Facial-Feedback Hypothesis

The facial-feedback hypothesis states that the contractions of the facial muscles may not only communicate what a person feels to others but also to the person himor herself. In other words, facial expressions are believed to have a direct influence on the experience of affect. This hypothesis goes back to Charles Darwin, [Page 341]who wrote that the expression of an emotion intensifies it, whereas its repression softens it. A second origin of the facial-feedback hypothesis is William James's theory of emotion, which states that the bodily changes follow the perception of an exciting fact and that the feeling of these bodily changes is the emotion.

Although Darwin and James differ in their view of the role of the eliciting stimulus, they agree that the behavior that accompanies an emotion exerts a causal influence on its experience. In particular, the skeletal muscles were identified as important contributors. While Darwin has assigned the facial muscles a special role as means of expression and has meticulously described their evolutionary significance (e.g., frowning), James's account is based on more global units of behavior (e.g., running away).

To test the causal influence of facial expressions on the experience of affect, three different procedures have been employed. In some experiments, participants were explicitly instructed to adopt an emotionally relevant facial expression. In another set of studies, the emotional meaning of the expression was not mentioned. Instead, the experimenter would point at the muscles that were supposed to be contracted. In yet a third method, facial expressions were induced by a procedure that required the contraction of specific muscles for a purpose that was void of any emotional meaning. For example, participants were told to hold a pen with either their teeth or their protruded lips to either induce or inhibit a smiling expression by extracting the zygomaticus muscle (one of the main muscles involved in making the mouth into a smile) or its antagonist. In a related study, golf tees were fixed on people's foreheads, which they had to move together by contracting the corrugator (frowning) muscle.

All procedures were successful in causing affective consequences either in people's self-reported mood, in specific emotions, or in the evaluation of emotional stimuli, like cartoons. However, the three facialinduction methods afford different theoretical interpretations. Specifically, the more likely it is that the induction of the facial expression is linked to the recognition of its emotional meaning, the more likely it is that people may infer their affective state on the basis of their expression. For example, they may draw the inference that if they smile, they must be amused. This mechanism is an extension of Bem's selfperception theory, which assumes that if internal cues are weak or ambiguous, people infer their attitudes from their behavior. Similarly, they may infer their emotional states from what they do. However, the fact that affective consequences can be obtained from facial expressions even if their emotional meaning is disguised suggests that more direct mechanisms may be operating as well.

While self-perception theory may account for the cases in which the meaning of the expressions is apparent, other models are necessary to explain the direct impact of the facial action. On a physiological level, it has been argued that facial expressions may regulate the volume and particularly the temperature of the blood that flows to the brain and therefore influence cerebral processes. It was suggested that an emotional event may cause peripheral muscular, glandular, or vascular action that changes the emotional experience. Another explanation that is based on evidence from the neurosciences comes from a study that identifies specific cortical activities that are connected to different facial expressions. Specifically, it was found that the facial expression of emotions that are linked to approach (e.g., joy) were associated with greater left frontal brain activity while avoidance emotions (e.g., fear and anger) were linked with greater right frontal activation.

...