Candel, I., Merckelbach, H., Loyen, S., & Reyskens, H. (2005). "I hit the Shift-key and then the computer crashed": Children and false admissions. Personality and Individual Differences, 38, 1381-1387.
Although a confession is generally viewed as irrefutable evidence of one's guilt, research into the phenomenon of false confessions has revealed that, under a variety of circumstances, a questioner may elicit a false confession from a suspect. To date, the literature in this area has identified three specific types of false confessions. A voluntary false confession is one that is made of one's own accord, without any external pressure to do so. A coerced-compliant false confession, on the other hand, is when a suspect has been heavily pressured, often throughout a lengthy interrogation process, to confess guilt. In this case, although the suspect still believes in their innocence, he or she may confess as a way of avoiding other potentially negative consequences (such as a longer jail sentence should she or he be found guilty at trial) or to gain a reward. The third type of false confession, and one which has received much attention in recent years, is the coercedinternalised false confession. According to researchers, in this condition suspects may not only confess to crimes that they did not commit, but may come to believe that they actually did commit the crimes that they have confessed to.
During the past decade, researchers have begun to make a concentrated attempt to identify the internal and external factors that increase the likelihood of obtaining a false, internalised confession. Interview techniques such as isolating the suspect, sitting close to the suspect while questioning them, and conducting lengthy interviews, have been found to increase the likelihood of eliciting a false confession. Researchers have been particularly interested in the personality characteristics of the suspect that may contribute to the likelihood of making a false confession. One factor that has received widespread attention is suggestibility, defined as the extent to which people are likely to accept messages communicated to them through questioning. This factor has been identified as one that may play a significant role in eliciting false confessions from children.
In an effort to examine the role of suggestibility in eliciting false, internalised confessions from children, Candel, Merckelbach, Loyen, and Reyskens (2005) recruited 50 children aged six to nine completed a brief interview-style questionnaire and computer task. The goal of this study was to determine the extent to which children will falsely confess to causing a computer malfunction when accused by a researcher, and whether this relates to their level of suggestibility. Similar studies have been conducted with adult and adolescent participants, with the main difference being that older participants were presented with false evidence in addition to the accusation. In this study, the children were falsely accused, but no false evidence was presented. In order to accomplish this goal, each child attended two sessions with the researcher and answered a series of questions designed to measure suggestibility and to complete a computer task, during which the computer supposedly "crashed".
Specifically, during the first session, each child completed the Dutch version of the Bonn Test of Statement Suggestibility (BTSS-NL). This measure consists of a brief narrative accompanied by four pictures, a ten-minute filler task (drawing a picture), and a set of 27 questions of three types: questions containing leading or misleading information, which comprise the Yield scale; questions that are immediately repeated after the participant has given an answer, implying that the initial response was incorrect (the Shift scale); and simple memory questions included in order to disguise the nature of the task. Suggestibility was measured by the participants' responses. …