To catch a liar, ask unexpected questions

NRC International, 18 January 2010 18:09, By Ellen de Bruin

A Dutch psychologist has developed a method for catching liars that is successful 70 to 80 percent of the time.

“Are you carrying any hazardous materials?” The question is commonly asked at airports – but seldom answered truthfully by those who actually are, like the Nigerian man who tried to blow up an airplane on Christmas Day. A foolproof lie-detection method would be a godsend to airport security.

According to psychologist Aldert Vrij, professor at the University of Portsmouth, it will be a long time before we develop one. Vrij has written countless articles on the subject of lie detection. He also advises the British police and military on the matter. “But I am not at liberty to discuss that,” he said.

Catching liars in the act

Investigations show that people currently perform little better than chance when trying to determine whether or not someone is telling the truth – and experts do little better. But we are getting there. Vrij has developed techniques that enable him to determine correctly whether or not someone is lying 70 to 80 percent of the time.

In one study he asked pairs of test subjects to steal money. They were then asked to deny it to an accomplice of his. Vrij also provided the couples with an alibi: they had lunch together. A control couple really did have lunch together. Later, the pairs were interviewed two by two.

“People cannot prepare for every contingency,” Vrij said in an interview. “So you need to ask them a couple of questions they will expect: what did you guys eat, what did you talk about? Then you continue asking questions that are logical in that context but unexpected. Questions that cannot easily be answered with ‘I don’t know.’ Questions about spatial orientation are an example: where were you situated relative to the door? On which side of your table did the waiter stand? Liars will start to talk more slowly. They will fall silent more often, indulge in less detail and contradict themselves more often.”

Which means they get nervous?

“No, their brain just becomes overloaded. Liars are not always more nervous. A lot of people think that, but someone telling the truth can also be nervous, especially when being interrogated by the police. Liars have to think more, especially when you ask them tough questions. So what it boils down to is: can you ask questions to magnify the difference between the truth-tellers and liars? As it turns out, sometimes you can. Another effective technique is asking people to draw things, like a map of the restaurant. Or have them recount events in the opposite order, another thing more difficult for liars.”

Another problem facing would be lie-detectors is that almost all research conducted focuses on acts that have already been committed. “After 9/11 however, it has become apparent that we need to be able to catch people lying about acts they haven’t committed yet,” Vrij said. That is what he and his colleagues are currently studying.

They had test subjects deliver a package to a designated building. The test subjects – all soldiers and police officers – were later approached by ‘secret agents’. If the agent failed to reproduce a certain password, the test subjects were supposed to lie about their mission. Otherwise, they were supposed to tell the truth. The agents struck at unexpected moments, both before and after the subjects had delivered their packages.

This not only enabled they scientists to differentiate between those telling the truth and those not, it also helped them determine if lying about acts one planned to carry out differed structurally from lying about a prior act.

The scientists studied the issue in a more everyday context as well. They asked some people at the airport to lie about a journey they planned to make and others to tell the truth. They then checked if the stories told by the liars differed structurally from those who told the truth.

The results, will soon be published in Applied Cognitive Psychology and Psychology, Crime and Law. It turns out people describe their plans for the future in just as much detail when lying as they do when telling the truth. When people lie about the past, however, their accounts are usually less detailed than truthful ones, because they fear getting caught out on the details. True stories about future plans differ from lies on one point: they are more plausible. This fact alone makes it easier to catch someone lying about intentions than it is to catch someone lying about the past. Also, it is easier to detect lies using a write-up of a conversation than it is to do so during a live conversation.

The latter conclusion poses obvious problems in practice.

“Indeed,” Vrij said. “The challenge therefore lies in seeking out verbal traits that distinguish those telling the truth from the liars. Once you figure that out you can train people to pay attention to those specific traits. I am sure it can be done.”

But if plausibility is the key characteristic in catching liars, won’t terrorists be able to avoid detection by preparing their stories better?

“There is another reason why plausibility is not an ideal variable: it is highly subjective. We are looking for objective characteristics to define plausibility. Indeed, terrorists will prepare their stories well.” However, he added, the technique of using expected and unexpected questions should help.

Won’t this set off an arms race for interrogators? Will suspects start preparing drawings in advance because of your research?

“That risk is always there. But my colleagues and I have written quite a lot about cheating on polygraph tests, and I do not have the impression a lot of suspects read those articles. The polygraph works by measuring arousal – heart rate, blood pressure, sweating. All based on the assumption that liars will tend to be more nervous. A naive, unreliable method that yields many false positives. At airports it is useless, since nearly everybody there is innocent.”

“The evidence that these tests work at all is very doubtful. In the US they are popular with the police and private military companies. They are more focussed on obtaining confessions there to begin with. A tool called the Behavioural Analysis Interview or BAI is also popular there. More than half a million people have been trained in applying it, asking a series of standardised questions while monitoring for indicators of anxiety: fidgeting, moving around, and looking away. But take the Nigerian who tried to bomb that plane at Christmas for instance: he made a very calm impression on his fellow passengers. That is consistent with research findings, but not what his fellow passengers would expect.

“The BAI can lead to false confessions. Someone who has committed a crime will realise: if I fidget too much I will look suspicious. Somebody who has a clear conscience might naively think he won’t get into trouble.”

Will we ever develop a sure-fire lie detection method?

“No. Even brain scans won’t do the trick. A ‘lie centre’ does not exist, we know that already. Some liars will always get off scot free.”

Aldert Vrij was awarded his doctoral degree in 1991 at the Vrije Universiteit in Amsterdam. He left for Portsmouth in 1994, where he became a professor in 2000.

Photo Maurice Boyer