What Facial Activity Can and Cannot Tell us About Emotions

Abstract

The belief that facial activity is linked to emotional states has a long history in science and in popular belief. Particularly the 1872 publication of Charles Darwin's book The expression of the emotions in man and animals (1998) has had a major impact on the role of emotional expressions in modern emotion research and has largely reinforced the commonly held belief that there is a link between emotions and expressive behavior. Milestone research, particularly by Paul Ekman and his colleagues has cemented the common notion that faces express emotions. However, essentially all of the involved researchers, including Darwin and Ekman have usually maintained a view that is far more differentiated than the distilled version that has resulted from repeated summarizing and synthesizing in the secondary and tertiary literature. In fact, no current researcher holds that all emotions are always reflected in facial or other nonverbal, activity. Inversely, almost all theorists will concur that oftentimes facial expressions are not associated with a concordant affective state, or for that matter, any affective state at all. This chapter will take stock of the current controversies in the debate regarding the relationship of facial activity and affective state. Specifically, I argue that there are strict limits to inferences regarding underlying affective states based on measurements of facial activity. The arguments raised here should be of particular interest to those who intend to apply findings regarding the relationship of emotion and facial actions, such as in clinical settings for initial diagnosis or accompanying the therapeutic process. In fact, any context in which attempts are made to interpret facial activity as a means of determining underlying affective states (or action tendencies, motivations, attitudes, or other states of mind) is affected by oversimplifications of the empirical data available to us today. Yet, regardless of the criticism offered here, I do not reject the potential usefulness of measuring facial activity for basic research and in applied contexts. In my mind there is no doubt that one of the origins of the production, and of the perception of facial actions is genetically determined and that, at times, there are indeed links between facial actions and the aforementioned states of mind, Instead, I argue that those interested in using measures of facial actions as tools to augment, or even replace, measures of subjective experience, should be aware of issues that have been identified and partially solved in connected domains, such as in psychophysiology.