Dec 1, 2006

Detecting Deception

One of the reasons I started this blog is out of frustration.

There happens to be a lot of misinformation on the web about how to detect deception. Don't get me wrong, there also happens to be a lot of useful information on the subject as well.

But as someone who has followed this issue for over a decade, it is somewhat troubling that the more questionable advice is often the easiest to find.

For example, one of the most popular sites offering advice about detecting deception contains many inaccuracies.

In fact, the first three tips for detecting deception by are just plain wrong.

  • decreased body movements
  • the lack of eye contact
  • increased facial touching, decreased touching of the chest/heart...
None of these statements are true.

Nonverbal cues of deception depend on the personality of the liar, the nature of the lie being told, and the context or setting in which the lie happens.

Some people gesture more when lying; some people gesture less. Eye contact does not necessarily change when lying or telling the truth. And increased facial touching also depends on the context.

There is NOT a consistent set of clues that can be applied across any given situation. It simply doesn't work that way.

A much better rule of thumb is to look for an inconsistency in a person's behavior.

Does a person's nonverbal behavior seem inconsistent with his or her normal behavior or with what is being said?

But, even this advice is probably of little use to most people (explained below).

One more example of the type of bad advice that's being offer - again, on another page at This time the advice deals specifically with eye movement and lying.
  • Does a person look up to the left?
  • Down to the right?
  • and so on....
Again, this advice is based on myth, not scientific evidence.

In fact, one of the primary researchers, who is involved with eye-movement research (NLP, as it is called), actually has the following advice to offer when asked questions about eye-movement, lying and detecting deception:
"But so far as I know, there's no absolute way to know if someone's lying. Human beings are far too complex for our current understanding to give definite results."
And a real human lie detector, has the following to say about watching eye contact:
"Please, please, please - if you are judging someone by their eyes alone - Stop!!"
So, what sites offer better advice and information on how to detect deception?
None of these sites, however, offers a simple technique for detecting deception. For the most part, these sites acknowledge just how difficult it can be to catch a liar.

Finally, there is little evidence to suggest that people can be taught to detect deception in any meaningful way.

It appears that detecting deception is a skill that only a few people have. A skill held by a very small set of human lie detectors.


Anonymous said...

You might want to read Aldert Vrij's work. His extensive studies indicate that "decreased body movements" are linked to deception, caused by cognitive complexity.

Tad said...

Yeah, I believe the claims are consistent with Vrij's work. Decreased body movement when lying has also been linked to individual differences and in some research the consequence or nature of the lie being told. While there might be a slight pattern (this will always be debatable given the inconsistent findings obtained across studies), there is simply too much variation within the pattern for it to be useful. In my opinion, it's very difficult to generalize or offer specific advice on how to detect deception based on any specific behavior. I think the accuracy rate for detecting deception based on nonverbal cues will always be very low. The cues vary too much and observers' perceptions are always biased. This is a very difficult combination to overcome.

Chris said...

I have good news and bad news for lie detectors out there. I'll start with the bad. The claim by Ekman and colleagues that there are certain individuals who can detect deception (e.g. "Who can detect a liar?", truth wizards) has been undermined (check out Bond, 2008 for a very brief and rather humorous account).

The good news? Our lab has demonstrated a key nonverbal and readily observable pattern of behaviour associated with deception (we're in the process of submission).

Chris Street