Catching lies

Psychologists in virtually every field have myths they have to fight. Personality psychologists have deal with the baseless but popular Myers-Briggs personality types^1,2. Developmental psychologists have to deal with the false claim that playing Mozart for babies improves their intelligence³. Like other deception researchers, I have to deal with people’s persistent and inaccurate beliefs about lies and how to catch them. I’d roughly estimate that – in my own limited experience – somewhere around 60% of the time, when I tell someone I study deception, they ask me if I can read people’s behavior and tell when someone is lying. My typical answer is, “No, I can’t do that. You can’t do that. No one can do that.”

Myths are often more appealing than reality. They’re frequently simpler, easier to understand, and more romantic than what science tells us⁴. People seem to enjoy the idea that lies can be caught by carefully observing a person’s body language or slight muscular movements in the face. It’s certainly a popular notion in fiction that we need to look at the right cues in order to accurately catch lies and deceit. The television show Lie to Me was built on the premise of catching lies through nonverbal behavior with astounding accuracy*. In Casino Royale, James Bond cleverly observes Le Chiffre’s eye twitch to catch him in a poker bluff**. In the musical Hamilton, Angelica Schulyer points out Alexander Hamilton’s fidgety hands as he deflects a question about his family. All this might make for entertaining fiction, but how do popular ideas about deception stand up to scientific scrutiny? Not well.

Researchers use the term “cue to deception” to refer to any observable behavior that might differ between someone lying and someone telling the truth. “Observable behavior” includes anything a human could detect without the assistance of technology. This includes verbal behavior (what a person says), nonverbal behavior (what a person does with their body), and paralinguistic behavior (the qualities of a person’s speech, such as pitch or speed). When we study cues to deception, we need to study situations in which we know for sure someone is lying or telling the truth. For example, we might instruct people to lie or tell the truth in a laboratory situation, or we might collect recordings of people whose lies (and truths) were later revealed (or backed up) with incontrovertible evidence, such as DNA testing***. Knowing who is lying and who is telling the truth is essential because if we don’t know for sure, we won’t be able to calculate how different lies and truths are. That is, we won’t be able to tell how well the behavior indicates deception. 

For example, if we think that liars blink more often than truth tellers (spoiler: they don’t), we will need recordings of a large number of people lying and a large number of people telling the truth. Then we need to watch all the recordings and count all the times each person blinked. We can then calculate how many times the typical liar blinks and how many times the typical truth-teller blinks and compare****. We can do this kind of counting and calculation for any kind of behavior we can see or hear.

Despite the widespread stereotype, there's no evidence that liars don't look people in the eye.

Thankfully, psychologists and other researchers have been studying cues to deception for nearly a century, so we have a huge amount of data about how people act when they lie and tell the truth – for more than 150 different behaviors⁵. The data tend to conflict with widespread beliefs about deceptive behavior. For instance, it’s a popular stereotype that “liars can’t look you in the eye.” This belief appears in a vast number of cultures across the world⁶. But there is, in fact, no evidence that liars don’t look people in the eye.

The accumulated evidence indicates that there indeed are some differences between truthful and deceptive behavior. For example, liars tend to be more tense, give less detailed descriptions, and tell less compelling stories. However, these differences are extremely small. How small? Researchers often quantify differences by putting them on a standardized scale, called an “effect size,” so we can compare them more easily. One way to do this is with a measure called Cohen’s d. The typical cue to deception differs between lies and truths with a Cohen’s d of 0.10⁵. This is about the same as the difference in height between 15- and 16-year-old girls⁷. It’s a real difference, but it’s practically imperceptible.

Imagine you had to guess whether a girl was age 15 or 16, based on her height and nothing else. Could you do it accurately? You couldn’t. No matter how skilled you are, there is simply not enough information there to make good guesses.  This is the same kind of challenge people are faced with when they try to detect deception.

In addition to studying how liars and truth tellers behave, for decades researchers have also been studying how well people can catch lies. With all the data we have, psychologists have calculated that, on average, people are about 54% accurate at distinguishing between truths and lies⁸. This is pretty unimpressive. If you were guessing using a coin flip, you’d be 50% accurate on average.

If 54% is the average, you might be wondering if some people are better at catching lies than others. Because the cues to deception are so weak, we wouldn’t expect there to be much variation in people’s skill at catching lies. And that’s exactly what we see when we calculate how much people vary in their abilities: There is virtually no variation in lie detection ability across people⁹. That is, everyone is equally bad at catching lies. However, there is a lot of variation in people’s skill at lying. Some people are quite skilled, and it is virtually impossible to distinguish between their lies and truths. Other people are more transparent, and it’s easier to tell when they are lying or telling the truth. You might be able to think of examples of times you’ve caught people lying – but you were probably able to catch them because they told a bad lie, not because you’re good at catching lies*****.

What about situations in which it’s really important that we make good judgments about whether someone is lying – when the stakes are high? Is it easier to tell if someone is lying in an interrogation room or on the witness stand? Probably not. In fact, when you compare accuracy rates in low-stakes and high-stakes scenarios, you see that the accuracy rates do not change (they’re still quite low), but both liars and truth-tellers are more likely to be judged to be lying⁸. That is, everyone seems to look more suspicious when the stakes are high. Psychologists call this “motivational impairment”¹⁰: high levels of motivation often make people perform worse than they would otherwise, like when a star athlete “chokes.” Liars and truth-tellers seem to be approximately equally impaired by high levels of motivation⁸.

