To Tell the Truth


When was the last time you told a lie?

As I’ve written before, I’ve been doing investigations and security consulting for a loooong time.  I just did the math in my head, and it’s a depressingly large number!  It’s been an interesting career, mostly good, and the source of endless variety.  I have done most of the things people think about private eyes doing, but usually it turns out being a bit less glamorous than movies and TV shows might suggest.  But my career has also led me to do strange, bizarre things I never could have anticipated.  Through it all, one of my favorite things to do has been interviewing people who are suspected to be lying, and trying to first determine if they are lying and if so, convince them to tell the truth.

Conducting this type of interview, some might call it an interrogation, is a fascinating endeavor.  I think the image most people have is of a dogged police detective in a small interview room, the suspect handcuffed to the table.  There may or may not be bright lights and rubber hoses involved.  To be sure, interview and interrogation skills are invaluable tools to the police investigator.  But having never been in law enforcement myself, convincing someone to be honest with you, sometimes against their own self-interest, requires a slightly different set of skills when you don’t have the authority of a badge behind you.  Someone might be motivated to come clean when they are facing the threat of imprisonment.  But if the investigator doesn’t have that kind of leverage, how does the job get done?

There are dozens of scholarly articles about interviews and deception detection.  There are several well-known training programs and methodologies that seek to instruct people on the art of deception detection and leading suspects to confessions.  I’ve studied these at length, and I think my favorite was one developed by a former agent of the Mossad, the Israeli intelligence agency, that applied some of these deception detection principles to analyzing written statements.  Nonetheless, the central theory of these programs is that deceptive communication simply looks, sounds, and “feels” different from honest communication.  Most of us have heard of things that “always” indicate deception like avoiding eye contact, covering one’s mouth when one speaks, or looking to the left (creating) instead of looking to the right (remembering).

In my experience, none of these are true, reliable indicators of deception.  Rather, the skilled interviewer must understand the suspect’s holistic communication style then look for deviations from that style.  It’s never just one behavior, but behavior has typically been a big part of deception detection.

Which is why I was intrigued a few years ago when one of my friends invited me to a symposium on the topic of deception detection.  My friend, a professor in the area of interpersonal communication, is one of the most intelligent people I know, so I take his opinions seriously.  His department was hosting Dr. Timothy Levine, Distinguished Professor and Chair of Communication Studies at the University of Alabama at Birmingham and Global Professor of Communications and Media at Korea University, Seoul.  Dr. Levine was speaking about his Truth Default Theory.  You can read about his theory in depth here, but allow me to summarize some key findings:

  • People tend to tell the truth
  • We tend to assume that other people are telling the truth, which is usually the case
  • This presumption of honesty essentially allows society to function
  • Deception is suspected when there is an obvious motive for deception, when they lack an honest demeanor, when we are predisposed to expect deception, or when communication content is self-contradictory or inconsistent with known facts
  • In clinical trials, deception detection efforts based on demeanor were less successful than deception detection efforts based on content.

It was that last point that was the subject of much… energetic… discussion between my friend and me.  I had years of practical experience telling me that behaviors, basically demeanor, were useful in detecting deception.  He was pointing to academic studies demonstrating that demeanor was not a useful tool.  We were in the midst of a fun, and not unique, conflict between practical and theoretical.

We at Murder Books are writers with real-world experiences in the world of investigations and the law, and our hope is that our experience will make for more compelling, grounded, and exciting thrillers and mysteries.  Understanding the psychology behind our characters, why they lie, why they kill, and why they are driven to find the truth adds depth and relatability to our characters, and the more authentic they are, the more our readers will recognize these characters as real and engaging.

My friend and I never really resolved our slight disagreement about the Truth Default Theory, but we agreed that the reasons people lie are varied, and can be complex and fascinating. When you read about a criminal who lies in the next mystery you read, maybe you’ll remember some of this, and you’ll be able to judge for yourself the reasons for their actions, and how well the investigator discovers the deception.

Or maybe I’ve been lying to you all this time.

–Ben Keller

9 thoughts on “To Tell the Truth

  1. Great topic, Ben. And lots of good great points. It’s easy to know a guy is lying when he says he’s innocent and you have 5 eye witnesses to the murder and his prints on the gun. It’s harder when you have to rely on what he says and does in the interview room. There was an old study that compared the lie-detection accuracy of the polygraph, voice stress analyzer, and experienced detectives with 1000+ hours experience interviewing suspects. All where the same at about 80%. Ask those experienced detectives how they knew the person was lying and they couldn’t point to any one thing, but they knew.


  2. We’ve been talking (behind the scenes) about this at MissDemeanors. I am a believer in the content test. Perhaps because sometimes I find myself looking off in the distance when talking/explaining something and I think – but I’m not lying…. I’m merely too deep into re-call to stay focused on the person I’m talking to! On the other hand we as individuals have ‘ticks’ and I suppose that if they are factored in the anomalies might point to the lies versus the truth telling.

    Either way, thanks for the great peek into the art of lie detection.


    1. Thank you, Tracee! I think everything, including behaviors, have to be taken into context. A good interviewer spends time letting the interviewee talk, so that a baseline can be observed. I’m glad you enjoyed this!



  3. I know you Ben Keller and you never lie…….well there was that one time when…..but thats another story.
    Interesting topic. I have found that if you just let the subject talk long enough they will usually make it obvous for you. Even without the rubber hoses.


  4. I have pretty good experimental evidence for content over demeanor. Its not that demeanor isn’t sometimes useful… it can be. Experienced investigators know this first hand. But, sometimes it is really misleading. I did an experiment with experienced federal interrogators showing them interviews. For one set, demeanor was diagnostic, and the experienced investigators got 100% right. But on a set of smooth liars, they got only 16%.

    And, its not just my lab. This field experiment provides the strongest evidence yet:

    The point about persuading people to be honest is a really good one. I my research, this is one of the things that makes an interrogator good.

    By the way, in academic research, more researchers would side with demeanor. My opinion is in the minority. Things are slowly shifting, but the demeanor approach has many advocates.

    People interested in my work can check out:


    1. Wow, I am honored, Dr. Levine! It’s not every day the person I cite comments on my post! Next time I’ll include a Bible quote and see if God shows up in the comments…

      I hope my trifling post in a mystery writers blog didn’t diminish or misrepresent your fascinating work. Hearing you speak years ago and subsequently reading of your research has challenged my thinking and in truth, has influenced the way I’ve conducted interrogations. I look forward to further research on the matter. If I can ever help with connecting you with industry practitioners for research purposes, please reach out.



  5. In my experience, status or role also has a place. For instance, parents and teachers are surely the best of the best when it comes to lie-detection. Although I am not a parent, if my memory of my childhood serves me correctly, my record of fib-success against my parental units was about zero versus a . . . uh . . . large number. I was, however, a teacher for nearly two decades, and in that business, one quickly develops a pretty sharp nose for the untrue. Whether this skill in parents and teachers is more heavily attributable to content or demeanor, I don’t know, but surely their familiarity with the baseline behaviors of the liars (ahem, I meant to write children and students) in question is very helpful. And in the closely allied field of BS-detection, I think parents and teachers are also probably way ahead of the pack. I was going to respond to this post earlier, but my dog ate my reply. In any event, this was a really interesting post, Ben. No, really . . .


    1. You made me laugh out loud, Roger!

      I think familiarity is a huge component. I’ve stated before that a good interviewer will invest time in learning the interviewee’s baseline communication style. If that time is 20 years instead of 20 minutes, then the ability to detect deception will only be greater!


Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Google photo

You are commenting using your Google account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s