Of course if you watch the Paul Ekman inspired programme “Lie to Me” you know it’s all in the NVC micro-expressions and only about 1 in 400 people have the natural ability to understand them.
Even people you might expect to do well at spotting liars such as police and security officers rarely do better than chance without extensive training. You might as well toss a coin. (If you don’t believe me have a go at judging whether or not these smiles are genuine).
There are methods experts can use in controlled settings, such as analysing written content, and getting people to illustrate their stories or tell them in reverse order but you can hardly do that in normal social situations.
Scientists no longer believe there is one single non-verbal signal that gives people away but more likely a cluster of NVC signals or behaviours.
Psychologists and scientists at MIT carried out a series of experiments to ascertain if NVC could convey trustworthiness. Strangers were asked to conduct a 5-minute conversation together either face-to-face or via web chat (but without using emoticons) to see whether or not they were better at judging trustworthiness face-to-face.
The participants then played a game that measured cooperation and self-interested economic behaviour.
Those who had chatted face-to-face beforehand were more accurate at predicting the trustworthiness of a stranger.
So what was missing from the text only conversation that was present in the face-to-face ones? Independent judges were asked to view the videotapes of the latter and identify all possible NVC clues such as smiling, nodding, touching etc.
Then they looked for specific clues when the volunteers successfully detected other people’s self-serving intentions. The body language that appeared repeatedly in those sessions was hand touching, face touching, crossing arms, and leaning away. Taken together they indicated deceitful behaviour and the more often they did these behaviours the less they were trusted.
But are these the only clues and what about people who fidget anyway? To take the experiment further the researchers used a robot called Nexi which is designed to emote with humanlike facial expressions and body language. They asked some of the participants to have a 10-minute conversation with it. The robot was manipulated to carry out the four behaviours linked with deceitfulness in different orders and with some repeats to stimulate normal fidgeting.
Other participants were also asked to have a conversation but during those conversations Nexi used gestures other than the four associated with deceitfulness. When Nexi used the deceitful movements people said they felt a sense of distrust but not with the other movements. And during a replaying of the game with Nexi they expected to be cheated and cooperated less.
An interesting aspect of the experiment was that although participants were skeptical of Nexi’s motives when it exhibited deceitful NVC they didn’t necessarily dislike it. Just as in real life you can like people you don’t necessarily trust.
Source: “How to Spot a Scoundrel” in Scientific American Mind September/October 2012
FYI Just how good are you at recognising emotions?