion might still linger on the face, in the form of microexpressions. In fact, unbeknownst to the participants, all of the photos showed a completely neutralexpression, randomly paired with the emotional descriptions. Participants were then asked to rate the extent to which they felt each person was expressing various emotions.

The results showed that women were more likely than men to perceive anger in the female faces. And this wasn’t because women were just generally more likely to over-perceive anger in all faces: Women were more likely to see anger in female than in male faces. Moreover, this pattern of results only occurred for anger, and not other emotions, suggesting that it’s specific to anger and not the case that women just read more emotions (of any type) into each other’s faces.

The authors also hypothesized that women who are sexual competition for other women, and thus potential targets for their aggression, would be the most prone to seeing anger in other women’s faces. Other research has shown that women who are sexually desirable or more sexually available tend to be the targets of this kind of aggression.4,5 Thus, the authors predicted that women who see themselves as sexually desirable would be especially prone to see anger in other women’s neutral face expressions.

In a third study of 56 adults, the authors repeated the emotional perception experiment, but this time, they also asked participants to report on their own sexual attractiveness and sexual availability (i.e., being open to casual sex and reporting more sexual partners). Their results once again revealed the bias for women to see anger in other women’s neutral faces—but, in addition, women who perceived themselves as desirable, or who were sexually available, exhibited this bias more than those who saw themselves as less desirable or available. The authors concluded that this bias is especially adaptive for such women because they are, in fact, more likely to be under threat from other women.

I would still be curious how these effects play out when women are actually interacting with one another, rather than looking at photos out of context. Would additional situational context make women more or less likely to see anger in neutral faces? And does this bias extend beyond facial expressions to women’s interpretation of tone of voice, gestures, or other behaviors? I would also suspect that the bias would be diminished or disappear if participants in the study had not beenexplicitly told that the people in the photos were masking their emotions.

This study also raises the question of whether or not this bias is likely to lead women to more accurately perceive other women’s emotions. In the last two experiments, all of the facial expressions really were neutral, but as the authors found in the first study, women’s neutral expressions with other women are more likely to mask angry feelings. So in the real world, the bias documented by the researchers may actually lead to more accurate perceptions. On the other hand, if actual neutral expressions are being seen as angry in a broad array of contexts, it would be more likely that the bias is leading women to false conclusions about each other.

The study’s authors explain their findings in terms of sexual competition. However, it’s also quite possible that social norms contributed to the results. Women may mask anger because anger is seen as especially socially unacceptable for women to display. As the first study shows, women are aware of their own tendency to mask anger, so they may infer that other women are similarly likely to mask anger. Therefore they may project their own tendencies onto other women they encounter. Men, who may be less aware of women’s tendency to mask anger, have less reason to suspect that a woman’s neutral face is hiding anger, and thus don’t have the same biased perceptions.

These findings still make me wonder if the perception of “resting bitch face” is born, in part, from sexual competition between women. Despite the popular notion that accusations of resting bitch face are products of sexism, the researchers found that men were just as likely to see anger in resting malefaces as in resting female faces. It’s women who seem more likely to perceive “resting bitch face.”