Polygraph Tests in the UK Criminal Justice System: An Analysis of the Myths

Polygraph testing, also known as lie detection, has become a topic of discussion in recent years, with more people wondering about its reliability and whether it should be used in certain situations. The Domestic Abuse Act 2021 has brought its use back into the spotlight.In this article, Benjamin Knight will explore what a polygraph is, the evidence for and against its reliability, whether it is admissible in criminal courts in England and Wales, where it is used in the UK Criminal Justice System, and how much the Home Office may have spent on polygraph testing in the past ten years.

The use of lie-detectors in UK Justice

What is a Polygraph?

The polygraph test, also known as a lie detector test, was invented in the early 20th century by a medical student named John Augustus Larson and a police officer named Leonard Keeler. The first polygraph machine was created in 1921, and it was based on the idea that lying causes changes in a person’s physiological responses, such as changes in blood pressure, pulse rate, respiration, and skin conductivity.

Larson and Keeler’s invention sought to revolutionise the field of criminal investigation by providing a new tool for detecting deception. The first applications of the polygraph test were in law enforcement and intelligence gathering, and it was used to investigate a variety of crimes, including theft, fraud, and espionage. It went on to be a staple of trashy television, of course, from Jeremy Kyle and Love Island to every police procedural ever made.

Since its invention, the polygraph test has been the subject of much controversy and debate, with some critics arguing that it is unreliable and can be easily manipulated, while others argue that it is a useful tool for detecting deception.

One of the main criticisms of lie-detectors is that they can be influenced by too many subjective factors, including anxiety, stress, and the examiner’s own biases.

Proponents of lie-detectors argue that they can be highly accurate when used correctly. According to a study published in the Journal of Forensic Sciences, the overall accuracy of polygraph testing is around 80%, although this figure varies depending on the circumstances in which the test is conducted. For example, the accuracy is generally higher when the person being tested is under suspicion for a specific crime, rather than when the test is being used as a general screening tool. It is not a universally accepted accuracy rate amongst scientists. Indeed, there are as many articles quoting percentages are there organisations with vested interests, one way or another.

Critics of lie-detectors argue that they are not reliable and should not be used as evidence in any legal proceedings. One of the main criticisms is that the physiological responses measured by the polygraph can be influenced by a number of factors that are not related to whether a person is lying or telling the truth. For example, a person may be nervous or anxious about taking the test, which could lead to false positives. In addition, there is evidence to suggest that many people can manipulate their physiological responses to produce false negatives.

Moreover, “…the very research programme in psychology (i.e. introspection) which propelled the polygraph into existence has been widely discredited as unscientific due to its lack of methodology, indefensible empirical basis and deficient validity”, according to SAGE Journal in 2020.

Despite these criticisms, polygraph testing remains in use in certain contexts, such as in the management of high-risk offenders and in certain counter-terrorism measures. Section 76 of the Domestic Abuse Act 2021 has extended the scope for polygraph use in offender monitoring by amending Part 3 of the Offender Management Act 2007.

The test itself usually takes between one and three hours, during which the examiner will ask a series of control questions to establish a baseline for the person’s physiological responses, and then ask a series of relevant questions that relate to the matter being investigated. The examiner will then compare the physiological responses to the control questions with those to the relevant questions to determine whether the person is being truthful.

The reliability of lie-detectors has been a topic of debate for decades, with some people believing that they are highly accurate, while others argue that they are little more reliable than chance. Either way, they are not reliable enough for the courts (other than the civil courts in limited circumstances).

Aren’t They Used All The Time in the United States?

Not really, no. The U.S. Supreme Court has held that polygraph evidence is generally inadmissible in criminal trials because it is not sufficiently reliable or scientifically valid. However, some States allow polygraph evidence to be introduced in certain circumstances, such as when both parties agree to its admission, or when it is used to impeach the credibility of a witness.

In addition, polygraph evidence may be admissible in other types of legal proceedings, such as administrative hearings, civil cases, and probation or parole revocation hearings. Some federal agencies also use polygraph testing as part of their background investigation and security clearance processes.

But to imagine that the “feds” whip out their lie detector kit at the drop of a hat is for television and film, I’m pleased to say.

Are They Ever Admissible in the Criminal Courts of England and Wales?

