Facial Expression - Universality of Facial Expressions

Universality of Facial Expressions

The Universality Hypothesis is assumption that certain facial expressions are signals of specific emotions (happiness, sadness, anger, fear, surprise, disgust) that are recognized by people everywhere, regardless of culture or language. The evolutionary basis of these kinds of facial expressions can be traced back to Darwin’s "The Expression of Emotion in Man and Animals." Reviews of the Universality Hypothesis have been both supportive (Ekman, Friesen, and Ellsworth, 1972; Izard, 1971) and critical (Woodworth and Schlosberg, 1954; Russell, 1994).

Support for the Universality Hypothesis

Ekman's work on facial expressions had its starting point in the work of psychologist Silvan Tomkins. Ekman showed that facial expressions of emotion are not culturally determined, but universal across human cultures.

To demonstrate the Universality Hypothesis, Ekman tested the South Fore people of New Guinea, a pre-industrial culture that was isolated from Western culture. The Fore were told brief stories about emotional events (happiness, sadness, anger, fear, surprise, disgust). After each story they were asked to select the matching facial expression from an array of three faces (Ekman & Friesen, 1971). The Fore selected the correct face on 64%-90% of trials – but had difficulty telling the fear face from the surprise face. Children selected from an array of only 2 faces and their results were similar to adults’. Subsequent cross-cultural studies with other cultures found also similar results (Ekman, 1987).

Questioning the Universality Hypothesis

People on both sides of this debate agree that the face expresses emotion. The question is: What specific emotional information do we read from a facial expression? Is a specific discrete emotion as claimed by the Universality Hypothesis? Or is it other more general information which we then piece together with other contextual information to determine how a person feels (Aviezer; Carroll & Russell, 1996)?

One difficulty with the evidence presented in support of the Universality Hypothesis is the method typically used to demonstrate universality inflates recognition scores (Russell, 1994). Although each factor may contribute only a small to the inflation, combined they can produce exaggerated scores.

1. The Universality Hypothesis is about our ability to recognize spontaneous facial expressions as they occur naturally. Yet the facial expressions used to test this hypothesis are posed. Studies of spontaneous facial expressions are rare and find that participants’ recognition of the expressions is lower than of the corresponding posed expressions (Matsumoto et al., 2009; Naab & Russell, 2007).

2. In most studies, participants are shown more than one facial expression (Ekman recommends six of each expression). But people judge facial expressions relative to others that they have seen (Thayer, 1980) and participants who judge more than one facial expression have higher recognition rates than those who judge only one (Russell, 1994).

3. The response format that is most commonly used in emotion recognition studies is forced choice. In forced choice, for each facial expression, participants are asked to select their response from a short list of emotion labels. The forced choice method determines the emotion attributed to the facial expressions via the labels that are presented (Russell, 1994). That is, participants will select the best match to the facial expression even if it is not the emotion label they would have provided spontaneously and even if they would not have labeled the expression as an emotion at all (Wagner, 2000).