In the classic Quentin Tarantino movie True Romance, Italian mafia boss Vicenzo Coccotti delivers a memorable speech about lying. He claims to be a world expert on spotting them, suggesting that the average man has seventeen signals that give the game away, whilst a woman has 20. He claims to know therefore, that his victim is lying, and proceeds to shoot him in the face.
Now, I’m not suggesting the average office encounter is going to end so violently, but nevertheless, being able to detect lies as proficiently as Mr Coccotti would be a valuable skill in any professionals armour.
It is of course a rather tricky thing to do, and previous tests show that so called experts often fare no better at detecting the liar than novices, both of which come in at around 50%. You may as well flip a coin and take your chances.
So when a study comes along that claims to be able to detect 97.8% of liars, that is perhaps something to take notice of.
The research, published in Human Communication Research, suggests that the key to detecting a liar has nothing at all to do with the supposed tells Mr Conccotti could detect. Instead, the key is to question your victim in the right way.
Here’s how the experiment panned out
Participants in the study were asked to play a trivia game. Unknown to their partner however, each participant was given the chance to cheat at the game. In one round of games nearly 45% of people cheated, whilst in the other round (clearly featuring a more honest bunch!), the number was just 12%.
What the participants didn’t know however is that they would be in for a grilling at the conclusion of their game. One group used the Reid Technique to grill their suspect. This method is commonly used in US police forces (although unclear if used in US crime dramas).
The Reid Technique employs a number of tactics, some of which may be well known to aforementioned crime drama aficionados. For instance, you might presume the suspect is guilty, or ask particularly loaded questions.
It undoubtedly did the trick, as all 33 of the cheating participants owned up to their crime under such cross examination.
Another group of hapless participants were then grilled by actual federal agents with years of interrogation experience under their belts. The pros managed to crack 97.8% of the liars, which turned out to be 87 of the 89 in the group!
Now, there is a slight caveat to consider here. Firstly, it’s suspected that such interrogation techniques can lead to false confessions. Important though the game was, it also offered little real incentive for the participants to hide their guilt.
Nevertheless, the success rate was incredibly impressive and suggests that lie detection is relatively straightforward, providing you use the right technique. As the researchers point out:
This research suggests that effective questioning is critical to deception detection.
Asking bad questions can actually make people worse than chance at lie detection, and you can make honest people appear guilty.
But, fairly minor changes in the questions can really improve accuracy, even in brief interviews.
This has huge implications for intelligence and law enforcement.
Difficulty often arrives in real life because we often like to presume innocence and honesty. The researchers suggest we’d be much better at detecting liars if we presume guilt and then question accordingly. All of which may be great at detecting lies, but perhaps not so good for our more honest communications.