Careful calculation or a leap of faith? A field study of the translation of CBCA ratings to final credibility judgements

This field experiment investigated the influence of Criteria‐Based Content Analysis (CBCA) ratings on ultimate decision accuracy regarding the credibility of children's statements of sexual abuse. Following a selection procedure, based on case facts independent of statement quality, 21 truthful accounts and 10 fabricated accounts of 6‐ to 17‐year olds were analysed. Two experts rated the presence of the CBCA criteria and made overall credibility judgements for each statement. Rater one achieved an overall hit rate of 84% (95% for truthful statements and 60% for fabricated statements) and rater two a hit rate of 81% (81% for both truthful and fabricated statements) but the raters did not always agree. The CBCA criteria appeared more often in the truthful statements compared to the fabricated statements. Additional factors that influenced raters' credibility judgements, besides CBCA scores, are discussed. Copyright © 2010 John Wiley & Sons, Ltd.     

Using an Example Statement Increases Information but Does Not Increase Accuracy of CBCA,RM, and SCAN

Verbal credibility assessment methods are frequently used in the criminal justice system to investigate the truthfulness of statements. Three of these methods are Criteria Based Content Analysis (CBCA), Reality Monitoring (RM), and Scientific Content Analysis (SCAN). The aim of this study is twofold. First, we investigated the diagnostic accuracy of CBCA, RM, and especially SCAN. Second, we tested whether giving the interviewee an example of a detailed statement can enhance the diagnostic accuracy of these verbal credibility methods. To test the latter, two groups of participants were requested to write down one true and one fabricated statement about a negative event. Prior to this request, one group received a detailed example statement, whereas the other group received no additional information. Results showed that CBCA and RM scores differed between true and fabricated statements, whereas SCAN scores did not. Giving a detailed example statement did not lead to better discrimination between truth tellers and liars for any of the methods but did lead to the participants producing significantly longer statements. The implications of these findings are discussed. Copyright © 2013 John Wiley & Sons, Ltd.     

Glaubhaftigkeitsbegutachtung – mehr als Merkmalsorientierte Inhaltsanalyse

The diagnostic value of criteria-based content analyses (CBCA) for assessing statement credibility has been widely acknowledged. However, theoretical considerations and empirical results show that CBCA is basically appropriate for discrimination between fabricated and truthful accounts. CBCA does not differentiate between truthful and suggested statements and can therefore not help to decide upon a suggestion hypothesis. Moreover truthful accounts do not always contain many CBCA criteria. It is argued that CBCA is for these reasons not a method for substantiating a truthful statement, but a method for rejecting a fabrication hypothesis As a whole Statement validity analysis is the process of systematically generating and evaluating hypotheses about potential origins of the given statement. Only in a subgroup of cases CBCA constitutes the major part of this diagnostic procedure.     

Criteria‐based content analysis of true and suggested accounts of events

Worldwide, the criteria‐based content analysis (CBCA) is probably the most widely used veracity assessment technique for discriminating between accounts of true and fabricated events. In this study, two experiments examined the effectiveness of the CBCA for discriminating between accounts of true events and suggested events believed to be true. In Experiment 1, CBCA‐trained judges evaluated participants' accounts of true and suggestively planted childhood events. In Experiment 2, judges analysed accounts of recent events that were experimentally manipulated to be a (a) true experience, (b) false experience believed to be true and (c) deliberately fabricated experience. In both experiments CBCA scores were significantly higher for accounts of true events than suggested events. However, this difference was not significant for participants classified as experiencing ‘full’ memories for the suggested event. Self‐report memory measures supported the findings of the CBCA analyses. Taken together these results suggest that the CBCA discriminative power is greatly constrained. Copyright © 2008 John Wiley & Sons, Ltd.     

Criteria Based Content Analysis: The Less Scientific Road to Truth

Criteria Based Content Analysis (CBCA) is a forensic tool that aims atdistinguishing true from false statements made by crime victims. As such,CBCA and similar techniques are regularly employed by psychologicalexpert-witnesses in court. However, it could be argued that the diagnosticaccuracy of CBCA is moderate, and too low for application in the forensiccontext. The current article discusses problems that arise if conclusionsabout credibility of testimonies based on CBCA are presented in court. Afundamental, yet so far hardly addressed flaw is that CBCA suffers from aninherent truth bias. That is, the underlying notion that credibility isexamined by means of searching for signs supporting truthfulness isscientifically less valid than the approach in which credibility isestablished by focusing on alternative explanations and signs of deceit.     

