About the SA Blog Network

Guest Blog

Guest Blog

Commentary invited by editors of Scientific American
Guest Blog HomeAboutContact

To Spot a Liar: An Offices of SciAm Mystery

The views expressed are those of the author and are not necessarily those of Scientific American.

Email   PrintPrint

I daresay, do you really think you could spot a liar?

I daresay, do you really think you could spot a liar?

Congratulations. You have been hired for your first detective gig. Report immediately to the downtown Manhattan office of Scientific American magazine. There’s been a robbery of their archival issues. They think it was an inside job. You will be interviewing members of the staff on their recent activities and whereabouts. Upon arrival, report to the Editor in Chief. She will fill you in on what has happened.

You hop in a cab and head downtown. On the train you realize, you were so excited about your first assignment, you haven’t brought any equipment with you other than a notepad, pen and your smartphone. How will you be able to catch the thief? How will you be able to tell who is lying and who is telling the truth?

You will have to rely on your quick wit and keen eye to spot the thief. The interviews may be your only chance to solve the case quickly and impress everyone on your first job. Hopefully speaking with the suspects will give you a clue.

You immediately google ‘eyes and lie detection’ while in the cab. Your search yields 1,240,000 results in 0.31 seconds. Wonderful. How will you sort through the information in time? Thank goodness for New York City traffic. You begin reading one of the first theories that popped up. It is on Neuro Linguistic Programming and eye movements and has been around since the 70’s. It states that if a right handed person looks up and to their right while answering questions, they are lying (recalling imagined or fabricated stories). While if they look up and to their left, they are telling the truth (recalling actual events that they saw happen). Great, you will use this to sift through those who are telling the truth from those who are trying to hide something.

Scene of the crime: the vault at Sci Am

Scene of the crime: the vault at Sci Am

As you arrive at SciAm, you are greeted by the Editor in Chief. She doesn’t add much to the story. The archives are gone. They were stored in a vault in the storage room. The vault showed no damage and there was no forced entry. That is why she is suspecting an inside job. She hands you a list of staff members to interview and suggests you get started right away. She has already placed the first person on the list in the conference room for questioning.

You open the conference room door, introduce yourself and take a seat next to him. He’s an editor and astrophysicist. The Editor in Chief says he had worked at the SciAm offices late one night covering a curious story about Mars. Hmmm. During your questioning he seems calm though he does keep looking up. He says he left the SciAm office that night around 1 am after covering the Mars Curiosity story and that he rode his bike back to his apartment. Peculiar. But still, he seems like a nice, straight-shooting, honest guy. Besides, how could he make off with several vintage issues of SciAm on his bicycle?

This NLP theory of looking up and lying begins to look a little far fetched so you search around for more on the topic and come across an experiment that just came out of the United Kingdom last month. The study debunks the NLP theory of eye movement and lying. You decide to read through their experiment. And ask the Editor in Chief to delay the next interviewee on the list a few minutes.

To test the NLP theory of eye movements and lying, Richard Wiseman et al. took a group of 32 college students who were all right handed. They took each student individually and instructed them to take the researcher’s cell phone, go into a particular office, put the phone into their pocket or bag and then come back. Upon returning, they were greeted by a different experimenter, who asked them what they just did inside the office. The second experimenter would not know whether each participant was taking their turn at the lie or truth portion of the study. Under the lie portion, participants were told by the first experimenter to fib to the second experimenter and say they went inside the office and placed the cell phone into a drawer. When asked what the contents were inside the drawer, participants had been told to lie convincingly and make up what might be in an everyday office desk drawer. They were also asked about the arrangement of items in the drawer. Students had never really opened any drawers, so they had to think up “plausible” objects that might have been in the drawer and to imagine how it might have been arranged. Their eye movements were recorded with a video camera focused on their face while they told their stories and answered questions and those videos were later analyzed by an independent party.

