News
No sure way to spot liars

Liars. The world is full of them. I run into them in my work all the time. I see students who want to convince me that they have ADHD so they can get drugs – or that they have a learning disability so they should get extra time to write exams.

Offenders deny their crimes and “misreport” stuff going on in institutions; police candidates apparently walk on water and have never made a mistake in their lives; people in minor motor vehicle collisions who apparently sustained horrible brain damage and need millions of dollars to compensate; athletes who were clunked on the head, don’t make sense and can’t think straight but insist they are “fine.”

Of course with my many years of training and keen powers of observation, I can spot a fraud a mile away and can always identify the liars.

Not.

It would be easier if liars were all like the proverbial used car salesperson in the plaid polyester suit and ducktail haircut, rubbing their hands with glee as they tell you that car you’re interested in was owned by a little old lady who only drove it to church on Sunday. The fact is that we have not yet come up with any really good ways to spot liars.

Think you’re pretty good at spotting liars – better than most people, anyway? HAH! You’re deluding yourself. You stink at it; we all do. This is a pretty well researched area and the fact is police, psychologists, FBI people, university undergrads, people on the street and pretty well anyone else you can think of have about a 50-50 chance of correctly separating out liars from truth-tellers. In other words, you can flip a coin and do just as well.

Still thinking, “well, I am better at this than most people? You are probably wrong. (You might also want to refer to some of my previous columns which have talked about how bad people are at assessing their own abilities. We stink at that too.)

If you’re in the US Secret Service, you MIGHT be a little better at it – and if you’re among the 0.5 per cent of people known as “wizards,” you might actually be pretty good; about 80 per cent accurate.

There are a very small group of people who seem to have been born able to detect liars. The TV show was based on one of these people.

A couple of researchers in California, Paul Ekman and Maureen O’Sullivan, have devoted a huge amount of time and energy trying to figure out the magic ingredient that makes people able to detect liars – and how to teach this skill to other people. Their eventual conclusions were:

1. We have no idea and B: we can’t.

Too bad, eh? Much as the idea of there being “truth wizards” is kind of appealing, there’s a lot of controversy about whether it’s accurate. Are people really just born this way – or is there some combination of inherent traits, along with years of slogging and studying, that leads to good results?

There is a little evidence that ability to pick out a liar does improve with age – that would argue for the “learned” part – but there are also all kinds of people offering all kinds of training, and little evidence that it makes a significant difference to daily performance.

Aside from the seat of the pants method, practiced by deluded people who are rather full of themselves and mistakenly think they are good at this, we have tried EEGs, voice stress analysis, MRIs, examining microexpressions, analysis of body language and nonverbal cues, cognitive load theories… on and on.

The Powers That Be in ancient China had suspected culprits chew on rice powder, then spit it out. The theory was that liars are more stressed, which decreases salivation, so if the spat-out rice powder was dry, you were a liar. While the mechanics of this approach are a little dated, the theory hasn’t changed a lot over the years.

Each approach contributed to overall knowledge in the field and each has some elements of validity – but the take home message is that none is good enough to rely on in the day to day, case by case context.

As a result, you will note that if you Google words related to lie detection or do an academic lit search, most of the hits will be from the early 2000s. There is not a lot of really recent stuff and that might be because the field has gone in a bit of a different direction (although there are still lots of ads from people trying to sell you courses on their chosen approach).

To a large extent, we’ve come to accept that we are pretty bad at this by nature, so we have to rely instead on specific tools and techniques that might assist us. In psychology and related fields, we have formal tests and measures that assess the extent to which someone is stringing us a line (and yes, these are used in police pre-employment screenings).

What about the policing field? Well, if you believe authors Aldert Vrij, Samantha Mann and Sharon Leal at the University of Portsmouth (and I tend to believe them), it’s best to stop worrying about gaze aversion (and dry rice powder) and use specific interviewing techniques. Their review in the October 2013 issue of the talks about several specific interviewing techniques that tend to widen the gap between liars and truth tellers.

Ask a suspect to tell you the details of an incident in reverse order – back to front – and observers were much more likely to be able to pick up cues to deceit and were therefore much more accurate – like about 60 per cent compared to only 42 per cent in the control condition. Ditto for asking suspects to maintain eye contact while telling their stories. Both approaches rely on cognitive load theory, which essentially says that lying is a lot of work – more work than telling the truth. Increase cognitive load by making a person do something more difficult and it gets harder to lie. (Mind you, 60 per cent ain’t so great either – but it IS better!)

There is also the “unanticipated questions” approach. Liars have to think up a story and answers ahead of time. If you ask them a question they didn’t anticipate, they may falter. The aforementioned authors describe a study in which one pair of subjects ate lunch in a restaurant (the Truth-Tellers) and another pair (the Liars) were off stealing a purse but pretended they also had lunch at the restaurant.

Both pairs could answer questions about what they had for lunch, who finished eating first, who was sitting near them – the usual stuff – but when asked to draw a sketch of the restaurant layout, a question the liars had not anticipated, it became obvious who was lying. The pairs of Truth Tellers came up with similar sketches and the Liars did not.

These authors also talk about strategies to use when you are more interested in examining concepts and ideas rather than events – handy for interviewing potential terrorists. It is called the Devil’s Advocate approach and I am not going go into it because I’m running out of space. There’s also the Strategic Use of Evidence (SUE) approach, where you get people to deny stuff you already know because you don’t tell them that you know it.

Bottom line? I suspect many of you know a whole lot more about most of this than I do, but it seems to me that if you’re still looking for minimal twitches of nose hairs or relying on other physiological signs of deception, you may well be missing the boat. Maybe one day we will get there… but not today.