a/justice· 31 October 2026 · 6 min read

Why the truth sounds like a lie in a police interview

Cover of The Signal Inversion Effect (Omxus Press)

In September 1988, on Long Island, New York, a 17-year-old named Marty Tankleff woke to find his mother dead and his father dying. Within hours, detectives were questioning him. One told him his father had come out of his coma and named Marty as the attacker. It wasn't true. His father never regained consciousness. Marty confessed.

The confession that took 19 years to undo

Marty took the confession back and said police had written it. A jury convicted him anyway, in 1990. He spent 17 years in prison. In 2007 an appeals court vacated the conviction, and the next year every charge was dropped. New York later paid him $3.375 million (CNN, 2014).

His case is not rare. Of the first 250 people cleared by DNA evidence in the United States, 40 had confessed, most in detail, to crimes they did not commit (Garrett, 2011).

How does an innocent person end up confessing? It starts earlier than the confession. It starts with the first question an investigator asks themselves: is this person lying?

What it's like

You walk into a room. Someone with authority asks you a question. You answer it honestly.

They don't believe you.

You try again. You add detail, because maybe they didn't understand. You stumble over your words because you are scared, and you are trying to remember exactly what happened, and you want to get it right.

They believe you less.

You start to cry because this is terrifying and no one is hearing you. They note that you are "emotionally manipulative". You stop crying and try to be calm and factual. They note that you are "flat".

You say "I didn't do this."

They hear: "I haven't confessed yet."

Nobody can tell

Is anyone in that room able to tell? Researchers have tested that thousands of times. Pooling 206 studies and 24,483 judges, people sorted truths from lies correctly 54% of the time. A coin gets 50%. Police, judges and psychiatrists did no better than students (Bond & DePaulo, 2006).

Training doesn't fix it. Investigators trained to read body language weren't more accurate. They were more confident, and quicker to call people liars (Meissner & Kassin, 2002).

The signs point the wrong way

Ask people how to spot a liar and most say the same thing: liars won't look you in the eye. Of 2,320 people in 58 countries, 63.7% named it, more than any other sign (Global Deception Research Team, 2006). Its real link to lying is d = 0.03, which is nothing (DePaulo et al., 2003). The sign carries no information. The belief in the sign carries all the power.

Set the signs the world believes in beside the largest review of what liars do, and 21 of 23 (91.3%) point the wrong way or nowhere (our analysis; binomial p < .0001).

Some signs run backwards. People telling the truth correct themselves more often and more often say "I don't remember". Liars' stories have fewer ordinary imperfections (DePaulo et al., 2003). In 121 clips of real courtroom testimony, people telling the truth said "um" and "uh" more often than people lying (Pérez-Rosas et al., 2015; our reanalysis). The rates were 5.14 fillers per 100 words against 3.03, an effect of d = 0.60 (p = .004). Remembering is hard work, and the work shows. A liar's story is ready, and comes out smooth.

So the honest witness who pauses, looks away to think, and says "I'm not sure" trips every alarm. The practised liar who looks you in the eye passes. Every move has a guilt-consistent reading. There's no box on the form for "telling the truth".

Then the interview takes over

The dominant interrogation method in the English-speaking world, the Reid Technique, has steps for handling denials and overcoming objections (Inbau et al., 2013). The denial is not weighed. It is overcome. Then the offered way out: "Did you do it because you were angry, or because you lost control?" Both answers are confessions.

A confession, once made, outweighs almost any other evidence in front of a jury (Kassin & Neumann, 1997), and it bends the rest: forensic examiners who learn of a confession see ambiguous findings as incriminating, alibi witnesses waver, eyewitnesses grow more certain (Kassin et al., 2012). Confessing proves you are guilty. Denying proves you are guilty and have not confessed yet.

It lands hardest on people who already read as "odd". Autistic adults telling the truth are rated as more deceptive and less credible than non-autistic adults telling the truth (Lim et al., 2022). Less eye contact, a flatter face, stimming and an unusual rhythm of speech are the things the system reads as lies.

Something that works

Publishing better cues won't fix this. Publish them and everyone learns them: liars stumble on purpose, truth-tellers perform their truth, and the signal collapses. We can't prove that with a measured result, but it's how signals behave when everyone knows them. What does work is not reading bodies at all. Across the studies, accusatory methods got more true confessions and more false ones too, while interviews that gather information kept the true ones and cut the false (Meissner et al., 2014). Better still, assess what a person says that can be checked against other evidence (Vrij et al., 2017). That isn't a better lie detector. It's the end of the lie-detector idea.

