Many of us watched the hit show Lie to Me on Fox every Monday evening. Thanks to the popularity of the show, people developed an interest in the topic of microexpressions and the world of nonverbal behavior.
Recently researchers at Michigan State University led by professor of communication, Timothy Levine, are putting Lie to Me viewer’s deception skills to the test in a new study entitled “The impact of Lie to Me on viewers’ actual ability to detect deception”.
The study which was published in the Journal of Communication Research, finds watching Lie to Me “increases suspicion of others but that is reduces one’s ability to detect deception”, according to an article written by Tom Jacobs of Miller-McCune.
Levine and his colleagues experiment involved 108 undergraduates at the university. Thirty-three of these individuals watched an episode of Lie to Me, another thirty-five watched a different crime drama called Numb3rs while the last third of the group did not watch either program. The group that did not watch either show served as the control group.
After they watched various episodes (or none at all), the participants saw a series of 12 taped interviews where half were telling the truth and the other half lied consistently. The participants were instructed to state whether they believed the person in the interview was being honest or deceptive.
Interestingly, according to the article written by Jacobs, the control group was the most accurate, correctly identifying the person as honest or dishonest 65.2 percent of the time. The Numb3rs group came in second, at 61.7 percent, while the Lie to Me group came in last at 59.5 percent.
The results of this study illustrated that Lie to Me viewers were “no better at distinguishing truths from lies but were more likely than control participants to misidentify honest interviewees as deceptive. Watching Lie to Me decreases truth bias thereby increasing suspicion of others while at the same time reducing deception detection ability” according to the study’s abstract.
It seems a larger sample size may be necessary in the future and we would be interested in reading the complete study. However, the show suggests what we have been suggesting all along: that viewers of Lie to Me shouldn’t accept all information that is presented on the show as accurate or think they know more about lie detection without getting formal training.
Hum, I find normal that viewers of Lie to me would be more suspicious, but what the article don’t say is if any of the groups where able to establish a base line on the group of people supposed to “lie consistenly”. Because if they are, in fact lying consistently, there is no scientific way to say if they are (as well illustrated by the episode Blinded). The base line is core of this science, so all this study shows at this point is that Lie to Me viewers are more suspicious, which seems natural as the motto of the show is “everybody lies”… 🙂
You bring up some good points. We haven’t read the entire article, but have requested to see it from Dr. Levine. We also think that the sample size is quite small, given that there were only a little over 100 participants. It would be interesting to see what the results would be if the sample size were larger. Remember that the average rate of detecting deception is only about 53%, which means all of the groups did quite well compared to the general public. But yes, it would be good to get some additional information regarding this study and what the methodology was.
According to the article, the reason viewers (who had just seen the pilot episode of “Lie to Me”) did worse was that the increased suspicion decreased the “truth bias” without improving detection skills. Thus the “Lie to me” group performed no different than the others in judging the deceptive interviews as lies, but did slightly worse in judging the honest interviews as honest. This was a novel finding, as ordinarily stimuli that increase suspicion tend to improve lie detection slightly.
The article also takes some potshots at some of the premises of the show (i.e. that real-time deception detection is feasible with Lightman’s accuracy rate).
I’d also suggest that watching ONE episode of Lie to Me would only encourage one to believe they’d learned some kind of “trick” to detecting deception that they could use at will. After reading Ekman’s books, it became abundantly clear to me that this science is not anywhere near as simple as portrayed in the show. Especially not one episode of the show. And certainly not if it were a recent episode rather than one of the first season episodes where more of the science was explained.
But I find this area of study fascinating and I think it’s great that someone took an interest in what the effects of the show are on people. This appears to jive with what I’ve heard from other laymen in the field on various websites like the readers of Eyes for Lies’ website.
Deception detection is extremely nuanced in my limited experience. I’ve not seen that made very clear in any form of entertainment media so far.
People who want a copy of my paper can drop me an e-mail.
https://www.msu.edu/~levinet/
The research out of my lab suggests that nonverbal behaviors are more highly misleading, and higher accuracy is obtained through content in context or with strategic questioning.
I am very pleased to see TV episodes of “Lie to me” increase in popularity as this is a science that was pretty much unheard of 30 years ago. More and more serious analysis of this research helps us to better understand the art of detecting deception. Detecting deception is not easy, nor the abilities to recognize truth. The majority of those who have an immediate need to be well versed in detecting deception do not take the time or the effort to train themselves on how to recognize the indicators of deception when it occurs. More often, it’s a guess, flip of the coin accuracy, a guess at best, with no serious articulation as to why.
