Faking It Isn’t Making It: Research Needs Spontaneous and Naturalistic Facial Expressions

Abstract Facial expressions play a pivotal role in shaping social interactions. However, the conceptualization of facial expressions as direct readouts of internal emotional experience has led to the conflation of three distinct question types. Specifically, there is confusion between questions concerning: (Q1) the production of facial expressions, (Q2) how accurately perceivers interpret expressors’ internal emotions from their outward expressions, and (Q3) perceiver responses to the outward appearance of expressions independent of the expressor’s internal emotional state. The disentanglement of these three question types highlights that, because the facial stimuli traditionally used in research are posed rather than reflective of internal emotions, they can only test perceiver responses (Q3), though they have often been interpreted as measures of perceptual accuracy (Q2). Moreover, due to their exaggerated and prototypical nature, these stimuli fail to capture the nuance and complexity of real-world expressions, potentially leading to ecologically invalid findings even for Q3. New data presented here also suggest that many of these stimuli are not perceived as genuinely emotional and may appear unnatural. We review evidence demonstrating that stimuli that are naturally- or spontaneously-elicited and/or appear genuinely emotional can produce different findings than traditional posed stimuli. Fortunately, naturalistic and spontaneous expression stimuli are now readily available for the field to move forward. We conclude with seven recommendations for advancing facial expression research.