Each student also performed the truth part of the experiment where they actually did put the phone in the drawer and therefore could truthfully name objects they saw in the drawer and its layout. Their eye movements were also recorded. Of interesting note, the contents of the office desk drawer were rotated for each trial and items were randomly picked out of an inventory of everyday office objects that might actually be in one’s drawer (like a stapler, an apple, an umbrella.)

Ok so looking upward, not necessarily a sign of lying.

Ok so looking upward, not necessarily a sign of lying.

After all of the students had performed both the lie and the truth conditions, the audio tracks of each interview were removed from the video clips and independent raters evaluated the eye movements by watching the video without sound, rating how many times and for how long each person looked up and to the left and up and to the right. They had no idea who was lying and who was telling the truth since the audio was removed and labels were hidden. The results showed no significance between liars and truth tellers in regards to the amount of time they took to answer and their eye movements were not indicative of the proposed gaze behavior theories of NLP.

So, your hunch about the astrophysicist was right, he was telling the truth (even though his head was a bit in the clouds.) Neuro Linguistic Programming was wrong, noting who looks up and to the right and up and to the left can not and will not help you to identify the liar. Then what will?

The Editor in Chief brings in the next person. You greet her and she sits. She introduces herself and you ask her to tell you a bit about what she does here. She explains she is also an editor and writes mostly about paleontology, anthropology and animal behavior. Could her animal-like instincts have told her to defy society’s rules and take what isn’t hers? Possibly. She seems a bit nervous and you begin to realize she is blinking her eyes quite a lot. Might this be a sign of a lying? You excuse yourself from the interview and do a quick internet search on blinking and lying. You come across a study that demonstrated that those who are trying to hide their emotions do indeed have a elevated blink rate. You go back into the room and feel like you just may have found your thief. You find her rubbing her eyes. After asking her if she’s ok you look down at her photo on the employee directory list. She is wearing glasses in the photo but today she is not. “What happened to your glasses?” You ask. “Oh, I’m trying out some new contact lenses,” she explains, “but they’re just killing me.” Contacts? Hmpf. That could be why she is blinking so much and rubbing her eyes. You definitely can not conclude she’s a liar based on her blinking. After thanking her for her time, you decide to cross her off the list of suspects for now and she’s dismissed.

Back at the conference room...

Back at the conference room...

Feeling like you don’t want to be misled again, you decide to stop the interviews for a little while until you have adequate time to research and read up a little more on the eyes and lie detection. You get a cup of coffee from the break room and then resume your research.

You come across another interesting study. Some researchers have reasoned that increasing a person’s cognitive load while asking them questions can act as a sort of distractor, making certain “tells” or nervous behaviors indicative of lying much more apparent. They assume that lying takes a lot of cognitive work, to consistently maintain the made-up story, to read the face of the interrogator to see if they are buying it and to be conscious to control your emotions and body movements so as not to be viewed as suspicious.

One study conducted in 2010 demonstrated that the act of asking individuals to maintain constant eye contact during an interrogation increased the individual’s cognitive load and made signs of deceit more detectable. Vrij et al. conducted an truth/lie experiment with a somewhat similar setup to Wiseman. They took 80 college students and divided them into two conditions, the liars and the truth tellers. For those who were unknowingly assigned to the truth section the experimenters asked them to play Connect 4 with another student in a particular room. The second student was really an experimenter. Soon after they began, they were interrupted by someone who came in to wipe off a whiteboard. Then they were interrupted a second time by a different person, this person claimed to be looking for their wallet and when they found it in the room they said there was money missing from it and that both of the participants who had been playing Connect 4 in the room would have to be interrogated to find out if he/she had or had not taken money from the wallet.

The liars bypassed the staged event of playing Connect 4 or being interrupted. They just were instructed to go into the room, take the money out of the wallet, hide it somewhere on themselves and then they given a sheet to study on what they would say happened when they were interrogated for taking the money. In their made up story, they were playing Connect 4 just like the truth tellers and the “script” laid out the staged event as it actually had happened to those in the truth condition. Each person in the truth and lie condition were motivated to be convincing as they were told they would receive 10 pounds if the officer believed their story.