Does this mean people lie about important things and get away with it a lot? Yes. Psychologists have conducted studies in which they ask people to report lies they’ve told. The participants are promised complete anonymity, to encourage them to be honest about their past deceit. In a study in which people were asked to describe serious lies they had told – including criminal behavior and marital infidelity – about 40% of the lies were never discovered¹¹. The vast majority of the lies people tell are small and harmless – intended to spare someone’s feelings or to keep relationships running smoothly. Some lies are serious and potentially harmful. But we are terrible at detecting the harmless ones and the serious ones alike.

The collected findings of the science of deception have serious implications. An important lesson to take from deception research is that we should be skeptical when we hear someone claiming they have a way to accurately catch lies. There’s no shortage of books, videos, and training packages that allege they can help you detect deception or read body language. Many (if not most) of these are not based on any scientific evidence. People accumulating fame and money from selling these products might not care^{12, 13}.

Another important lesson to take from research is that we must be highly cautious about the training we invest in and the policies we enact as they pertain to deception and deception judgments. In legal contexts, correctly judging whether someone is lying or telling the truth may have consequences for people’s lives and liberty: Mistaking a true denial of a crime for a lie can lead to a wrongful conviction; mistaking a false denial for an honest denial can allow a perpetrator to elude justice.

Transportation security is one domain in which the science of deception has been mishandled.

If we don’t use what we know about deception, it can also be highly wasteful. Governments and law enforcement agencies have not always made decisions consistent with the science of deception. For example, the Transportation Security Administration (TSA) spent nearly $900,000,000 on a program called Screening Passengers by Observational Techniques (SPOT) – a package of training and security measures largely based on the premise of visually examining airline passengers for behavioral cues to deception¹⁴.  The program was intended to help identify potential threats to transportation security. The techniques used in this program were inconsistent with the research on deception, and there was no scientific reason to think it would work. It was likely a colossal waste of resources.

Catching lies with unerring accuracy might be fun to watch in movies or on TV. But it’s wasteful, if not dangerous, to mistake myths and fantasy for reality. These fictional portrayals of deception detection are just that – fictional.

Timothy Luke wrote and Will Crozier edited this post

Notes

* For some time, Lie to Me claimed to be based on science, but this assertion was later removed from the promotional materials. Ironically, exposure to Lie to Me might actually make people worse at catching lies. See Levine, T. R., Serota, K. B., & Shulman, H. C. (2010). The impact of Lie to Me on viewers’ actual ability to detect deception. Communication Research, 37, 847-856.

** Of course, this ends up spectacularly backfiring when Le Chiffre learns that Bond has discovered his “tell.”

*** For example, some researchers have examined video recordings of people pleading to the media for the public to provide information about a relatively who had gone missing. In some of the cases, it was later proven that the pleader in fact killed their relative. In other cases, it was proven they were innocent.

**** Point of interest: One of my first research jobs was counting the number of times people blinked in interrogations. It was one of the most mind-numbing things I’ve done in my life.

***** By definition, any time you fail to catch someone lying to you, you don’t know it happened. For that reason, people often have an inflated sense of how good they are at catching lies: You can remember the times you caught lies successfully (which can give you the impression you’re good at it), but you simply have no knowledge of all the times you’ve been duped (and it could be a lot).

References

[1] Pittenger, D. J. (2005). Cautionary Comments Regarding the Myers-Briggs Type Indicator. Consulting Psychology Journal: Practice and Research, 57, 210-221.

[2] https://www.vox.com/2014/7/15/5881947/myers-briggs-personality-test-meaningless

[3] Pietschnig, J., Voracek, M., & Formann, A. K. (2010). Mozart effect–Shmozart effect: A meta-analysis. Intelligence, 38, 314-323.

[4] Lewandowsky, S., Ecker, U. K., Seifert, C. M., Schwarz, N., & Cook, J. (2012). Misinformation and its correction: Continued influence and successful debiasing. Psychological Science in the Public Interest, 13, 106-131.

[5] DePaulo, B. M., Lindsay, J. J., Malone, B. E., Muhlenbruck, L., Charlton, K., & Cooper, H. (2003). Cues to Deception. Psychological Bulletin, 129, 74-118.

[6] Global Deception Research Team. (2006). A world of lies. Journal of Cross-Cultural Psychology, 37, 60-74.

[7] https://www.cdc.gov/growthcharts/data/set2clinical/cj41c072.pdf

[8] Bond Jr, C. F., & DePaulo, B. M. (2006). Accuracy of deception judgments. Personality and Social Psychology Review, 10, 214-234.

[9] Bond Jr, C. F., & DePaulo, B. M. (2008). Individual Differences in Judging Deception: Accuracy and Bias. Psychological Bulletin, 134, 477-492.

[10] DePaulo, B. M., Kirkendol, S. E., Tang, J., & O'Brien, T. P. (1988). The motivational impairment effect in the communication of deception: Replications and extensions. Journal of Nonverbal Behavior, 12, 177-202.

[11] DePaulo, B. M., Ansfield, M. E., Kirkendol, S. E., & Boden, J. M. (2004). Serious lies. Basic and Applied Social Psychology, 26, 147-167.

[12] Eriksson, A., & Lacerda, F. (2007). Charlatanry in forensic speech science: A problem to be taken seriously. International Journal of Speech, Language and the Law, 14, 169-193.

[13] Masip, J., Herrero, C., Garrido, E., & Barba, A. (2011). Is the behaviour analysis interview just common sense? Applied Cognitive Psychology, 25, 593-604

[14] Government Accountability Office (2013). TSA Should Limit Future Funding for Behavior Detection Activities. GAO-14-159. https://www.gao.gov/assets/660/658923.pdf