Monday, 12 November 2012

Most people can fake a genuine "Duchenne" smile

For years, the literature on the psychology of smiling has claimed that fake smiles can be easily and reliably distinguished from genuine smiles by the absence of crinkling around the eyes. The eye crinkling of a supposedly real "Duchenne smile" (named after a French physician with a fondness for electrodes) is caused by activation of the orbicularis oculi muscles, which raises the cheeks. The traditional view is that this muscle is not within voluntary control, unlike the zygomatic major muscle that bends the mouth upwards into a smile. Fake smiles therefore feature the upturned mouth but there's something missing in the eyes, or so it was long claimed.

Doubts first emerged in a 2009 paper, in which Duchenne smiles were produced just as often when participants pretended to be amused, as when they were genuinely amused. Now a research team led by Sarah Gunnery has provided more evidence that undermines the old beliefs about Duchenne smiles being a reliable sign of true positive emotion.

Gunnery and her colleagues had 96 student participants (49 men) pull smiling faces into a camera while role-playing genuine positive emotion (e.g. pleasure at a good exam grade) or while role-playing fake positive emotion (e.g. smiling in response to a gift that's not really liked).

Overall, 28 per cent of the smiles were rated by two experienced coders as Duchenne smiles, with the characteristic crinkling around the eyes. This broke down as 31 per cent for positive situations and 24 per cent in the fake positive situations. When naive viewers rated these smiles, they tended to say the Duchenne smiles were more genuine, but this was largely because eye crinkling tended to go hand in hand with more expressive smiling around the mouth.

Next, the participants were presented with a photograph of a person pulling a Duchenne smile and another showing a "fake" smile with no eye crinkling, and their task was to imitate both. Seventy-one per cent successfully imitated the Duchenne smile, and 69 per cent successfully imitated the fake smile.

These results explode the myth that it's not possible to fake a "genuine" Duchenne smile. They also hint at this being a skill that varies from person to person. It was the same participants who tended to display Duchenne smiles in the various conditions of the experiment. Moreover, these Duchenne participants reported feeling that they'd done a good job in the tasks, and they said they were able to pull fake expressions in their daily lives, all of which suggests they have good insight into their facial abilities.

A weakness of the study is its reliance throughout on staged emotion. While the evidence is clear that many people can fake the Duchenne in neutral conditions (albeit while imagining emotional scenarios), we don't know how easy it is for people to do this under conditions in which they truly are experiencing negative emotion. On the other hand, because there were no explicit instructions in the role-playing tasks to pull a Duchenne smile, nor were there any consequential outcomes to provide extra motivation, the prevalence of the ability to fake Duchenne smiles in neutral conditions may actually have been underestimated.

"Findings from the present study strengthen the argument that people can volitionally activate their cheek raiser muscle and put on a Duchenne smile," Gunnery and her team concluded. "Future research will further investigate individual differences, and will use behavioural outcomes to measure similarities in people who deliberately produce the Duchenne smile."


Gunnery, S., Hall, J., and Ruben, M. (2012). The Deliberate Duchenne Smile: Individual Differences in Expressive Control Journal of Nonverbal Behavior DOI: 10.1007/s10919-012-0139-4

Post written by Christian Jarrett for the BPS Research Digest.


charlesgraves said...

Online education programs include the same options that are available in a conventional university or college setting. Liberal arts and sciences are partly available online. Other programs that will require a physical presence can also be partially taken online. Click here for the details.

David Winter said...


I've always been suspicious about the 'you can't fake a Duchenne smile' claim. Have these people never watched actors?

Through an admittedly non-scientific process of observation, I have a suggestion for a slightly better indicator of genuineness which is to do with the onset and decay rates of a smile (or other expression). Onset rate is the speed at which muscles contract to produce the smile and decay rate is the speed at which the muscles relax to remove the smile.

