Can Smiling Make Things Funnier?

New research has tested the facial feedback hypothesis

Posted Dec 27, 2018

A classic study in social psychology had participants rate how humorus they found various written comics. The kicker was that half the participants rated the comics with a pencil between their teeth, and half with a pencil between their lips. What this was testing was whether or not the act of smiling — by having the pencil between your teeth — increased the amount of enjoyment people found in the comics. 

The result, that people when made to smile using this simple technique found more humor in the comics, became a classic study and arguably the foundation of the facial feedback hypothesis. This hypothesis states that people's facial movements can influence their emotions. That is, it isn't just that we smile when we are happy, but to an extent we are also happier, because we are smiling, or have been smiling. 

The facial feedback hypothesis has come under much criticism due to failures to replicate the various findings in the field. One recent manuscript reported 17 failed attempts, for instance, to replicate the original facial feedback study.

A group of psychologists at Hebrew University of Jerusalem, however, noticed an inconsistency between the original study testing the facial feedback hypothesis and these failed replication attempts. Specifically, the replication attempts involved the usage of a video camera recording the study. The original study did not have a camera, nor were people being observed while looking at the comics.

This is problematic to the facial feedback hypothesis, because when people are observed, they rely less on internal feelings when making observations and judgments. As such, a camera being present might reduce the amount that people link their internal feelings to their judgements of humor, basically destroying the very foundation of the facial feedback effect. It doesn't matter if a person is smiling or not when rating comics if they are not relying on the emotions that it generates, in other words.

Consistent with the facial feedback hypothesis, the original finding replicated when people were not being observed with a camera. It did not replicate, however, when people were being observed with a camera. That is, putting a pencil between your teeth (smiling) increased the humor in the comics, but only if people were not being observed.

These findings are of note for several reasons. First, they demonstrate that the facial feedback hypothesis exists. We can, even if minimally, alter our enjoyment of stimuli through our facial reactions. Second, they show the power of being observed on human judgments. The impact of the experimental manipulation — what face people were forced to make — was altered merely by the idea that people might be being watched. Third, it shows that when a study fails to replicate, it might be because it doesn't mirror the design of the original study close enough, not because the original study is flawed. 


Noah, T., Schul, Y., & Mayo, R. (2018). When both the original study and its failed replication are correct: Feeling observed eliminates the facial-feedback effect. Journal of Personality and Social Psychology, 114(5), 657-664.