People are already moving. Wicklander-Zulawski, a firm that had taught the Reid method to investigators since 1984, announced in March 2017 that it would stop. "Confrontation is not an effective way of getting truthful information," its chief executive said (Innocence Project, 2017). Reid's own company says false confessions come from misuse of its method, not the method. Still, a firm that had made its living teaching this looked at what it was doing and changed. So can anyone else.

This evidence has been published for twenty years. If you were questioned, told the truth, and weren't believed: you weren't crazy. They couldn't tell. They never could.

What you can do

  1. If police want to question you about something serious, ask for a lawyer before you answer, even if you've done nothing. Especially then. Being innocent is not a defence against this.
  2. If you sit on a jury, remember that a confession can be false, and that a nervous person isn't a guilty one.
  3. If you hire or judge or teach, stop reading eyes. Ask what can be checked.
  4. Ask your state police how they train interviewers, and whether they record every interview from start to finish.

Sources

  • Bond, C. F., Jr., & DePaulo, B. M. (2006). Accuracy of deception judgments. Personality and Social Psychology Review, 10(3), 214–234. https://doi.org/10.1207/s15327957pspr1003_2
  • CNN. (2014, January 8). New York man wrongly imprisoned for 17 years gets $3.375 million settlement. https://edition.cnn.com/2014/01/08/justice/new-york-wrongful-imprisonment-settlement
  • DePaulo, B. M., Lindsay, J. J., Malone, B. E., Muhlenbruck, L., Charlton, K., & Cooper, H. (2003). Cues to deception. Psychological Bulletin, 129(1), 74–118. https://doi.org/10.1037/0033-2909.129.1.74
  • Garrett, B. L. (2011). Convicting the innocent: Where criminal prosecutions go wrong. Harvard University Press. https://doi.org/10.4159/harvard.9780674060982
  • Global Deception Research Team. (2006). A world of lies. Journal of Cross-Cultural Psychology, 37(1), 60–74. https://doi.org/10.1177/0022022105282295
  • Inbau, F. E., Reid, J. E., Buckley, J. P., & Jayne, B. C. (2013). Criminal interrogation and confessions (5th ed.). Jones & Bartlett Learning.
  • Innocence Project. (2017). Leading police consulting group will no longer teach the Reid technique. https://innocenceproject.org/?p=47873
  • Kassin, S. M., Bogart, D., & Kerner, J. (2012). Confessions that corrupt: Evidence from the DNA exoneration case files. Psychological Science, 23(1), 41–45. https://doi.org/10.1177/0956797611422918
  • Kassin, S. M., & Neumann, K. (1997). On the power of confession evidence: An experimental test of the fundamental difference hypothesis. Law and Human Behavior, 21(5), 469–484. https://doi.org/10.1023/A:1024871622490
  • Lim, A., Young, R. L., & Brewer, N. (2022). Autistic adults may be erroneously perceived as deceptive and lacking credibility. Journal of Autism and Developmental Disorders, 52(2), 490–507. https://doi.org/10.1007/s10803-021-04963-4
  • Meissner, C. A., & Kassin, S. M. (2002). "He's guilty!": Investigator bias in judgments of truth and deception. Law and Human Behavior, 26(5), 469–480. https://doi.org/10.1023/A:1020278620751
  • Meissner, C. A., Redlich, A. D., Michael, S. W., Evans, J. R., Camilletti, C. R., Bhatt, S., & Brandon, S. (2014). Accusatorial and information-gathering interrogation methods and their effects on true and false confessions: A meta-analytic review. Journal of Experimental Criminology, 10(4), 459–486. https://doi.org/10.1007/s11292-014-9207-6
  • Pérez-Rosas, V., Abouelenien, M., Mihalcea, R., & Burzo, M. (2015). Deception detection using real-life trial data. Proceedings of the 2015 ACM International Conference on Multimodal Interaction, 59–66. https://doi.org/10.1145/2818346.2820758
  • Vrij, A., Fisher, R. P., & Blank, H. (2017). A cognitive approach to lie detection: A meta-analysis. Legal and Criminological Psychology, 22(1), 1–21. https://doi.org/10.1111/lcrp.12088

Keep reading