The best known, least understood behaviour
Everybody yawns: babies in the womb, adults in meetings, dogs at the vet. In the 1980s Robert Provine was struck by how little yawning had been studied, and began to study it the way an ethologist studies a bird’s courtship dance. He asked participants to think about yawning, recorded the yawns that came and measured how long they lasted.
A yawn lasted 5.9 seconds on average and, once begun, ran to completion with what Provine called the inevitability of a sneeze. Participants who tried yawning with clenched teeth said it was not satisfying at all, as if they were stuck in mid-yawn.
Seeing, reading, thinking
In Provine’s experiments, 55% of people watching a video of yawning faces yawned; with smiling faces in the same set-up the figure was 21%. But seeing was not required. Of those who read about yawning for five minutes, 30% yawned, against 11% of those reading about hiccups; counting people who yawned or felt the urge to, the figures were 75% and 11%. Simply being told to “think about yawning” was enough to set off yawns.
In 1989 Provine went a step further and looked for the part of a yawning face that makes us yawn. The answer was unexpected: not even the gaping mouth was necessary. A yawning face with the mouth removed could still trigger yawns. Observers responded not to a single feature but to the face as a whole.
Why a comparison group?
People also yawn spontaneously when bored or tired. Provine and Hamernik found that people who watched a dull colour-bar test pattern yawned about 70% more than people who watched music videos. In other words, looking at a screen, reading or being bored can make you yawn on their own.
So to tell whether a yawn is contagious, two groups have to be compared: one sees yawning faces, the other sees the same faces doing something else, such as smiling. The difference between the groups shows the contagious share of yawning. In our experiment, too, half of the participants see yawning faces and half see smiling ones.
Yawning and empathy
The idea that contagious yawning is linked to empathy is popular. Some studies reported that yawns from close people are more contagious, or found links with empathy scores. But the picture is mixed.
In a study of 328 people, Bartholomew and Cirulli (2014) found that once age was taken into account, measures such as empathy, emotional contagion, sleepiness and time of day were not related to susceptibility to contagious yawning; only age explained a small share. Gallup and Massen (2016) likewise questioned a claimed difference between women and men in contagious yawning, pointing out that many earlier studies had found no such difference.
Who yawns and who doesn’t?
In the same study, 68% of people who watched a three-minute yawning video yawned at least once. 94% of participants believed they sometimes, often or usually yawned when someone else did, so people somewhat overestimated their own susceptibility. 59% of those who did not yawn said they did not even feel the urge: most were not holding back, they simply did not yawn.
Even more interesting, a person’s susceptibility was very consistent over time: measured about two and a half months apart, in the lab or online at home, the same people gave similar results. This consistency suggests that susceptibility to contagious yawning may be a stable, individual trait.
What does this experiment do?
For ninety seconds, six different drawn faces appear on screen; each yawns or smiles for about six seconds. In the yawning group, short texts about yawning run under the faces; in the comparison group the same space holds everyday topics such as maps, bread yeast or compasses. You press the button each time you yawn.
At the end we ask whether you felt the urge to yawn and whether you held back a yawn, because some people stifle yawns when they feel watched. What we measure in the crowd is the difference between the two groups in the share of people who yawned at least once, together with its 95% confidence interval.