PNAS (Proceedings of the National Academy of Sciences) is a well respected body in America and beyond.
It has been interesting to see and hear reactions to this claim and also to the credibility of PNAS publishing what some claim to be a poor quality paper.
Cliff Lansley has shared his PERSONAL view on this study within the LinkedIn Forum for Facial Action Coding System (FACS) + Emotional Awareness
I struggle with the whole alignment of this study. The research does not support the claim. The hypothesis that ‘facial expressions of emotion are not culturally universal’ is about the display of the emotion. The universality hypothesis does not claim that all humans communicate emotions universally – the universality hypothesis is more to do with, as the paper states later, the argument that “six basic internal human emotions (i.e., happy, surprise, fear, disgust, anger, and sad) are expressed using the same facial movements across all cultures”.
These are two different statements.
This study seeks to evaluate subjective, mental representations of facial expressions of emotions and this injects a major contaminant. This alone is reason enough reject the claim of the research. Paul Ekman’s research focuses on those displays which happen, unbidden, as a result of experiencing an emotional episode.
The ‘Discussion’ in the report attempts to deal with this perception versus production argument. But it doesn’t. It argues that, “the facial movements perceived by observers reflect those produced in their social environment because signals designed for communication (and therefore recognition) are those perceived by the observer”. As soon as the paper moved onto “cultural groups who use them (facial expressions) for social communication” this took this, for me, firmly into the field of conscious gestures – not emotionally triggered facial expressions.
We know from the extensive research over the last 40 years that facial expressions of some (7) emotions are displayed universally across cultures, sometimes without consciousness, though they are not always judged accurately. Nor can they be always reconstructed consistently. For example, when some people are asked to draw or imitate a sad face, a common expression that is created consists of pursed lips, tight eyelids and brows down – similar to when a child sulks. Yet genuine felt sadness is universally displayed with inner brows up, relaxed eyelids and mouth corners down – the first and the third components here being very hard to manipulate at will by most people.
Eyeball movements do not feature in the universality of expression hypothesis – yet they play a major part in this study. Display rules of different cultures are mentioned yet not developed enough in terms of these being learned choices about what to display on the face and sometimes come into play after the genuine emotional display has occurred on the face – thereby creating the micro-expression of the genuine, universal emotion followed by the (often) conscious suppression or masking that occurs in some groups, families and cultures when we are in the presence of others. This might explain the differences in intensity scores. Blended emotions where one emotion moves into another feature here too of course.
So has this study helped to stimulate debate? – yes. Does it successfully argue that ‘facial expressions of emotion are not culturally universal’ – no.
What it might do is help to describe the differences in conscious displays of emotion in certain cultures – though these are not necessarily connected with the evolved, unbidden, facial expressions of emotion that good research has shown to be universal.
Update: January 2013
PNAS have now published a letter that critiques the Glasgow study and states the results “do not refute “universalists” accounts of emotional communication”.
Sauter and Eisner question the methodology and highlight that the claim is not supported by their data.