Detecting Concealed Information from a Mock Crime Scenario by Using Psychophysiological and Rt-Based Measures

ABSTRACT The Concealed Information Test has been widely used in polygraph testing in order to assess whether a suspect holds information that he/she does not want to reveal. Recently, behavioral measures based on reaction times (RTs) have also been shown to reliably reveal the concealed information effect in an oddball paradigm, thereby called an RT-based Concealed Information Test (RT-based CIT). The present study aimed to assess the detection efficiency of the two methods (i.e., the polygraph-based CIT vs. the RT-based CIT) in a mock crime paradigm, using psychophysiological measures from answers to questions in the polygraph test, and RTs to images in the RT-based test. The results confirmed the identification of concealed information effects using both assessment techniques and revealed that the two methods have similar accuracy rates for the detection of deception. KEYWORDS: deception detection, Concealed Information Test, polygraph, reaction times INTRODUCTION The polygraph (also known as lie detector) is an instrument commonly used in forensic practice in order to assess the truthfulness of a suspect. During a polygraph test, several physiological indices are measured and recorded: blood pressure, pulse rate, respiration, and skin conductance. The examinee is asked a series of questions and he/she responds to every question with yes or no. The traditional testing format is The Control Question Test (or Comparison Question Test - CQT; see Raskin & Honts, 2002, for a review), a test that contains three types of questions: relevant (directly related with the crime investigated), comparison (addressing general misconducts, related if possible with the type of crime under investigation) and irrelevant (focusing on neutral issues). This testing format has been widely criticized in the literature because it lacks theoretical foundation, it is insufficiently standardized, can be easily contaminated, is vulnerable to countermeasures and biased against innocent subjects (Ben-Shakhar & Furedy, 1990; Ben-Shakhar, 2002; Iacono & Lykken, 1999; Lykken, 1998). Nevertheless, the debate continues in the scientific community and the CQT is still widely used in forensic practice (Bus & Visu, 2004; Iacono, 2008). As an alternative, Lykken (1959, 1974) introduced a testing format named the Guilty Knowledge Test (later known as the Concealed Information Test - CIT, or the Concealed Knowledge Test - CKT). It is essentially a recognition test in which the subject is presented with several items, among which a critical or relevant item is embedded. The rationale is that the critical item will be recognized only by the "guilty" subjects, but not by the "innocent" subjects. An example of a question used in CIT could be: "The getaway car had a specific color. Was it: a) blue? b) green? c) white? d) red? e) black?", with white being the correct answer. It is absolutely necessary that the details of the crime are not known by the public; otherwise this testing format cannot be used. The CIT is aimed at disclosing the possession of information and is not meant to directly reflect deception. A repeated pattern of reacting differentially to the relevant items would suggest that the examinee has knowledge regarding the crime committed. Physiological responses assessed during CIT are the same as those recorded in the case of CQT: blood pressure, pulse rate, respiration and skin conductance. Among these, numerous laboratory studies showed that skin conductance is the most efficient, reaching the highest discrimination rate (Ben-Shakhar & Elaad, 2003). The laboratory studies have shown that CIT has accuracy rates from 76 to 88% for guilty subjects and from 83 to 99% for innocent subjects (Vrij, 2008). The few field studies published so far display a similar pattern of accuracy rates: 42-76% for guilty subjects and 94- 98% for innocent subjects (Vrij, 2008). It is also worth mentioning that in Japan, the CIT is used exclusively in forensic practice (Nakayama, 2002). …