The truth tellers and liars were each interrogated in a separate room by a man dressed like a British police officer. The interrogation was videotaped and the video and audio were later evaluated by an independent rater who was blind to the premise of the study and who carefully watched and evaluated the video and audio from each interviewee for pauses, speech hesitation, speech errors, frequency of arm and hand movements, hand and finger movements, eye blinks, leg and foot movements and gaze aversion.

Some of the liars and some of the truth tellers were told to maintain direct eye contact with the police officer at all times and to not look away. Even though they were told this, it was near impossible for each person to continue looking at the officer constantly but the independent raters who judged the videos later were able to detect the cues of deceit I mentioned earlier easier when the liars had been told to maintain direct eye contact versus when they had not.

With so many things to hide, a liar has his hands full and sometimes can ‘drop the ball.’

With so many things to hide, a liar has his hands full and sometimes can ‘drop the ball.’

Think of it as a cognitive juggling act, the liars had to keep their story straight, give details, maintain an air of calm and constant eye contact all while controlling their possible “tells.” I can see why they may slip a little bit here and there with increased cognitive load (a lot to keep track of and control), you are giving them more and more things to keep up in the air. No wonder some blow their cover.

You decide to use this in your next interview. You tell the Editor in Chief to send the next person. As you reach for a handshake, you get a hug instead. Odd. You ask him questions on what he does here and he explains how he comes and goes often and splits his time between two states. You instruct him to keep looking you in the eye. He looks away as he speaks of how he has his own keys to the place and how he has a deep love for science. He tries his best to hold continuous eye contact but he inevitably looks away. He fidgets, he looks away again. He’s nervous.

You call in the Editor in Chief. “I have found your guy. It was him. He took from the archives.”

Shocked and stunned, the suspect breaks down and confesses everything. “I just wanted something to read for the long train rides home,” he cries.

“Tell it to the judge,” you remark. Though the Editor in Chief herself shows sympathy for the science thief and explains that as long as he brings back all of the vintage issues he took she will not press charges.

“Judging from the looks of him, I knew you wouldn’t. Seems like his only crime is reading science a little too much.”

“If you only knew”, she says, “thanks for everything. Job well done.”

“Don’t thank me,” you say, “thank those who really search for truth. Thank the research scientists.”

Afternote: The story you have just heard is fake but the research studies are real. The names of employees at Scientific American magazine were not given to protect the innocent. The sole purpose of this story was to reveal to you the results of some recent studies on eye movements and lie detection. No vintage copies of Scientific American were harmed during the making of this story.*

*Just to be clear, no one has stolen vintage Scientific American magazines from the SciAm offices but SciAm has made it quite easy for you to sneak a peek at classic issues ‘From the Archives’ without running the risk of a criminal record. Simply download from Scientific American Digital and make a quick getaway.

Photo credits: stock xchng photos: The Great Detective (Thomas Romer); Secure (Frank Kohne);  alert but aloof (sxc user); Meeting Room (Razvan  Caliman);  Juggler (Emiliano Spada); Author pic: (Erica Angiolillo/Gotcha by Erica).


Bandler, Richard and John Grinder. Frogs into Princes: Neuro Linguistic Programming. Real People Pr, June 1979. Print. ISBN-10: 0911226192

Porter S, Ten Brinke, L. Reading between the lies: identifying concealed and falsified emotions in universal facial expressions. Psychol Sci. 2008 May; 19 (5): 508-14. PMID: 18466413

Vrij, Aldert, Mann, Samantha, Leal, Sharon and Fisher, R.  ‘Look into my eyes’: can an instruction to maintain eye contact facilitate lie detection? Psychology, Crime and Law. 2010 May 16 (4). pp. 327-348. doi: 10.1080/10683160902740633

Wiseman R, Watt C, Ten Brinke L, Porter S, Couper SL, Rankin C. The eyes don’t have it: lie detection and neuro-linguistic programming. PLoS One. 2012; 7(7) e40259. Epub 2012 Jul 11. PMID: 22808128

Cheryl Murphy About the Author: Cheryl G. Murphy is an optometrist and freelance science writer living and working in New York State. She began writing about vision science on her blog,Science Hidden in Plain Sight, in 2008. Links to her previous contributions to Scientific American’s guest blog can be found here. Follow her on Facebook or Twitter. Follow on Twitter @murphyod.

