Detecting deception

Most people know Pinocchio; the wooden puppet whose nose grew immediately when he told a lie. Although we are all experts at lying, we find it difficult to recognise lies in others. There is no physical reaction that occurs only when we lie and no behaviour that is a flawless diagnosis of lying. There is also no brain area that is active only when we lie, because lying is a very complex matter involving several areas. Indeed, when lying, one has to know and suppress the truth on the one hand and on the other hand come up with a convincing lie while appearing as credible as possible. 

Nowadays, erroneous assumptions are still common that there is a connection between signs of nervousness, anxiety and stress with lying. Even experienced investigators regularly say that in the interrogation room they can tell if the suspect is lying by their body language (sweating, blushing, shaking and fidgeting). However, they cannot distinguish whether the suspect is afraid of being discovered or afraid of not being believed (the Othello effect). Poor lie detection accuracy is mainly related to relying on weak and unreliable evidence of deception. For example, despite widespread belief, liars do not avert their gaze more than truth tellers. Very often, such erroneous assumptions are subliminally based on fear, stress and nervousness. However, there is no proven connection between this kind of nervousness and lying.

However, in high stakes situations, such as criminal proceedings, accurate and reliable deception detection techniques are very important. Deception detection methods should be based on a sound scientific framework, reliably indicate possible involvement in a crime and avoid wrongful imprisonment. But is this possible?

Even currently used techniques for detecting dishonesty, such as the polygraph, are based on the idea that a fear or stress response reveals deception (In Germany, the use of such polygraphs is not permitted in criminal proceedings). Such unreliable lie detection tests based on stress-related cues should be avoided because they mainly falsely identify innocent people as liars. There is, however, a method to detect memory traces, which is about detecting intimate crime details and not about detecting deception. This test is called Concealed Information Test(CIT). Where with the polygraph one tries to establish deception by interpreting answers to interrogation questions such as "Did you kill Mr X?" to detect deception, the purpose of the CIT is to check whether the suspect has knowledge of certain criminal information. For example, the CIT asks whether the murder weapon was a bomb, a firearm or a knife. This method is therefore not called a lie detection test, but a memory recognition test. The CIT is thus more reliable and less prone to false positives: the risk of an innocent person being falsely indicated as a liar by this test is very unlikely. 

The way people talk can also be an indicator of lying. People who tell the truth are more likely to give spontaneous details: Colours, smells, times, and witnesses. They correct themselves if they have left out details and admit it if they have forgotten something. Liars often stick to the one story they have memorised. They are less detailed and can give fewer details. Research also indicates that people who tell the truth can make statements that are factually verifiable, whereas liars make statements that are vague and uncontrollable. 

Because various techniques can assist in the detection of deception, the decision remains complicated. Especially when there are major legal consequences involved, it is important to assess other evidence objectively.


  • Ben–Shakhar, G. (2012). Current research and potential applications of the Concealed Information Test: An overview. Frontiers in Psychology3: 342.
  • DePaulo, B. M., Lindsay, J. J., Malone, B. E., Muhlenbruck, L., Charlton, K., & Cooper, H. (2003). Cues to deception. Psychological Bulletin129(1), 74.
  • Nahari, G. (2018). The applicability of the verifiability approach to the real world. In Detecting concealed information and deception (pp. 329-349). Academic Press.