Surprise! Facial expressions aren’t necessarily universal

Different people can interpret facial expressions differently. Rishi S

You can tell a lot about a person’s emotional state by looking at their face. A quick glance can give you an idea of whether a person is, say, happy or angry, allowing you to modify your behaviour accordingly.

The rapid and accurate recognition of some emotional states – particularly fear or anger – would have been advantageous in our evolutionary history. For instance, being able to determine when someone is angry with you might give you time to run away before they attack.

For this reason, you might think the way emotions are expressed on the face would be the same across all races and not substantially influenced by culture.

But new research, published in the Proceedings of the National Academy of Sciences by psychologist Rachael Jack and colleagues seems to show this isn’t the case.

What are you looking at?

Facial expressions are produced by the movement of facial muscles, with distinct patterns of movement thought to convey certain expressions.

For instance, the activation of the zygomaticus major muscles leads to an upturned mouth and activation of the orbicularis oculi muscles enhances creases around the eyes, both of which are often associated with happiness.

It is often claimed there are at least five discrete, basic emotions – anger, happiness, disgust, sadness, fear – and perhaps more (surprise and contempt), each with their own characteristic facial expression.

Creases around the eyes can be a sign of happiness. sean dreilinger

One of the first to propose a limited number of biologically based and universal expressions was Charles Darwin in 1872, in The Expression of the Emotions in Man and Animals.

And there is strong support for this position. Certain facial expressions appear in a similar form in humans and primates, as well as infants and young children.

These same facial expressions are even depicted in identical ways by people who are born blind (that is, without the ability to see and copy expressions).

But perhaps the most convincing evidence in support of Darwin’s universality hypothesis came a century after it was first proposed, with psychologist Paul Ekman’s research on cross-cultural facial expression recognition.

Ekman and colleagues demonstrated that a preliterate culture in Papua New Guinea, the Fore, as with 21 literate cultures also studied, could label facial expressions representing anger, happiness, sadness and disgust (although they could not discriminate between surprise and fear).

The Fore also generated facial expressions of basic emotions that were well recognised by other cultures.

Therefore, it seems the capacity to generate and recognise these basic expressions was not learned through media or other social influences. Instead, it had developed in isolation, providing compelling evidence in support of the universality hypothesis.

But Ekman’s research has not been without its critics. Over the past four decades there has been considerable debate about whether facial expressions of emotion are, in fact, universal or whether they are instead shaped by one’s culture.

Which brings us back to the research of Rachael Jack and colleagues.

Their paper, unambigiously titled Facial expressions of emotion are not culturally universal, casts doubt on the universality hypothesis.

Faces on a screen

Jack and colleagues created a computer program that randomly generated thousands of 3D facial movements, some of which formed characteristic patterns of facial expressions (see video below).

Observers, half of whom were Western Caucasian and half East Asian (with little experience of each other’s culture), were asked to categorise each animation as one of six basic emotions – anger, happiness, fear, sadness, disgust or surprise – or “other”, and judge the emotional intensity of each animation.

For Western Caucasian observers, the facial movements categorised as each of the six basic emotions appeared distinct and all observers identified the same emotions.

By contrast, these same distinct movements were not seen in the categorisations made by East Asians, who showed a high degree of overlap in the categorisation of certain movements. This was particularly the case between fear and surprise, as well as anger and disgust.

Other differences were also apparent between the observations of Western Caucasians and East Asians. In particular, the perception of emotional intensity in some expressions was cued by rapid changes to the eye region for East Asians, whereas Western Caucasians took their cues from other parts of the face.

This study, along with other research over the past decade, is forming a consistent argument: there are cultural differences in the kind of facial expression movements thought to constitute certain emotions.

Jack and colleagues argue that if facial expressions were once universal, they have since evolved. If this is the case, one can only wonder what kind of facial signals will evolve in the future to communicate not just internal emotions, but intentions and deceptions.

What then, does this mean for communication between cultures in our increasingly globalised world?

Well, it seems obvious that we can’t assume our facial expressions communicate the same meaning to everyone.

This in turn poses interesting questions: are the feelings that underlie our facial expressions also culturally diverse? Or is it merely the external depiction of our emotions, in the form of facial expressions, that have been shaped by culture?

As always, there is much still to learn.

The Conversation is a non-profit + your donation is tax deductible. Help knowledge-based, ethical journalism today.