The views expressed are those of the author and are not necessarily those of Scientific American.

Comments 18 Comments

Add Comment
  1. 1. julianpenrod 11:49 am 08/16/2012

    This may cause this not to be printed, but this article is filled with characteristic failings in the fraud called “science”.
    How many noticed the emphasiws on coolelge students for the studies. How many realize that college students can be considered a specialized population without many of the qualities to be seen so much more in the general population? No life and death responsibility, off time capable of being more devoted to fun, things like poses of superiority being more eminently important. They “proved” eggs cause cholesterol rise by interviewing only a population of individuals who went for treatment for high cholestrol. Carefully, they ignored the likelihood that their subjects had a particular tendency to devleop high cholestrol!
    Consider, too, the eminent failure in the “experiments” to consistently ignore that, if someone is tpold to lie, maybe they won’t act like someone lying on their own choice! Again, the lack of eminent responsibility! You won’t get blamed, you won’t pay if you’re found out!
    And note the casual reference to assuming that a lie requires extra cogitation to keep the lie straight. What of someone lying who follows the rule, “Keep it simple”?
    And why doesn’t the “detective” check the second individual’s eyes to see if she really has contact lenses in them?
    Basically, a contrived story to “validate” certain “studies” while conveniently undermining previously held ideas. An example of the diffide3nt application of “Old wives’ tales are always wrong”, conveniently ignoring the fact that an “old wives’ tale” wouldn’t persist if it didn’t have validity. But that goes with another conceit trundled to con so many, namely, “Everybody back then was stupid! Everybody today is a genius!” Apparently, even if “back then” was only a few decades ago. Apparently selling the idea to those so shallow their mentality only sretches back a couple of years. The New World Order must be planning a lot of overt lying in the near future and this is intended to con people so they don’t realize when they’re being lied to.

    Link to this
  2. 2. rogersgeorge 12:50 pm 08/16/2012

    Whereas I figured the info about eye movement was the point of the story, I also figured it was a red herring for the mystery. I can’t imagine that the first guy, someone covering the Curiosity landing on Mars, would leave at 1 AM! The landing took place at 1:32.

    Link to this
  3. 3. saganista 12:54 pm 08/16/2012

    After that first comment, I’m almost embarrassed to add my two cents. For what it’s worth: I would have been more suspicious of the first interviewee since he claims he went home at 1am, and the Curiosity landing happened after 1:30am EST. I know the intent of the article was to illuminate the various techniques for detecting lying but fact-checking the subject’s statements seems to be the simplest and most accurate method.

    Link to this
  4. 4. Bora Zivkovic 12:59 pm 08/16/2012

    Not sure if, in his general disdain about science (and why then read a science site?) @julianpenrod totally misread the article? After all, the post is critical of many of these studies and explains why.

    Link to this
  5. 5. toohardforscience 1:56 pm 08/16/2012

    Oh, Bora, that’s crazy talk. Why should commenters demonstrate reading comprehension regarding articles they comment on? The Internet’s great for allowing reaction without analysis or much in the way of consequence. ;-) — CQC

    Link to this
  6. 6. julianpenrod 2:29 pm 08/16/2012

    The utter simplicity of the mistakes those who atack me make.
    Bora Zivkovic claims I “misread the article”. Where? The article condemned older statements about eye movements and blinking, using purely artificial condition “experiments” apparently fabricated to overturn the older ideas. But I pointed out that the populaton chosen is conveniently not considered as being unrepresentative of the population in general, and I also indicated the patently eminent fact that, if people are instructed to lie, and they won’t suffer if they’re found out, it could change the way they act during the lie! The older experiments show no sign of being based solely on special poipulations under special conditions, which is why they do not eserve the mistrust of the modern “experiments”. The fact is, a lot if not most of modern “science’ is just concocting fraudulent “experiments” to “disprove” previous discoveries, based on the constant discovery paradigm, as interpreted by the dim, but ignoring the fact that, if a truth is discovered, presumably, it isn’t different if you perform the “experiment” at a later time.

