Lie detectors: body language tells us surprisingly little about whether someone is being honest

Do you ever wonder if you could pass a lie detection test or imagine what it would be like to read people’s body language? Reading body language may be great for adding tension to action movie interrogation scenes, however, the truth is, there isn’t much evidence you can detect lies by watching someone’s body language.

When you try to discover whether someone is lying in an interview, your sources are the behaviour the person displays or the information they provide. Nonverbal lie detection (body language) is more popular than verbal lie detection as people think that lie tellers can control their speech but not their behaviour. But verbal cues for deceit are far more telling.

People often assume lie tellers will be anxious. For example, that a lie teller might look away from the interviewer, fidget with their hands, sweat or swallow frequently. There is no scientific evidence for this belief. The problem is truth tellers also get nervous during interviews and may display the same behaviour as lie tellers.

Lie tellers are more concerned about their credibility, whereas truth tellers are more likely to think that the truth will shine through. However, if lie tellers and truthful people opt for body language strategies, they will do the same thing: avoid displaying signs of nervousness.

But the spoken tactics truth tellers and lie tellers use differ. Truth tellers are forthcoming and willing to provide information. They typically do not provide all the information they know at first, because they don’t know how much they are expected to offer. They may also lack the motivation to provide a lot of information. Truth tellers think their honesty is obvious to observers. Why put so much effort into providing details they think are irrelevant when the truth is clear? Plus, at first, they may be unable to retrieve everything that is stored in their memory.

Talking the talk

Lie tellers try to keep their stories simple. They are afraid what they say may give leads to investigators that they can check. They fear they won’t be able to repeat all they said when interviewed again later, or that an elaborate lie will require too much thinking time.

Studies analysing deception research have shown that not only are verbal cues more revealing than nonverbal cues about deceit but also people are better at lie detection when they listen to speech than when they observe behaviour.

Interview protocols in most professions, such as border control and police, have been developed by deception researchers aimed at exploiting the different verbal strategies truth tellers and lie tellers use in interviews. The protocol interviewers choose normally depends on the evidence.

If the interviewer has independent evidence (for example, an email showing that someone attended an event) the strategic use of evidence (SUE) is the best choice. This is when interviewers ask questions about the event without revealing the evidence they have. Truth tellers who have nothing to hide will speak freely and provide details, whereas lie tellers will deny they attended the event, will be reluctant to give specifics and may deflect questions. Lie tellers are more likely than truth tellers to contradict the evidence.

The professional approach

Sometimes interviewers do not have evidence, but it is possible the interviewee can provide it. When using a verifiability approach (VA) interview technique, interviewers ask interviewees whether they can provide evidence the interviewer can check. VA research has found truth tellers are more likely to volunteer such evidence (for example, mentioning other people who were at the event) than lie tellers.

Suppose that the topic of investigation is not whether the interviewee attended an event but whether the interviewee tells the truth or not about what they discussed with someone at the event. SUE and VA are not appropriate for this situation. An email showing someone attended the event will not reveal what happened there. If the interviewee did not record the conversation, the interviewee won’t be able to offer verifiable information. In that situation, cognitive credibility assessment (CCA) can be used, an interview protocol that only considers the quality of a statement.

In a CCA interview, an interviewee is initially asked to report what happened during a narrow time period. The interviewee is then given prompts that raise expectations about what to say (let them listen to an example recording of someone giving the amount of detail you would like to hear), increases motivation to talk (by giving the impression that you listen to the best story you have heard in your life) or facilitates memory recall (by asking people to sketch out details of what they experienced while reporting their experiences).

In a CCA interview, interviewees are asked to tell their story several times over. CCA research has shown that truth tellers volunteer more extra information during these successive recalls than lie tellers who keep their stories simple.

It’s impossible to tell what information is inside someone’s head. For now, people’s thoughts are private as we simply don’t have the technology to unravel what someone is thinking. It may be less glamorous than a lie detector machine, but simply listening to the words someone says can reveal more about the state of their mind than they’d like.

Aldert Vrij receives funding from the Economic and Social Research Council (ESRC, present) and the Federal Bureau of Investigation (FBI, present).

He also received many grants in the past including from:
Engineering and Physical Sciences Research Council;
Dutch Government;
Singapore Government;
Leverhulme Trust