Don’t say we didn’t warn you. You might be able to fool some people, but you’re unlikely to do the same thing with a computer. A new study from researchers at the University of California in San Diego and the University of Toronto suggests that computer software is able to spot genuine or fake expressions of pain far more accurately than humans are able to do.

The study, “Automatic Decoding of Deceptive Pain Expressions”, reveals that computers are able to identify the distinct dynamic features of facial expressions that are often missed by humans. According to the paper, there are two main components of the software – computer vision and machine learning – which is able to achieve an 85 percent accuracy rate compared to just 55 percent for humans, even after the latter has had many hours of practice and training.

Marian Bartlett, lead author of the study and research professor at Institute for Neural Computation in UC San Diego, says that their tests are proof that humans leave a lot to be desired when it comes to deciphering which expressions of pain are real and which are not. While most people can fake expressions well enough to convince others, it’s far more difficult to fool computers into believing the same thing.

“Perceptual processes that are very easy for humans are hard for computers,”said Bartlett. “This is one of the first examples of computers being better than people at a perceptual process.”

The scientists’ methodology was based on the idea that genuine and fake expressions take different pathways in the brain, and also on the Facial Action Coding System described by psychologist Paul Ekman in the 1970s, which states that each movement we make is built on a specific muscle or set of muscles.

“In highly social species such as humans, faces have evolved to convey rich information, including expressions of emotion and pain. And, because of the way our brains are built, people can simulate emotions they’re not actually experiencing – so successfully that they fool other people. The computer is much better at spotting the subtle differences between involuntary and voluntary facial movements,” said Kang Lee, a professor at the Dr. Eric Jackman Institute of Child Study at the University of Toronto, who also authored the study.

The researcher’s tests show that the most common giveaway if you’re faking an expression is the way you open your mouth. When someone is faking they’re in agonizing pain, they tend to open and close their mouths in a regular pattern, whilst those in genuine pain do so more randomly.

The software has obvious implications for sports like soccer, where faking injury to win free kicks is a common problem, but the researchers think their system could be used in other areas too.

“In addition, our computer-vision system can be applied to detect states in which the human face may provide important clues as to health, physiology, emotion, or thought, such as drivers’ expressions of sleepiness, students’ expressions of attention and comprehension of lectures, or responses to treatment of affective disorders,” said Bartlett.

The paper is called “Automatic Decoding of Deceptive Pain Expressions” and is published in the latest issue of Current Biology.

photo credit: Vox Efx via photopin cc