Mock juror's perceptions of a child witness passing or failing a truth and lies discussion or promising to tell the truth

This study examined the effect of a child passing or failing the UK truth and lies discussion (TLD) compared with the Canadian promise to tell the truth on mock jurors' decisions regarding witness credibility and truthfulness and defendant guilt. Ninety-two participants read a vignette that described a child witnessing his father physically attacking his mother. The vignette was manipulated for witness age (age 4 years and age 8 years) and TLD performance/promise. Supporting the hypotheses, participants rated the witness's credibility and truthfulness significantly higher after a witness passed a TLD and after promising to tell the truth. The age of the child witness did not significantly affect jurors' decision making. The results are discussed in relation to arguments regarding the abolition of the UK's TLD in favour of introducing a promise to tell the truth.     

Judgment of the credibility of rape allegations: a content analytic field experiment]

This study explores the issue of whether allegations of rape can be correctly identified as true or false on the basis of a set of content criteria used successfully in previous studies with respect to children's accounts of sexual abuse. A sample of 30 authentic protocols of police interrogations of women reporting rape were selected for the study. Fifteen cases were classified as false accounts because the complaints eventually confessed to having fabricated their allegations. The remaining cases were classified as true rapes on the basis of medical and other corroborating evidence. A group of experienced police officers (N = 30) judged these cases in terms of their credibility on the basis of the 19 content criteria suggested by Steller & K?hnken (1989) and gave an overall judgment of the true vs. false nature of each statement. A second group of police officers (N = 22) judged the credibility of the statements without reference to these criteria. The findings from a series of discriminant function analyses showed that the set of criteria as a whole is successful in distinguishing between true and false rape allegations. The results of the stepwise discriminant function analyses reveal which criteria in particular contribute to the correct identification of true and false cases. However, the higher percentage of correct classifications in the group using the content criteria was only marginally significant compared to the second group judging the cases without reference to the criteria. Altogether, the findings of this study support the basic tenet of statement analysis that the linguistic features of a statement contain essential clues for assessing its credibility.     

Statement Validity Analysis

Summary

All information regarding 103 Dutch sexual abuse cases was analyzed using the Validity Checklist, the second part of the Statement Validity Analysis procedure. The fulfillment of the VC items was compared for four different age groups. Only two VC items differed significantly between the age groups. The systematic summarization of content analysis and validity checks revealed only a weak relationship between the fulfillment of the CBCA criteria and the VC items together and the outcome of the cases. The results suggest that SVA cannot yet be used as a scientifically validated instrument for judging the truthfulness of allegations of child sexual abuse.     

Children report suggested events even when interviewed in a non-suggestive manner: what are its implications for credibility assessment?

An important question in the legal context is how suggested statements about fictitious events develop over a course of various interviews. Sixty-seven first-graders were interviewed four times about one real and one fictitious event, applying various suggestive techniques. A fifth, non-suggestive, interview was conducted by blind experts. Over the course of the interviews there was a considerable increase in assents to the fictitious events. Moreover, few significant differences in criteria-based content analysis-criteria were revealed between true and fictitious statements in the fifth interview. Accordingly, experts had difficulties in discriminating between true and suggested accounts. Furthermore, several children were convinced of the truth of their reports, even after having been partially debriefed in a sixth interview. Implications for statement validity assessment are discussed. Copyright © 2004 John Wiley & Sons, Ltd.     

Fantasy proneness as a confounder of verbal lie detection tools

This study investigated whether high fantasy‐prone individuals have superior storytelling abilities. It also explored whether this trait is related to specific linguistic features (i.e. self‐references, cognitive complexity, and emotional words). Participants high (n = 30) and low (n = 30) on a fantasy proneness scale were instructed to write down a true and a fabricated story about an aversive situation in which they had been the victim. Stories were then examined using two verbal lie detection approaches: criteria‐based content analysis (CBCA) and linguistic inquiry and word count (LIWC). Irrespective of the truth status of the stories, independent observers rated stories of high fantasy‐prone individuals as being richer in all nine CBCA elements than those of low fantasy‐prone individuals. Furthermore, overall, high fantasy‐prone people used more self‐references in their stories compared with low fantasy‐prone individuals. High fantasy prones' fabricated stories scored higher on various truth indices than authentic stories of low fantasy prones. Thus, high fantasy‐prone people are good in creating a sense of authenticity, even when they fabricate stories. Forensic experts should bear this in mind when they employ verbal lie detection tools. Copyright © 2010 John Wiley & Sons, Ltd.     

Effects of the putative confession instruction on perceptions of children's true and false statements

The putative confession (PC) instruction (i.e., “[suspect] told me everything that happened and wants you to tell the truth”) during forensic interviews with children has been shown to increase the accuracy of children's statements, but it is unclear whether adults' perceptions are sensitive to this salutary effect. The present study examined how adults perceive children's true and false responses to the PC instruction. Participants (n = 299) watched videotaped interviews of children and rated the child's credibility and the truthfulness of his/her statements. When viewing children's responses to the PC instruction, true and false statements were rated as equally credible, and there was a decrease in accuracy for identifying false denials as lies. These findings suggest that participants viewed the PC instruction as truth‐inducing. Implications for the forensic use of the PC instruction are discussed.     

