The detection of change in facial expression of emotion

People use facial expressions of others as important cues with which to regulate their own behaviour. facial expressions are also monitored during attempts to regulate the emotions of others. If you see that your friend is sad, you might tell a joke. Then you will look at your friend’s facial expression to assess the impact of the intervention. If you see a glimmer of a smile, this might encourage you to pursue your attempts at emotion regulation. But if your friend frowns, you might stop trying to be funny. Changes in facial expressions are monitored in order to assess the attitudes of others. However, perceivers are not always free of bias. People sometimes exaggerate or minimize the nature or the intensity of an emotion expressed facially. A robust predictor of accuracy in the decoding of facial expression of emotion is gender. On average, women decode facial expression of emotion more accurately than men. An exception on this is the facial expression of anger. Men seem to decode anger more accurately and sometimes faster than women. This is particularly the case if the expressor is male. This article presents a new experimental procedure for examining the detection of changes in facial expression of emotion. The method involves the use of computer-generated movies. In these movies, facial expressions change gradually from an emotional expression to a neutral expression. The participant needs to detect the offset of the initial facial expression in the movie.

The perception of facial expression

Much is known about basic processes of face perception, but relatively little is known about the influences of emotion in the perception of facial expression. In most studies, scientists found that people tend to see displayed on other’s faces the same emotion that they themselves are currently feeling. The effect most often observed in relevant research is what can be called an emotion congruence effect. One interpretation of emotion congruence is that emotion-congruent information has facilitated encoding. This means that any emotion congruent facial information should be preferentially encoded and also weighted more heavily in judgments of the nature of the facial expression. There is also another mechanism that can account for the emotion congruence effects. Perceivers can also use the emotional impact of the face on his/her internal emotional state in order to evaluate the facial expression of the target. So, the perceiver could ask himself/herself how the face makes him/her feel. If someone is happy and sees an ambiguous facial stimulus, he/she might misattribute his/her feelings of happiness to the face and thus judge the face to be expressing happiness. In this case, the emotion congruence observed is due to a process of misattribution of emotional state

The study

In the present study, the writers looked at the facial expression of emotion and more concretely, at the perception of expression change. The participants first watched 12 minutes of film clip. This was done to induce certain emotions (happiness, sadness or neutral). They then had to fill in a questionnaire to assess their emotion. Afterwards, participants saw photographs of actors displaying happiness and sadness. These photographs were digitally blended with a photograph of the same actor expressing neutral emotion to create 100-frame movies in which a face initially expressing an emotion became gradually neutral. The participants were told that they would be seeing a series of photographs. A sliding bar was displayed at the bottom of the screen and the participants had to drag the bar from left to right to play the movie and thus change the facial expressions to a neutral emotion. The participants had to move the sliding bar to the right and to stop at the first frame at which they perceived that the face no longer expressed its initial emotion and then to click on a certain button.

The manipulation was successful. Participants in the happy condition were much happier than were neutral emotion participants and sad condition participants. Participants in the happy condition perceived the offset of a happy expression at a later frame in the morph movie than they saw the offset of the sad expression. The participants in the sad condition saw the offset of a sad expression at a later frame in the changing morph movie than they saw the offset of the happy expression. So this means that happy participants saw happy expressions to be present on an increasingly neutral face for a longer time than sad expressions. For sad participants, the opposite pattern was found.

Future research should look into other factors that influence the perception of facial expression of emotion. For instance, anxiously-attached people are extremely attentive to negative emotions and these individuals may see expressions of such emotions persist for a very long time. Avoidently-attached people deny their own negative emotions and they might therefore see expressions of negative emotions disappear very quickly from a face.