We may think we know the telltale signs of lying, be it shifty eyes or nervous fidgeting. Professional interrogators look for such tells, too, assuming a suspect’s nervousness betrays his guilt. But interrogation can rattle even the innocent, so nervousness alone cannot distinguish liars from truth-tellers.

Scientists looking for better ways to detect lies have found a promising one: increasing suspects’ “cognitive load.” For a host of reasons, their theory goes, lying is more mentally taxing than telling the truth. Performing an extra task while lying or telling the truth should therefore affect the liars more.

To test this idea, deception researchers led by psychologist Aldert Vrij of the University of Portsmouth in England asked one group to lie convincingly and another group to tell the truth about a staged theft scenario that only the truth-tellers had experienced. A second pair of groups had to do the same but with a crucial twist: both the liars and the truth-tellers had to maintain eye contact while telling their stories.

Later, as researchers watched videotapes of the suspects’ accounts, they tallied verbal signs of cognitive load (such as fewer spatial details in the suspects’ stories) and nonverbal ones (such as fewer eyeblinks). The eyeblinks are particularly interesting because whereas rapid blinking suggests nervousness, fewer blinks are a sign of cognitive load, Vrij explains—and contrary to what police are taught, liars tend to blink less. Although the effect was subtle, the instruction to maintain eye contact did magnify the differences between the truth-tellers and the liars.

So do these differences actually make it easier for others to distinguish liars from truth-tellers? They do—but although students watching the videos had an easier time spotting a liar in the eye-contact condition, their accuracy rates were still poor. Any group differences between liars and truth-tellers were dwarfed by differences between individual participants. (For example, some people blink far less than others whether or not they are lying—and some are simply better able to carry a higher cognitive load.)

All this makes it hard to put the study’s findings into practice—especially out in the field, where the people most likely to lie are those who are good at lying. “In the real world, there’s no Pinocchio-like cue that distinguishes liars from truth-tellers,” says study co-author Ronald Fisher of Florida International University. Magnifying subtle differences may be the next best thing. [For more on lie detection, see “Portrait of a Lie,” by Matthias Gamer; Scientific American Mind, February/March 2009.]

Note: This article was originally printed with the title, "The Load of Lying."