ONE OF MY GUILTY PLEASURES is the long-running TV show NCIS, a drama focused on the Naval Criminal Investigative Service. The hero is a former marine, now Special Agent Jethro Gibbs, a disciplined detective with an uncanny ability to observe and question criminal suspects. Gibbs doesn’t say much or display a lot of emotion in the interrogation room—indeed, his cool demeanor is his trademark—yet he is a keen lie spotter.
Psychological scientists are fascinated by real-life versions of the fictional Gibbs. Detecting lies and liars is essential to effective policing and prosecution of criminals, but it is maddeningly difficult. Most of us can correctly spot barely more than half of all lies and truths through listening and observation—meaning we are wrong almost as often as we are right. And half a century of research has done little to polish this unimpressive track record.
But scientists are still working to improve on that, and among them is social psychologist Aldert Vrij of the University of Portsmouth in England. Vrij has been using a key insight from his field to improve interrogation methods: the human mind, despite its impressive abilities, has limited capacity for how much thinking it can handle at any one time. So piling on demands for additional, simultaneous thought—or cognitive “load”—compromises normal information processing. Because lying is more cognitively demanding than telling the truth, these compromised abilities should be revealed in detectable behavioral clues.
Why is lying more demanding? Imagine for a few minutes that you’re guilty of a murder, and Gibbs is cross-examining you. To start, you need to invent a story, and you also have to monitor that tale constantly so it is plausible and consistent with the known facts. That task takes a lot of mental effort that innocent truth tellers do not have to spend. You also need to actively remember the details of the story you’ve fabricated so that you don’t contradict yourself at any point. Remembering a fiction is much more demanding than remembering something that actually occurred. Because you’re worried about your credibility, you’re most likely trying to control your demeanor, and “looking honest” also saps mental energy. And you’re not just monitoring yourself; you’re also scanning Gibbs’s face for signs that he might be seeing through your lie. That’s not all. Like an actor, you have the mental demands of staying in character. And finally, you have to suppress the truth so that you don’t let some damning fact slip out—another drain on your mind’s limited supply of fuel. In short, the truth is automatic and effortless, and lying is the opposite of that. It is intentional, deliberate and exhausting.
Testing the Limits of Lying
So how could Gibbs and other detectives exploit the differing mental experiences of liars and truth tellers? Here are a few strategies that Vrij and his colleagues have been testing in the laboratory, which they describe in a recent issue of the journal Current Directions in Psychological Science.
One intriguing strategy is to demand that suspects tell their stories in reverse. Narrating backward increases cognitive load because it runs counter to the natural forward sequencing of events. Because liars already have depleted cognitive resources, they should find this unfamiliar mental exercise more taxing than truth tellers do—which should increase the likelihood that they will somehow betray themselves. And in fact, that is just what happens in the lab: Vrij ran an experiment in which half the liars and truth tellers were instructed to recall their stories in reverse order. When observers later looked at videotapes of the complete interviews, they correctly spotted only 42 percent of the lies people told when recounting their stories without fabrication—below average, which means they were hard to spot—but a remarkable 60 percent when the liars were compromised by the reverse storytelling.[break]
Another tactic for increasing liars’ cognitive burden is to insist that suspects maintain eye contact with their questioner. When people have to concentrate on telling their story accurately—which liars must, more than truth tellers—they typically look away to some motionless point, rather than directly at the conversation partner. Keeping eye contact is distracting, and it makes narration more difficult. Vrij also tested this strategy in the lab, and again observers spotted lies more easily when the liars were required to look the interrogator in the eye.
Drawing Out the Truth
A third strategy that could be surprisingly effective is to ask suspects to draw a picture. Putting pencil to paper forces people to give spatial information—something that most liars have not prepared for as part of planning their lies and that, therefore, overtaxes their mental resources. When Vrij and his colleagues asked volunteers what their offices looked like, after instructing half to tell the truth about their occupations and half to lie, both truth tellers and liars gave the same amount of detail in their verbal responses. But when Vrij asked them to draw their offices, the liars’ drawings were much less detailed than those of the truth tellers. In another of the experiments, volunteers were questioned about a lunch date that only some subjects had actually attended. The liars’ verbal descriptions of the restaurant did not match up as well with their drawings as did the truth tellers’—and the inconsistencies exposed the lies.
All these tricks may seem like overkill when we think about the fictional detectives we know, including NCIS agent Gibbs, who seem able to ferret out every fib they hear without using any strategies other than their intuition. But in real life, such people are exceedingly rare; psychological scientists call them “wizards” because of their seemingly supernatural lie-detection skills. Researchers have been trying—without a lot of success—to unravel these wizards’ strategies. Until they do, less sophisticated lie catchers may be able to exploit the mind’s cognitive deficits, using tricks such as Vrij’s, to catch the bad guys in their deceptions.