start » program » presentations

"In a sense, the face is equipped to lie the most and leak the most, and thus can be a very confusing
source of information during deception"
Ekman and Friesen, 1969, p.98

“And how can you possibly know that I have told lie?”
“Lies, my dear boy, are found out immediately, because they are of two sorts.
There are lies that have short legs, and lies that have long noses.
Your lie, as it happens, is one of those that have a long nose.”
Collodi, “Pinocchio”, 2007, p.74

The smile is universally recognized from facial expressions or so it has been claimed. It has been observed in many studies that happy or smiling faces are recognized more quickly and with higher accuracy that other facial expressions (Ekman & Friesen, 1982). Nevertheless, judgments of observers whether the person is lying or not are not better than chance (e.g. Ekman & Friesen, 1974; Ekman, 1988). Poster presents two studies on real versus fake smile detection. First study focuses on the effect of time exposure (200 ms, 5000 ms), nationality (Poles, Swedes) and gender (male, female) on detecting fake smile from photographs. Time, nationality and gender turned out to be significant. The longer time the higher participants judge sincerity, authenticity and trust of presented people. The aim of second study, beside studying the accuracy of recognition Duchenne and fake smile, was to find out where participants search for clues about authenticity of smile. Two series of pictures presenting real and fake smile were presented to subjects in two exposure time 500 ms and 5000 ms. Subjects were asked to assess sincerity of presented person, authenticity of smile, trust and liking for them. In addition, eye tracker was used to record data on eye movement and fixation on the interesting areas of the face: eyes and mouth. The results showed that subjects made significantly more fixations on eye area compared to mouth area in both time exposure conditions. Smiles from pictures, both real and fake, were significantly higher assessed in case of short exposure compared to longer one. No statistically significant differences were obtained between the assessment of real and fake smiles - subjects were not able to distinguish real from fake smiles. Results given above stand for "Othello Error" (Ekman, 1992) which is when a truthful person (in this case people presenting Duchenne smile) is perceived as to be lying. Low accuracy in real versus fake smile detection might be inconsistent with significantly longer time spent on fixation on the eye area, because it might means that subjects are able to see engagement of orbicularis oculi muscle (which orbits the eye, moves cheeks up, bags skin below eye and make crow’s feet wrinkles) and they are unable to perceive it as a clue for realness of smile. All results will be discussed in poster presentation.

Keywords: detection, facial expression, fake smile, smile, Othello error