A study carried out by the Universities of California, Buffalo and Toronto has discovered that a computer system is more accurate at spotting fake facial expressions than people.
Leading recruiters and coaches look out for micro-expressions, the tiny little facial movements we make without realizing what we’re doing. When we’re faking a smile, our eyes tend not to crease and the areas between our eyebrows and eyelids don’t contract. As these movements occur in milliseconds, a computer can pick up on these minutiae. People find it difficult.
The researchers of this latest study used a computer expression recognition toolbox that was able to work in real time to assess whether people were telling the truth. The participants of the study put their hands in ice water – some people suffered real pain, others faked it. People found it hard to distinguish between the fakers and the real pain sufferers, and even when they were trained in micro-expressions and what to look for, they still were accurate only 55 percent of the time. The computer got it right 85 percent of the time.
Knowing if someone is telling the truth is critical in the recruitment process. An employer wants to know if candidates are being honest, candidates want to know that they aren’t being over-sold to. It doesn’t take much stretching of the imagination to foresee candidates of the future being given the once over by a recruiting robot!
The study, “Automatic Decoding of Deceptive Pain Expressions,” is published in the March 2014 issue of Current Biology.