Your Brain on Lies, Damned Lies and ‘Truth Serums’

question the answerPork pies, fibs, whoppers, untruths, tall stories, fabrications, damned lies…not to mention statistics.

Apparently, every person lies an average of 1.65 times every day. However, since that average is self-reported, maybe take that figure with a pinch of salt. The truth is, most people are great at lying. The ability to conjure up a plausible alternative reality is, when you think about it, seriously impressive, but it takes practice. From about the age of 3 young children are able to make up false information, at a stunning rate of one lie every 2 hours – though admittedly the lies from a toddler’s wild imagination are relatively easy to identify.

When we lie, brain cells in the prefrontal cortex – the planning, ‘executive’ of the brain – work harder than when we tell the truth. This may be reflected in the physical structure of our brains as well: pathological liars have been shown to have more white ‘wiring’ matter and less grey matter in the prefrontal cortex of their brain than other people. But how can we tell if someone is telling a lie, or telling the truth?

Back in the day – 2000 years ago – in ancient India, people would use the rice test to spot liars. When someone is lying, their sympathetic (‘fight or flight’) nervous system goes into overdrive, leading to a dry mouth. If you could spit out a grain of rice, you were seen to be telling the truth. If your mouth was parched and you couldn’t spit the grain out, you were lying. Since then, several different methods of catching out liars have been used – to varying levels of success.

In several books and films (Harry Potter, True Lies, The Hitchhiker’s Guide to the Galaxy and many more), a ‘truth serum’ is used to elicit accurate information from the recipient.  In actual fact, however, truth serums don’t exist.  Apart from in fiction, their title is an ironic misnomer. Having said that, scientists have tried for decades to develop a failsafe ‘veritaserum’ in order to catch out liars.

wineAlcohol has been used as a sort of lie preventor for millennia, as the Latin phrase ‘in vino veritas’ (in wine [there is] truth) demonstrates. Alcohol acts in the brain by increasing the activity of GABA channels, leading to a general depression of brain activity. This has been thought to suppress complex inhibitions of thoughts and behaviours, loosening the drinker’s tongue. However, drinking alcohol doesn’t prevent people giving false information, and it by no means prompts people to tell ‘the truth, the whole truth and nothing but the truth’.

A drug called scopolamine was used to sedate women during childbirth in the early 20th century when a doctor noticed that the women taking the drug would answer questions candidly and accurately. Scopolamine acts on GABA receptors in a similar way to alcohol, and so a person intoxified with the drug is just as likely to give false information as someone who’s had a few stiff drinks.

Barbiturates are sedatives such as sodium amytal that work on the brain in a similar way to alcohol – by interfering with people’s inhibitions such that they spill the beans. Sodium amytal was used in several cases in the 1930s to interrogate suspected malingerers in the U.S. army, but the drug does not prevent lying and can even make the recipient more suggestible and prone to making inaccurate statements.

headIn the 1950s and 60s, the CIA’s Project ‘MK-ULTRA’ tested drugs such as LSD on unconsenting adults and children. If LSD proved a reliable truth serum, it would be an invaluable tool in the Cold War. The tests showed that LSD would be far too unreliable and unpredictable to use in interrogation.

Despite the repeated lack of success in the search for a ‘truth serum’, scientists have continued trying to develop alternative technologies for busting liars. The polygraph, used by respected institutions including the CIA , FBI and The Jeremy Kyle Show, measures changes in arousal – heart rate, blood pressure, sweating, and breathing rate – in order to detect deception. However, there is a lot of scepticism surrounding polygraphy. In particular, there are several hacks to avoid getting caught out by a polygraph – most notably biting your tongue, difficult mental arithmetic, or tensing your inner anal sphinchter without clenching your buttocks (thanks for that factual gem, QI).

The improvement of brain imaging methods – in particular functional magnetic resonance imaging or fMRI – has extended the scope of detecting liars. On the internet, one might stumble across ‘No Lie MRI’, an American firm that offers a lie detection service for individuals, lawyers, governments and corporations. They claim that this service could be used to “drastically alter/improve interpersonal relationships, risk definition, fraud detection, investor confidence [and] how wars are fought.”