A bit of history as to why not: in Fennell v. Jerome Property Maintenance Ltd, The Times, November 26, 1986, QBD, it was held that evidence produced by the administration of a mechanically, chemically or hypnotically induced test on a witness so as to show the veracity or otherwise of that witness is inadmissible. That was a civil case and they were concerned with hypnosis but it is a good comparator. The criminal courts are more resistant and have a higher standard of proof.

Notably, Practice Direction 19A.1 of the Criminal Procedure Rules governs the admissibility of expert evidence: “expert opinion evidence is admissible in criminal proceedings at common law if, in summary, it is relevant to a matter in issue in the proceedings, it is needed to provide the court with information likely to be outside the court’s own knowledge and experience; and the witness is competent to give that opinion.”

In R v. Dlugosz; R. v. Pickering; R. v. S. (M.D.) [2013] 1 Cr.App.R. 32, the Court of Appeal observed: “in determining the issue of admissibility the court must be satisfied that there is a sufficiently reliable scientific basis for the evidence to be admitted. If there is then the court leaves the opposing views to be tested before the jury.”

In Trochym v The Queen, 216 C.C.C. (3d) 225, the Supreme Court of Canada held that a party wishing to rely on novel, scientific evidence must first establish that the underlying science is sufficiently reliable to be admitted in a court of law.

The Privy Council in Lundy v R [2103] UKPC 28 provided further guidance, which should be considered when assessing the admissibility of new or novel scientific techniques.

Bluntly, polygraphs do not pass any of the above tests.

But the present government, acting on approval by several police forces, have amended the Offender Management Act 2007 to expand the use of mandatory polygraphs to help to monitor the conduct of offenders released after serving prison sentences for more and more offences. Nevertheless, section 30 of the 2007 Act specifically forbids the use of the results of those tests in any criminal proceedings against the person concerned. By ‘results’, the Act clearly includes:

  • any statement made by the released person while participating in a polygraph session; and
  • any physiological reactions of the released person while being questioned in the course of a polygraph examination.
  • The offences (other than terrorism-related offending) for which the tests can be mandatory are limited but may be a little surprising. They are:
  • Breach of a restraining order
  • Specified violent offences (Part 1, Sch. 15 of the CJA 2003)
  • Controlling or coercive behaviour in an intimate or family relationship
  • Sexual offences attracting notification requirements
  • Offences under Part 2 Sch. 15 of the CJA 2003
  • [The Scottish and NI equivalent offences]
  • Equivalent service offences (i.e. in the context of courts-martial)

Testing is only imposed upon offenders who are:

  • Aged 18 years and over.
  • Assessed as very high or high risk of serious harm using nationally accredited risk assessment tools.
  • Convicted of one or more of the following offences: murder, a specified violent offence, or breach of a restraining order where the offence involved domestic abuse; controlling or coercive behaviour in an intimate or family relationship.
  • Sentenced to a term of custody of 12 months or more and released on licence.

Most importantly (when advising lay clients as to their risk of recall due to testing), it should be noted that:

  • People on supervision subject to testing cannot be recalled to custody for failing a polygraph test,
  • They can be recalled for making disclosures during the test that reveal they have breached other licence conditions or that their risk has escalated to a level whereby they can no longer be safely managed in the community.
  • Those failing the test can, however, be tested more frequently and, in addition, they may be given a formal warning or made subject to additional licence conditions.
    Information gathered from a failed examination may be shared with the police who are able to conduct further investigations that may or may not result in charges being made,
  • Offenders who attempt to ‘trick’ the polygraph test, or who refuse to take it can be recalled to custody. This raises questions as to how an attempt to ‘trick’ a test can be identified with sufficient certainty to deny a person their liberty.

All of this is despite the limitations known by parliament, which has had numerous discussions and briefings on the efficacy (or lack thereof) of polygraph testing in criminal investigations. The same conclusions are usually reached. It is a good threat, in effect. The threat can cause admissions to be made but that’s about all.

A 2014 ACPO (as it was called at the time) statement on polygraphs in investigations strongly discouraged its use and pointed out that the risk to the integrity of the investigation (and to any prosecution case) had to be considered. The statement provides that if a polygraph has been used, it must be revealed to the Crown Prosecution Service, and therefore it will likely be disclosed to the defence. It even gives three examples of how the use might taint an investigation and prosecution case:

Where someone has been eliminated from the investigation directly because of the use of a polygraph or on the basis of lines of enquiry following a polygraph test. It could be argued that the whole enquiry was flawed because too much reliance was placed on an invalid technique;

If a suspect is implicated from polygraphic evidence, it could be argued that the investigation followed lines of enquiry that were heavily influenced by ‘confirmation bias’ based on a flawed technology;

If a victim of a crime is seen to be deceptive because of a polygraph, then it would be a mistake merely to drop those enquiries.

