2Over the last decade, Europe has seen a marked increase in the use of the polygraph for the detection of deception. Belgium and Finland, nowadays regularly use polygraph tests in criminal investigations, and the United Kingdom and the Netherlands have adopted its use in the treatment and monitoring of sex offenders (Meijer & van Koppen, 2008). Still the use of the polygraph for the detection of deception has been debated in the scientific literature for ages. In this contribution we highlight the promises and perils of the use of the polygraph for the detection of deception.
The polygraphThe words 'polygraph' and 'lie detector' are often used synonymously.Technically, the term 'polygraph' refers only to the recording device that is used for registering different physiological parameters. Polygraphs that are specifically designed for lie detection tests used to be briefcase-sized machines that registered physiological signals from the subject by recording them with multiple pens on a lengthy roll of paper.Nowadays, a small amplifier/digitizer and a laptop recording the signals are used in most cases. The sensors attached to the subject are generally (1) two expendable bands positioned around the thorax and the abdomen measuring respiration, (2) two electrodes attached to the inside of the hand measuring electrodermal activity and (3) an inflatable cuff positioned around the upper arm registering blood pressure. These physiological parameters co-vary with a number of psychological processes, including attention and emotion. Consequently, inferences about psychological processes can be made from polygraph recordings. And although it is widely accepted that there is no unique physiological pattern associated with lying, polygraph recording can still be used to infer guilt or innocence.
Detection of deception with the polygraphThe most widely used polygraph technique is known as the Control Question Test (CQT; Reid, 1947). In this type of test, the suspect answers several questions while connected to the polygraph. Among these questions are relevant and control questions. The relevant questions specifically deal with the incident under investigation, for example "on the 25th of March, did you hold up the post office in Chorley?" The control questions have a more generic nature, but also deal with undesirable behavior, for example "In the first 25 years of your life, have you ever done anything illegal?" It is reasoned that the most threatening questions will elicit the strongest physiological responses. The crucial assumption is that the relevant questions are most threatening for guilty suspects, whereas the control questions form the biggest threat to innocent suspects. This latter assumption requires some explanation. A CQT is preceded by a lengthy interview that serves to convince the suspect that the polygraph can determine to an extremely high degree of accuracy determine whether the suspect is lying or not. This means that an innocent examinee can confidently and honestly answer "no" to the relevant questions...