a/justice· 25 September 2026 · 4 min read

Can police tell when someone is lying?

No. People judge lies correctly about 54% of the time, and police do no better than students. The cues most of us trust, like eye contact, carry almost no information, so honest people who are frightened, grieving or autistic often look "guilty".

54%: barely better than chance

In 2006 Charles Bond and Bella DePaulo pooled 206 studies of lie detection covering 24,483 judges. On average people sorted lies from truths correctly 54% of the time. Guessing gets 50%.

The errors lean one way. People spotted 61% of truths as true but only 47% of lies as lies. Most of us assume we are hearing the truth, which makes daily life workable. It also means the rare person flagged as a liar is often an honest one caught by a bad signal.

They also found that people who judge lies for a living, such as police officers, did not do meaningfully better than untrained people.

More confident, not more accurate

Saul Kassin and colleagues filmed prisoners giving two confessions each: a true one to their own crime and an invented one to a crime they did not commit. Police investigators and students then judged which confessions were true.

The investigators scored 53.9%, no better than a guess. Students did slightly better. The investigators felt much more sure of their answers, and leaned towards calling confessions true. When told half the confessions were false, they lost that lean but gained no accuracy.

That gap between certainty and accuracy matters most. A confident wrong judgement early in an investigation shapes everything after it.

The eye-contact myth

The Global Deception Research Team asked people in 75 countries how to spot a liar. About 64% named the same sign: liars look away. It was by far the most common belief worldwide.

It is wrong. A review of more than 100 studies by DePaulo and colleagues found that eye contact had almost no link with lying. Liars do not reliably look away, and truth-tellers do not reliably hold your gaze.

So people read lies from cues that carry no information. And many honest reactions, such as looking away while trying to remember, stumbling over words, crying or going flat and quiet, get read as evasive or cold.

Autistic people pay a higher price

Averted gaze, flat affect, repetitive movement and literal answers are common in autism. They are also what many people think lying looks like.

Ailsa Lim, Robyn Young and Neil Brewer at Flinders University showed 1,410 people videos of 30 autistic and 29 non-autistic adults. When telling the truth, autistic adults were rated more deceptive and less credible. The judgement came from the person's overall manner, not from any single behaviour.

Why false confessions happen, and stick

Confession-driven interrogation, such as the Reid technique used widely in the United States, starts by judging whether the suspect is lying. It then treats denials as something to overcome. If that first judgement rests on cues that do not work, innocent people end up in the interrogation room.

Brandon Garrett studied the first 250 people freed by DNA evidence in the United States. More than 40 had confessed. Almost all of those confessions contained accurate crime details the public did not know, details most likely passed on, knowingly or not, during questioning. Juries heard those details as proof of guilt.

Kassin calls this "corroboration inflation". Once a confession exists, witnesses become more certain, forensic examiners read unclear results as matches, and the confession outweighs evidence of innocence.

What Australia changed

From the 1990s, Queensland police and then other Australian states moved from Reid-style interrogation to the British PEACE model. PEACE aims to collect a full account rather than a confession, and it drops behaviour-based lie detection. Lisanne Adam and Celine van Golde reviewed how far that change reached and found it applied unevenly across forces. The model on paper is better than the practice.

What this means for you

  • Your demeanour is not evidence. If you are interviewed, frightened reactions do not prove anything, and good interviewers know that.
  • Ask for a lawyer before you answer questions. A lawyer can tell you what your right to silence covers in your state.
  • If you judge others, on a jury, at work or at home, look for evidence that can be checked: records, times, places. Leave out how someone looked while they spoke.
  • If the person is autistic or neurodivergent, say so early, and ask that a support person attend.

Sources

  1. Bond, C. F., Jr., & DePaulo, B. M. (2006). Accuracy of deception judgments. Personality and Social Psychology Review, 10(3), 214–234. doi.org/10.1207/s15327957pspr1003_2
  2. Kassin, S. M., Meissner, C. A., & Norwick, R. J. (2005). "I'd know a false confession if I saw one": A comparative study of college students and police investigators. Law and Human Behavior, 29(2), 211–227. pubmed.ncbi.nlm.nih.gov/15912725
  3. Global Deception Research Team. (2006). A world of lies. Journal of Cross-Cultural Psychology, 37(1), 60–74. doi.org/10.1177/0022022105282295
  4. 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. PDF
  5. 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. Flinders University
  6. Garrett, B. L. (2010). The substance of false confessions. Stanford Law Review, 62(4), 1051. SSRN
  7. Kassin, S. M. (2012). Why confessions trump innocence. American Psychologist, 67(6), 431–445. PDF
  8. Adam, L., & van Golde, C. (2020). Police practice and false confessions: A search for the implementation of investigative interviewing in Australia. Alternative Law Journal. doi.org/10.1177/1037969X19874415

Keep reading