Currently James Holmes, the man charged with injuring 70 and killing 12 at the Batman cinema shooting in Aurora, Colorado, is on trial. The judge has ruled his consent to a “medically appropriate” narcoanalytic interview and polygraph. That is, Holmes could be interviewed under the influence of sodium amytal or other similar drugs in order to determine whether or not he is feigning insanity. The use of these drugs may contravene the U.S. Constitution’s 5th Amendment ‘the right to remain silent’. Clinical psychiatrist Professor Hoge says, “The idea that sodium amytal is a truth serum is not correct.  It’s an invalid belief. It is unproven in its ability to produce reliable information and it’s not a standard procedure used by forensic psychiatrists in the assessment of the insanity defence, nor is polygraph.”

The potential benefits of a 100% reliable, valid method of lie-detection are obvious, although there are ethical grey areas that scientists and the legal/ethical community would need to tackle if the technology is ever found. For now I think the evidence for using current lie detection methods, especially for anything more serious than The Jeremy Kyle Show, is far too sparse.

Post by Natasha Bray

3 thoughts on “Your Brain on Lies, Damned Lies and ‘Truth Serums’”

  1. Well now, you would think the author of this posting would fess up, those machines are now a reality, not with out there negative affect however, hence a ethical question arises regarding there use.
    This speculation, if correct, would certainly change the face of intelligence gathering and assessment, plus be of great strategic importance.

    • Dear Frank,

      Thank you for your comment! These machines (I am assuming you are referring to fMRI and polygraph) are a reality, despite having shown varying rates of successful lie detection. Polygraph, for instance, has shown anything between 50-95% success, and fMRI has shown up to about 90% success. You are very right about there being ethical issues arising from this.

      There the question of the success rate not being good enough in the ‘real world’, for example in a court of law. Personally, I think that if one in ten ‘lies’ are not/falsely identified it is not acceptable in situations where the outcome matters. I don’t think that 1/10 fulfils ‘beyond reasonable doubt’. There are several differences between the trials on the general public in a laboratory vs. real life situations. For example, some mental illnesses and personality disorders interfere with the nervous system, so the success rate may vary in these individuals considerably more than in the people in the trials. Also, in the laboratory trials, the researchers can put in an incentive to lie (e.g. money), but they could not put a negative incentive (e.g. time in prison) on the test subjects. The brain has complex reactions to fear and stress, so the prospect of a punishment if someone is wrongly accused of lying could trigger a stress response that could make it look like that individual was lying. As far as I know, there are no experiments that test these techniques in the field (e.g. on criminal defendants) to see whether they are valid in these situations.

      Secondly, there is the question of whether a defendant/witness should be subjected to this type of technique. Don’t get me wrong, in some places judges have deemed the techniques to be suitable, and I am not in any way legally qualified. However, I think it is a legally contentious point in terms of the right to free speech and the right to a fair trial based on airtight evidence, not to mention the treatment of mental illness. If a person refused to sit in a scanner to prove their innocence, would that be evidence of guilt or simply exercising the right to remain silent? How would the questions for lie detection chosen? Could the person conducting the test conceivably be able to ask questions relating to sensitive information outside the boundaries of their commission? There are so many ethical questions that would have yet to be resolved for even a hypothetical 100% accurate lie detector.

      If you’d like to read more about the ethics behind these lie detection methods, this is a good review: It was written in 2008 so although fMRI technology may have advanced in the last five years, the ethical arguments still stand.

      Again, thanks for your interest in the blog, we really appreciate people’s feedback on our articles!

      Best wishes,

  2. Problem with those scans/drugs.

    Psychopaths’ brains operate differently. They have been found to not have the same links between thoughts/words/actions as normal humans do. Lie detectors, including MRI’s have been shown to be ineffective.

    So maybe we should just use that MRI to see that they are a psychopath and use more discretion in what they say, because they tend to be better liars than the rest.

Comments are closed.

Share This