Skip to main content

There's No Foolproof Way to Detect a Lie

Except for the rare band of "truth wizards" among us, people are not skilled at catching liars.
This machine is not reliable. (Photo: mnfiraq/Flickr)

This machine is not reliable. (Photo: mnfiraq/Flickr)

The first long con I fell victim to involved a golden retriever named Daisy. I was in the first grade, and so was Daisy’s co-owner, a friend I’ll call Kim. One day that spring, Kim walked up to the long conference-style lunch table we shared in our Catholic school cafeteria, dropped her hot lunch tray on the table, and, once we’d all stopped talking to look up at her, told us she had big news.

“My family got a puppy,” she said, sitting down now that she had our attention. “Her name is Daisy, she’s a golden retriever, and we rescued her from the alley.” Everyone was excited, or pretended to be. Kim was popular enough that you had to pretend to care about everything she said, even if it wasn’t that big a deal. Plenty of us had pets. I had a very good dog named Hannah at home, but you didn’t see me making public announcements about it. But I was not as cool as Kim.

Soon, Daisy was a lunchtime fixture. Recess, too. Kim would tell us about the fancy new toys her parents had bought the dog, or implausible sounding tricks she’d learned. (I seem to remember something about her being able to bark the alphabet.) In spite of ourselves, we grew curious. “When can we meet her?” we asked. “Not yet,” said Kim. “My parents say she’s too young for visitors.”

I was catfished by a dog, and I never saw it coming.

I bragged to my mom about Daisy—hopefully out of Hannah’s earshot. “That’s interesting,” my mom said. “Mrs. Miller never told me they got a dog.”

We kept asking Kim when we could come over and play with the puppy. She kept saying “Not yet.” Then, tragedy struck: Daisy got sick. Too sick for visitors. She had to stay overnight at the vet’s, Kim told us, whimpering. “What’s wrong with her?” we asked. The vet didn’t even know. That’s how bad it was.

A week later, Daisy was dead, and Kim was inconsolable. She cried, tearlessly, every lunch. We gave her our chocolate milk boxes and homemade desserts. At recess she was too sad to play kickball. She sat on a swing and we gathered around her feet like disciples.

It wasn’t long after that my mom and Kim’s mom ran into each other in town. My mom told Kim’s mom she was sorry to hear about Daisy. “Who’s Daisy?” Kim’s mom asked.

The reason I remember this story is that my diary from that year has an entry that reads, in its entirety: “DAISY WAS A LIE.”

It’s easy to see the telltale signs now, of course. But then, I had no idea. I was catfished by a dog, and I never saw it coming.

MOST PEOPLE AREN’T VERY good at knowing when they’re being lied to. In experimental settings (Ekman and O’Sullivan, 1991; Malone and DePaulo, 2001), when people are asked to assess whether a number of individuals are telling the truth or lying, they’ll get it right about half the time—typically, just barely better than chance. There are speech patterns and facial expressions said to be “cues,” but these are often unreliable, and can be overcome, particularly when the liar in question doesn’t care if you believe her or not. (Wanting to be believed stresses us out, which can lead to giveaways like averted eye contact and stammering.)

Even the polygraph, which measures several physiological indicators for possible deception—including blood pressure, pulse, breathing, and sweat production—isn’t infallible. In fact, many scientists categorize the machine as pseudoscience; some say that the inherent nature of deception means there can be no single reliable way to detect it. We can test for the presence of certain physiological reactions, but that’s all. Polygraph “validation” rates are determined by confession or jury verdict, both of which are subjective. The former could still be a lie, and the latter could be wrong.

Still, the polygraph has its proponents—namely, the CIA and FBI. According to a 2006 article in the Washington Post, experts argue the polygraph's greatest utility lies not in actual efficacy, but in the fact that people taking it believe it works. Again, though, that the tool makes people more likely to admit to things does not necessarily make those things true.

As New York University professor and former president of the American Society of Criminology Jerome Skolnick once wrote: “There actually is no independent means of checking the phenomenon of lying, of confirming that an individual designated as a liar actually lied.”

There do, however, seem to be certain people who are better at guessing. In the late 1980s, researchers Maureen O’Sullivan and Paul Ekman began a study that aimed to examine individual differences in people’s ability to detect deception. The study would span over 20 years and analyze 20,000 participants. Of those 20,000, O’Sullivan and Ekman eventually found 50 who had an above-average ability to detect lies. These people were given the very cute nickname “truth wizards.”

Truth wizards, according to O’Sullivan and Ekman, were able to accurately detect deception at least 80 percent of the time—a substantial improvement over the average 50 percent. The authors wrote that these were people with a knack for spotting micro-expressions, body language, and emotional cues. Most were introverted and observant. Truth wizards came from a variety of professional backgrounds, but (perhaps unsurprisingly) Secret Service members were among the most skilled.

The Wizards Project, as it has come to be known, is not without criticism; Charles Bond and Ahmet Uysal argued that the wizards’ performance was nothing more than chance. As academics are wont to do, O’Sullivan and Ekman have responded to Bond and Uysal’s criticism with criticism of their criticism. Regardless of the exact proportion of people who can skillfully detect deception, both pairs of authors agree that the number is very, very low.

Researchers are still trying to figure out how the truth wizards got to be so good, and if it’s possible that the rest of us can learn to be better. Ekman offers a variety of training tools for purchase on his website, though testing one’s potential improvement presents another challenge. All we know for sure is that spotting a lie is hard. I am reassuring myself here as much as anyone else, but if you have no idea when someone’s inventing a dog, you aren’t gullible. You’re normal.