A Meta-Analysis of the Facial Feedback Literature: Effects of Facial Expressions on Emotional Experience are Small and Variable

The facial feedback hypothesis suggests that our facial expressions influence our emotional experience. In light of Wagenmakers et al.’s (2016) failure to replicate Strack, Martin, and Stepper’s (1988) seminal demonstration of facial feedback effects, a meta-analysis was conducted on 286 effect sizes derived from 136 facial feedback studies. Results revealed that the overall effect of facial feedback on affective experience was significant, but small (d = .20, p < .000000005).Approximately 70% of variation in facial feedback effect sizes is due to heterogeneity, which suggests that facial feedback effects are stronger in some circumstances than others. Eleven potential moderators were examined, and three were associated with differences in effect sizes: (1) Type of affective reaction: Facial feedback influenced emotional experience (e.g., reported amusement) and, to an even greater degree, perceptions of stimuli’s affective quality (e.g., funniness of cartoons). However, after controlling for publication bias, there was little evidence that facial feedback influenced perceptions of affective quality. (2) Presence of emotional stimuli: Facial feedback effects on emotional experience were larger in the absence of emotionally evocative stimuli (e.g., cartoons). (3) Type of stimuli: When participants are presented with emotionally evocative stimuli, facial feedback effects were larger in the presence of some types of stimuli (e.g., imagined scenarios) than others (e.g., pictures).