In response to petitioning by civic groups, the Control Yuan has agreed to review polygraph testing — commonly known as a lie detector test — which is often taken as proof of guilt in criminal judgements. This news highlights the suspicion that polygraphs, which the judiciary regards as highly dependable, might be pseudoscience.
The precondition for the test to be effective is whether lying can cause specific physiological reactions that are beyond a person’s control, and that these reactions can be measured using special equipment and the results interpreted by experts. However, while this premise may seem scientific, it has always been in doubt.
Those who question the effectiveness of polygraph testing do so because they doubt its objectivity. Questions of particular concern are whether the examiner has sufficient expertise, whether the equipment used is working effectively, whether the testing environment is normal and so on. All these factors can influence the outcome of the test. Furthermore, given that each person is physiologically different, how can there be consistent criteria for judgement? By comparison, other forensic techniques, such as DNA profiling, do not involve the aforementioned variables, and are more accurate and can be verified by other experts. Relatively speaking, polygraph tests are subject to many forms of interference and lack a feature that is very important in science: replicability. This means that results cannot be checked for accuracy because it cannot be repeated and this is the Achilles heel of lie detection tests.
Although the criminal justice system does not exclude polygraph testing as evidence, it does require examiners to have expert training and experience. It requires the instruments used to be of good quality and functioning normally. For a polygraph test to be accepted, the test environment must be free of interference and it can only be done when the examinee is in a normal physical and mental state and fully conscious.
To ensure that examinees are in the required state, examiners should tell them that they have the right to refuse testing and inform them about the possible effects and consequences of the test. Test subjects should first be medically examined to ensure they are in a fit mental and physical state to be tested. The fact that there are so many requirements exposes the concern that lie detector tests carried out under duress violate a suspect’s right to defense, and that the results are likely to be distorted.
Although judicial practice has strict requirements around polygraph results, in reality, these requirements may not be sufficient, because when the test environment is controlled by the interrogator and the examinee is in a state of isolation, it is doubtful whether the examinee can be in a normal physical and mental state even if they are innocent. Furthermore, lie detector tests are usually done when prosecutors think the accused is not telling the truth. Examiners might be prejudiced and it is hard to be sure that they are truly objective. Even if a suspect does not confess, these factors will make it hard not to fail the test. Such results are therefore a quasi-confession, which seriously violates a person’s right to non-self-incrimination.
In 1997 a soldier named Chiang Kuo-ching (江國慶) was executed after being wrongfully convicted of killing a five-year-old girl. Not only was he forced to confess under torture, but investigators treated his failure of a lie detector test as iron-clad proof of his guilt. This false result, plus a lot of other bogus evidence, meant an innocent man was executed. This is precisely why all so-called scientific forms of evidence, including polygraphs, need to be re-examined.