“Lie to me” demonstartes it to you, as it occurs, in the content where it can be read and understood. However, novices who watch the show are probably void of the baselines needed to comprehend where and when deception occurs in reality. Its so much easier to read when your in your easy chair, in front of the TV.
I totally believe that people who watch Lie To Me become less able to detect deception for the following reason: Lie To Me uses actors, not real liars, and there is no way to compare acting to real lying. Actors may be very good actors, but they are thinking about what they need to express about how they feel. Too much cortical involvement to really show microexpressions. It just can’t be done. That’s why I didn’t shell out 60$ and buy the program on training to detect microexpressions. It used actors, not real liars.
If the tell tale signs of lying involve involuntary actions on the side of the liar, then you simply can’t use actors to learn about detecting lies. They are acting, not lying.
Dr. Fay,
You’re correct about micro-expressions in the show in general, although there have been a few times when the actors have done remarkable jobs of nailing one. But in terms of the training tools offered here and probably elsewhere, that isn’t really an issue because of the way they function.
I haven’t used any of Humintell’s training programs yet, though I plan to in the future. But if their micro-expression training is similar to the METT training that Paul Ekman created, then it’s not using actors to “perform” micro-expressions in real-time. What it (Ekman’s METT training anyway) uses are a photo of a model (or possibly grad student) showing a neutral face, and a photo that same model showing a particular expression, and flipping rapidly back and forth between the two.
Given that micro-expressions are so fast, at about 1/25 of a second, they only last as long as about 3 (or 4 on higher end displays) frames of video refresh on a typical computer monitor, and only 1 frame of actual recorded video with either film (24 frames per second) or traditional video camera video (30 frames per second). So doing the training this way really doesn’t incur any actual loss in terms of facial-movement-over-time within the training. This also allows the training developers to select only those photos in which the expressions are being properly displayed.
It’s also worth noting that the training is only designed to improve one’s accuracy in spotting and identifying which micro-expression is being shown, and it’s still up to the person using the training to carry what they’ve learned into the real world with real-life face-to-face interactions.
The training is worthwhile despite it’s using models to pose the facial expressions (I’ve only tried Ekman’s METT Original thus far).
Also bear in mind that micro-expressions themselves don’t indicate deception or honesty, they only indicate that a felt emotion is being consciously hidden. It’s still up to the observer to identify whether that masked emotion might indicate deception or is simply an innocent masked emotion in that particular instance. They’re an important tool, but they’re not a Pinocchio’s nose, only an additional detail to make note of.
I hope that helps! 🙂
Hey Keith, On a side note, our MiX training is pretty much the exact same as the METT, since METT was co-created by Dr. Ekman and Dr. Matsumoto. Our other training tools, MiX 2, SubX, etc. unique and aren’t found anywhere else. However, the training tools we offer are similar to the METT in that they show a neutral face, then flash to a target face.
well the hardest thing in spotting deception , is how we interpret the things we see. and by just watching the show every Monday doesn’t prove that the person will get better at spotting deception, this needs some studies, researches, and practice, so from that small sample, people can misinterpret things they saw. let’s not start judging everything.
I would be very curious to know how many attorneys and Judges watch this show and attempt to use it daily in courtrooms around the country to substitute for facts and evidence. Every day we officers of the court using hunches and opinions in place of facts and evidence causing historically high rates of appeals (only for those who can afford it, leaving the poor left out), causing legal traffic jams of the judicial systems. Plea deals are made at a rate of over 75% of all cases just because people are sitting in jail for an average of 3-9 months waiting for a trial, even for shoplifting charges. There should be a line of inquiry at the beginning of every hearing to ascertain who, if anyone has watched these types of shows and do they understand they can’t apply the fictional “science” in our courtrooms.
There was a case in Texas where a psychotic mother killed 5 children and there was something that came into the trial about an episode of law & order and they accused her of watching it and trying to get away with murder. The entire case was reversed on appeal because of what the prosecutor said about the tv show.
Don’t get me wrong, I love the show and I also believe there is in fact a great deal to be learned about nuero linguistics, micro expressions, body language and deception. It has long ago been proven that lie detectors are not reliable and we are getting better every day with technology and understanding how the brain functions. One thing that stood out to me in the show was- Botox- nobody ever asked about Botox. Everyone gets them now and it isn’t always in medical records. They get them at girl parties, luncheons, spas, you name it! You can even order it online now. Talk about a curve ball….
Just my .02 that means little to nothing