It goes on to state categorically that, ‘[t]here are no typical cues to deceit, either through non-verbal behaviour, verbal behaviour, or physiology that can be used within the UK criminal justice system to accurately and consistently discriminate between lies and truth’.

Nevertheless, police forces are “inviting” suspects to take part in polygraph testing, according to anecdotal evidence as recent as March 2023.

The Government is also briefing journalists that polygraph testing is a useful tool in risk management – seemingly on the back of figures from police and NOMS which may or may not corroborate the Home Office’s assertions but for which no methodology has been published.

What Do You Mean “May or May Not”?

The UK Government claims that more than 5,000 polygraph tests have been used on sexual offenders. They go on to claim that two-thirds of those tests have “resulted in significant disclosures” but this is as unscientific as the polygraph. Those disclosures may have been made at the threat of testing, even if no test had been utilised, or may have been declared ‘significant’ according to highly subjective standards. There is no publicly available record of this data or the circumstances in which it was obtained.

Since the amendments to the law on polygraph use, they have been subject to a 3-year pilot (which is not yet complete). The most recent figures I can locate for the confirmed use of polygraph testing at the expense of the tax-payer is for the period between 1st January 2017 to 31 December 2019. During that period, 1,435 offenders were required to undergo a polygraph test. All of them were offenders tested after their release from prison. That has almost certainly changed and does not include testing that individual police forces have implemented, despite the utter uselessness of any outcome derived from the tests. Moreover, compared to the “all time” figures, this suggests a sharp increase in polygraph use (and cost).

Is Parliament Keeping Polygraph Use Under Review?

In a word, “yes”. Westminster seems rather fixated on the use of polygraph testing, despite the scientific scepticism.

The House of Commons Work and Pensions Committee’s report on benefit sanctions (2018) included a section on the use of polygraph testing in employment screening. The report recommended that the use of polygraph testing in this context should be subject to appropriate safeguards and ethical guidelines, and that further research should be conducted into its reliability and validity.

The House of Lords Science and Technology Committee’s report on forensic science (2010) included a section on the use of polygraph testing in criminal investigations. The report concluded that there is no scientific basis for the use of polygraph testing as a reliable form of lie detection, and recommended that the use of polygraph testing in this context should be subject to appropriate ethical guidelines and safeguards.

Politics Over Proof

So, why the obsession with it? Perhaps the answer lies elsewhere than in science. Polygraph testing is sometimes viewed as a way of enhancing public confidence in the criminal justice system, by demonstrating that law enforcement officials are taking proactive steps to prevent crime and to ensure public safety. In this sense, polygraph testing can be seen as a political tool, rather than a scientific or evidence-based one.There may be some institutional momentum behind its continued use and development, despite the scientific scepticism surrounding it.


The field of lie detection has seen some advancements and evolutions over the years. For example:

  • Functional Magnetic Resonance Imaging (fMRI) – This is a technique that uses magnetic fields to map brain activity by measuring changes in blood flow. Studies have shown that fMRI can possibly be used to detect deception by measuring differences in brain activity when a person is telling the truth versus when they are lying.
  • Eye-tracking – Eye-tracking technology can be used to detect deception by measuring eye movements and pupil dilation. Research has shown that people tend to look away or avoid eye contact when lying, and eye-tracking technology can be used to detect these subtle changes in behaviour.
  • Voice Stress Analysis – Voice stress analysis is a technique that analyses changes in a person’s voice to detect deception. The idea is that lying causes changes in vocal pitch and tone, which can be detected by analysing the sound waves of a person’s voice.
  • Computer Algorithms – Recent advances in machine learning and artificial intelligence have led to the development of computer algorithms that can detect deception based on linguistic cues, such as the use of certain words or phrases.

While these techniques may show promise, they are still in the early stages of development and are not widely used in legal or investigative contexts anywhere in the world (despite what television may have you believe). As with polygraph testing, experts caution against relying solely on these techniques as the sole basis for making decisions about guilt or innocence and none of these methods are likely to be found reliable enough to play a part in trials any time soon. Just like profiling, they are likely to remain the stuff of dramas and expensive but fruitless government studies.