My casual observations suggest that a genuine, involuntary smile is characterised by a rapid onset (the point of involuntary reflexes is to kick in quickly) and a slow decay (as the firing of the involuntary reflex gradually dies down or as your voluntary systems struggle to take control of your muscles from the involuntary system).

Fake smiles are likely to have a slower onset (as you consciously exert control over muscles you don't use habitually) and a faster decay (as you release those muscles back into their natural state).

I suspect that, with practice, even this can be faked.

This is one of the reasons why I'm always suspicious of any research into facial expressions which is based on static photographs. That's not how we pick up signals from facial expressions in the wild.

I don't know of any research along these lines.

Cliff Lansley said...

I like your points about onset and offset David - I notice that genuine emotions are characterised by a smoother offset of the key facial expressions in most people.
In the hundreds of people we have engaged in facial expression related work I have also noticed that the genuine emotions are smoother (onset and offset), longer (a few seconds) and more symmetrical than when they are faked (by forcing the reliable muscles that create the expression) - though this has not been documented or published yet.
Paul Ekman's research with Mark Frank ("Not all smiles are created equal"- see below) highlighted the five main differences as follows:
1. the presence of the orbicularis oculi[OO] action (outer eye crunch) in conjunction with the zygomatic major (lip corner raise) - I will come back to this,
2. symmetrical action of the zygomatic major[ZM]
3. smooth ZM action
4. consistent durations of ZM action from one enjoyable episode to the next
5. synchronicity between OO and ZM - peaking at the same instant.

The first one is the point of this article and I would challenge most people to voluntarily engage their outer OO (not a squint with the inner OO) to produce the crunching that often creates crows feet and a narrowing of the upper eye cover fold - WITHOUT engaging the mouth.

Of course most of us can all generate these genuine smiles if we are amused or recall a happy moment.

The challenge for those reading faces is that many people can produce the EFFECTS of OO contraction by forcing a strong ZM which pushes the fatty tissue of the cheeks upwards to create a similar appearance. Most of those we train or those who have been through FACS training ARE able to differentiate this from the genuine smile.

Article can be read here
Humor 6-1 (1993), 9-26 c Walter de Gruyter

Unknown said...

I have total ability to "crinkle my eyes" whenever I want. I can flash a Duchenne smile at will. And yes, people seem to respond better to my Duchenne smile than just a fake smile.

I this Duchenne crinkle effect alone sufficient to give people the impression that a person is happy? This I have often wondered about. I know when I am trying to look happy, but don't feel like smiling I just do the eye thing, but I have no idea what people think of this. I think it makes me look more relaxed and happy.

Anonymous said...

I naturally did this since a child. I just connected the emotion to the smile and was able to perform the "Duchene Smile" when I didn't feel "that" way. So when I perform a "Duchene Smile" I'm basically synthesizing the emotion in order to make it. I assumed everyone could do that. I guess I picked up on people's reaction when I smiled a certain way. I didn't want to appear rude to people.

Unknown said...

The physician with a fondness for electrodes was French, not Dutch. Interesting subject, though.

Unknown said...

now corrected, thanks.

Anonymous said...

Yup, same here.
The nice thing is that when I really turn on the Watts, doing the sparkly eyes bit, I feel kinda warm n fuzzy inside.

Anonymous said...

Okay well this just destroyed half the research for my project ... which is due tomorrow...

Maikeru1333 said...

body language can also influence your mood, ie fake a smile and start feeling happier...‎

Maikeru1333 said...‎

body language can also influence your mood, ie if you engage the muscles that smile, it can lead to you actually feeling happier, same with confidence and power, or the opposite, weakness and helplessness...

Anonymous said...

Are you people serious? What a bunch of bladders full of hot air. No one ever said you can't fake it or crinkle your eyes. The point is that when the original study was done, they didn't tell participants to fake a smile when they get stimuli. They just tested them. When they tested with a stimulus that was clearly not really going to illicit a genuine smile, people faked it. When they gave genuinely smile inducing stimuli, participants produced a duchenne smile.

Post a Comment

Note: only a member of this blog may post a comment.