If only our kids smiled more, we'd buy them more T-shirts (but probably not vodka)

On our recent trip to Europe, we had a hard time getting the kids to smile for pictures. Most of our pictures of Nora ended up looking something like this (actually this one's a self portrait, but you get the idea):

Here her expression is basically neutral, and if it wasn't such a dramatic shot, it would be a bit boring.

When we could get her to smile, often the smile was inauthentic -- posed, or even sarcastic, like in this shot:

Here she's expressing mock excitement over her parents' excitement about the figure depicted in the statue: Leonardo Fibonacci, the great mathematician (click on the photo for a closer view).

Authentic smiles generally came at inopportune moments for photography, such as this one:

Why couldn't Nora have smiled like this in front of the Parthenon and saved her inauthentic smile for laughing at the irony of being served water in a vodka glass?

The primary feature of a genuine smile was identified by Duchenne de Boulogne in 1862: "crow's feet," or wrinkles in the outside corner of the eyes (click on the picture above to see them on Nora). Subsequent research has found that people readily differentiate between real and posed smiles, and like the people in pictures better when they're sporting a real smile. Other research has found that people respond more positively to advertisements when the models in the ad are smiling -- but until now no study has examined the difference between real and posed smiles in models.

Victoria Peace, Lynden Miles, and Lucy Johnston filled that gap by asking female volunteers to rate T-shirts modeled by women with different facial expressions. The t-shirts were actually identical except for a color manipulation in Photoshop, and the different facial expressions were made separately by pulling from a database of emotional expressions. The same model was either smiling genuinely, smiling artificially, or showing a neutral expression. Thanks to some nifty Photoshop work, viewers saw her wearing each shirt, and with each facial expression. After each photo, they indicated whether they liked the shirt. Here are the results:

Significantly more viewers preferred the shirts when they were attached to the authentically smiling face than either the posed smile or the neutral face. In fact, there was no difference in responses to a posed smile or a neutral face.

Next Peace's team decided to see how quickly the effect occurred. In a new experiment they flashed each picture for just 150 milliseconds, then asked whether viewers liked each shirt. While some shirts were preferred over others, there was no effect of facial expression. So while facial expression has an impact on preferences, it takes at least 150 milliseconds for that impact to be observed.

But if facial expression can't affect preferences quickly, perhaps it's not a lingering effect. Maybe if we're distracted by another task, the authenticity of the smile won't matter. In a third experiment, the team told viewers they simply needed to identify the color of each T-shirt as quickly as possible. This time there were just three shirts, and each viewer saw only one facial expression with each shirt. As soon as the color was identified, the picture disappeared. At the end of the test, viewers were asked to say which shirt they preferred. Even though the facial expressions were evenly distributed among all three shirts, significantly more viewers preferred the shirt worn by the genuinely smiling model compared to the neutral or posed smile.

So even when distracted by another task, we still prefer items paired with a genuine smile compared to posed smiles.

And that gets me thinking.... You know what I'd like right now? A nice, cold glass of vodka! (It's 5 o'clock somewhere, right?)

Peace, V., Miles, L., & Johnston, L. (2006). It doesn't matter what you wear: The impact of posed and genuine expressions of happiness on product evaluation. Social Cognition, 24(2), 137-168.