    Link to this
  7. 7. Bora Zivkovic 2:47 pm 08/16/2012

    @julianpenrod – I am sorry the blogger did not write the post you wanted her to write. Instead, she wrote the post she wanted to write. And I edited it. And I like it. And your addition is irrelevant to it.

    One of my jobs here is to protect my bloggers from commenters who take discussions on useless tangents. Stay on topic, and be nice to our guests here on the Guest Blog.

    I use the “three strikes and out” moderation system here. This was your first strike.

    Link to this
  8. 8. murphyod 3:08 pm 08/16/2012

    @rogersgeorgeand @saganista- I did say “around” 1 am… perhaps I should have said around 2 am instead ;)

    Link to this
  9. 9. SRSwain 4:07 pm 08/16/2012

    Too clever by far. Why not just tell the facts and leave the cleverness to the Huffington Post? Their stuff is often largely fictionalized or otherwise adulterated (no pun intended), anyway. I suspect that most of your readers don’t have as much time to get down to the bottom line as this article assumes. “Now brevity is the soul of wit…” (Shakespeare: Hamlet: Act 2, Scene 2).

    Link to this
  10. 10. Deception_researcher 4:21 pm 08/16/2012

    I quick reply to one comment. It is true that college students are a special population, different in many ways from most humans. It is also the case, that in this particular area of research, college students perform pretty much the same as non-student samples. I use both student and non-student samples, and have always found the same thing regardless. And, its not just me. Meta-analysis on the issue is clear. Student-non-student is not a predictor of study outcomes on this particular topic.

    That said, there are always issues with validity. In my read, the eye direction study was convincing. I’m more skeptical of the cognitive load ideas because sometimes it is difficult to remember the truth and just easier to make things up. In this, the post makes a good point. McCornack (1997) makes this argument persuasively.

    My advise is if you want to catch a lie, listen to what is said and fact check if possible. Look for inconsistencies with facts. Ignore nonverbal behaviors.

    Link to this
  11. 11. Cassai 4:28 pm 08/16/2012

    Thanks for helping debunk the NLP eye-movement hogwash, @murphyod. It’s a stubborn myth that just doesn’t want to die.

    The cognitive load technique is promising. It’s often used by law enforcement in the form of having the person of interest tell their story backwards (from the end to the beginning). Studies show that recalling actual events in backwards order is much easier than recalling stories either partially or totally fabricated. Overt deception “tells” are more likely to show up even if the liar gets his/her facts straight in reverse order, because, as mentioned in the article, it takes more brain juice to lie convincingly*.

    *[People high on the psychopathy scale, and/or habitual or pathological liars, deceive without any added cog load or physiological symptoms. They are often "invisible" to all known forms of deception detection]

    Spotting lies through neuro-imaging of the brain is the most successful technique so far, but the idea of someday having that technology available in every cop shop and courtroom is a scary prospect.

    Link to this
  12. 12. murphyod 5:05 pm 08/16/2012

    @SRSwain- Just wanted a new way of presenting a topic. I don’t think it was too too long but see your point. I think most who wouldn’t have time to take the ‘fictional journey’ of the article would have probably stopped reading after the first paragraph anyway, don’t you? =)

    Link to this
  13. 13. murphyod 5:17 pm 08/16/2012

    @Deception_researcher- Thanks for the additional input and you bring up some interesting points. I could see how some could just be very cool under pressure and very good liars (or actors if you will) and therefore cognitive load may have less of an effect if any on them.

