IE 11 is not supported. For an optimal experience visit our site on another browser.

You can't read my poker face, science confirms

When Serena Williams or Rafael Nadal make a great save or win a tennis match their faces fill with joy and they excitedly pump their fists. When they miss a save or lose a match, their faces flood with disappointment and they angrily punch their fists.  It turns out that extreme excitement and extreme anger look very similar. We believe we distinguish these feelings by gazing at people’s faces, reading them for subtle changes in emotion. But researchers recently found that people more accurately understand mood when they examine body language.

“When people rate a whole image, it was clear to them—they saw winners and losers,” explains Hillel Aviezer, now an assistant professor of psychology at the Hebrew University of Jerusalem. “When you see the faces alone, it is really confusing.”

Aviezer—who was a postdoctoral student with Alexander Todorov, a professor of psychology at Princeton University and a coauthor of the paper—wanted to see how accurate people were at predicting emotions when they examined facial expressions and body language. Psychologists often ask subjects to describe what someone is feeling by looking at a series of pictures of actors mimicking emotions. But Aviezer wondered how participants would react to more realistic, less scripted visages.

The researchers asked 45 undergraduate students to join one of three groups. One group looked at complete images of tennis players, meaning the photos included faces and bodies, as the athletes reacted to either winning or losing. The second cohort looked only at the faces of tennis players as they triumphed or failed. And the third set examined only the bodies of the tennis players responding to wins or losses.

Students that looked at either the complete photo, including face and body, or just an image of the body more accurately predicted the tennis player’s feelings. Those who saw only the faces had a 50/50 chance of correctly predicting whether the person was happy or angry.

“Many of these findings are pretty surprising because they went against the standard textbook [answer],” says Aviezer.  

Even though it seems that body language leads to more accurate impressions of mood, the participants did not realize it. Fifty-three percent of the participants that ranked the entire image believed they understood the tennis player’s emotion because of her facial expressions.

 “More than half of the participants believe they are reading facial cues … this study gave different results.”

Aviezer also asked another group of participants to look at photos where the face showed one emotion, such as extreme happiness, but the body displayed another feeling, such as extreme anger. Then he asked the subjects to mimic the facial expressions, generally a simple lab activity. If the face looked happy, but the body language projected anger, the students made angry faces, not happy faces.

“They changed their faces based on the bodies. We thought they would be good at imitating a posed face,” Aviezer says.

Overall, the report finds that body language plays a much larger role in understanding emotions. 

“We think the face is very important. [But] we don’t give the body much credit with how much information [it shares].”

The report appeared in the journal Science