    Also to that point, some individuals may just be anxious to be interrogated or they may believe they are telling the truth (when actually they themselves have the facts wrong… the book The Invisible Gorilla is a great read and brings up a case where someone who was attacked and is completely convinced of the identity of their attacker only to find out years later, through DNA testing and another criminal’s confession that the attacker was be someone else entirely. The victim THOUGHT they were telling truth.)

    So I agree completely with what you said, “if you want to catch a lie, listen to what is said and fact check if possible. Look for inconsistencies with facts. Ignore nonverbal behaviors.”

    Lie detection through the eyes or other “tells” is not the best way to solve a crime (obviously). Though it is fun to learn about.

    Link to this
  14. 14. murphyod 5:31 pm 08/16/2012

    @Cassai- Thank you. =)

    Funny, I was actually debating using the cognitive load technique of telling the story backwards as for one of my suspects but thought the forced (as constant as humanly possible) direct eye contact would be what people could relate to the most …(I think at one point or another in our lives we have all either said or been told, “look me in the eyes and tell me the truth…”

    I also saw the study that came out this spring on the ‘lie detecting machine’ but decided to stay far away from that one…

    I think “lie detection” is a very tricky technique as I said before and should not be the primary tool used to solve a mystery. =)

    Link to this
  15. 15. Cassai 9:43 pm 08/16/2012

    I liked your story, murphyod! Putting these studies in the context of “real-world” situations (or fictional real worlds) breathes some life into this branch of research. Research that aims to predict how humans behave in real-world situations, but is rarely put in that context without being hopelessly mangled (ie The Mentalist, dontgetmestarted).

    Deception detection is becoming more mainstream in corporate hiring practices too, where, for better or worse, interviewers are loading up on the latest studies, attending “classes”, and trying to spot the lies as they happen. Bad idea for all of the reasons you and other commenters have pointed out. Still a fascinating subject to me though :) .

    Link to this
  16. 16. murphyod 10:55 pm 08/16/2012

    Thank you Cassai! I really appreciate that. I find the topic fascinating as well! =) I really enjoyed writing it. I actually had two versions written of this article (the fictional storyline and a more, straight-forward stereotypical science article), I am glad people read and enjoyed this version.

    Thanks again =)

    Link to this
  17. 17. Deception_researcher 4:55 pm 08/17/2012

    A couple more deception findings for those interested:

    Psychologist Charlie Bond has found that the most common folk wisdom about liars is that they don’t look you in the eye. Interestingly, this belief appears to hold across cultures.

    Eye gaze and eye contact are completely uncorrelated with actual honesty/deception (See DePaulo et al. meta-analysis).

    Arousal/anxiety behaviors seem to be linked with deception, but not that strongly. The problem is there are really big differences from person to person. Some people come off as nervous even when honest, and some liars are really smooth.

    One recent article by Levine suggested that there are a small subset of people, maybe 10% of the population, that are really bad liars and who give themselves away with their nonverbal behaviors (i.e., a few leaky liars) but that for most people, you really can’t tell based on nonverbal behavior.

    For people who interview suspects, the most promising findings are Hartwig and Granhag’s Strategic Use of Evidence (SUE) approach. So, if you have some case facts/evidence, initially withhold it, and see is the suspect contradicts a known fact. Then you can confront them with evidence a little at a time, watch the story change, and hopefully let them either dig themselves into lies that are clearly lies, or get a confession.

    Link to this
  18. 18. murphyod 10:23 am 08/20/2012

    Thanks @deception_researcher … Very interesting! I especially love the SUE approach, I would like to read up on that some more… Fascinating! I would love to hear a case where the person’s alibi begins to shift and then crumble as evidence is slowly stirred in…I wonder if most confess at that point or just change their story.

    Link to this

Add a Comment
You must sign in or register as a member to submit a comment.

More from Scientific American

Email this Article