A Case Study of Witness Consistency and Memory Recovery Across Multiple Investigative Interviews

Access to audio recordings of five interviews (Interviews 2–6), and to the interviewer's contemporaneous notes during an initial unrecorded interview, made it possible to assess consistency across repeated attempts by a 9‐year‐old to describe her older sister's abduction from their shared bedroom. Information provided in each of the interviews was systematically analysed to determine whether each unit of information was new, consistent (repeated) or contradictory in relation to earlier reported information and whether any informative detail provided in the witness' initial interview was subsequently omitted. In addition, the witness' accounts were compared with details provided by the victim upon her rescue. This case analysis is particularly informative in light of widespread professional concerns about the effects of repeated interviewing on the quality and accuracy of children's accounts of experienced events. Copyright © 2011 John Wiley & Sons, Ltd.     

Assessing children's statements: the impact of a repeated experience on CBCA and RM ratings

This study examined the extent to which the Criteria‐based Content Analysis (CBCA) technique and the Reality Monitoring (RM) technique were affected by the number of times children had experienced or imagined an event. Children (age 10–13, N =87 ) participated in an experiment where half the sample experienced a health examination (either one or four times), and the other half imagined (either one or four times) that they took part in a health examination. One week after the final occasion, the children were interviewed. The results showed that RM was sensitive to both the authenticity of the statements (increased presence of the criteria for real events) and whether the event had been repeatedly experienced/imagined (increased presence of the criteria for the repeated actions). The CBCA did not successfully distinguish the real from the imagined. Copyright © 2004 John Wiley & Sons, Ltd.     

Differentiating accounts of actual,suggested and fabricated childhood events using the judgment of memory characteristics questionnaire

Statement analysis procedures are used in forensic settings to classify reported events as experienced or non‐experienced. These procedures are typically validated using accounts of actual events and intentionally fabricated events. However, people can also unintentionally develop false memories. To examine whether inclusion of accounts of suggested events affects classification accuracy, we validated the judgment of memory characteristics questionnaire (JMCQ) statement analysis procedure using all three statement types. Participants attempted to recall two actual events and one suggested event from their childhood over two cognitive interviews, then intentionally fabricated an account of another childhood event. Fourteen of the 34 participants (41%) reported having experienced the suggested event. Independent raters then used the JMCQ to analyse and classify each type of statement from this participant subset. Inclusion of accounts of suggested events did not reduce classification accuracy. Raters tended to classify accounts of both fabricated and suggested events as non‐experienced. Copyright © 2010 John Wiley & Sons, Ltd.     

The Timing of the Defense Opening Statement: Don't Wait Until the Evidence Is in1

Defense attorneys often reserve their opening statements until after presentation of evidence by the prosecution, a practice advocated by some experts. The current study, involving 291 subjects, varied the timing of the defense attorney's opening statement so that it preceded the prosecutor's opening statement, immediately followed the prosecutor's opening statement, or was reserved until after the prosecutor's case presentation. The type of opening statement (content vs. noncontent) and the testimony (altered vs. unaltered) were varied for purposes of increasing generalization. Materials were based on an actual case of auto theft and were presented in written form complete with judge's instructions. The results showed that subjects perceived eyewitness testimony, the prosecutor's opening statement, the prosecutor's closing statement, the defense attorney's closing statement, and the effectiveness of the attorneys differently depending on the timing of the opening statement. Each of these items favored the defense more than the prosecution if the defense opening statement was earlier rather than later. These effects did not interact with type of opening statement or the testimony alteration variable. Individual verdicts, when weighted by the subjects' confidence in their verdicts, were also affected by the timing variable with verdicts more favorable to the defense when the defense opening statement was given earlier rather than later. The consistency with which the timing variable affected subjects' impressions suggests that defense attorneys who take their first opportunity to make an opening statement, rather than delay, end up with a stronger case for their client. Possible exceptions to this conclusion are discussed.     

Embedding lies into truthful stories does not affect their quality

When given the opportunity, liars will embed their lies into otherwise truthful statements. In what way this embedding affects the quality of lies, however, remains largely unknown. This study investigated whether lies that are embedded into truthful stories are richer in detail and contain higher quality details compared to lies that are part of entirely fabricated statements. Participants (N = 111) were asked to provide a statement that was either entirely truthful, entirely fabricated, or had the fabricated element of interest embedded into an otherwise truthful story. Results indicated that lies embedded in a fabricated statement are not qualitatively different from lies embedded in an otherwise truthful statement. Supporting Bayes factors provided moderate to strong evidence for this conclusion. Accordingly, verbal credibility assessment tools based on the verbal content measured in this study may be robust against the embedding of lies.     

Täuschungsstrategien von Kindern, Jugendlichen und Erwachsenen

Criteria-Based Content Analysis (22) is part of the Statement Validity Assessment (SVA), a method used for assessing the credibility of witness’ statements. CBCA relies on two assumptions, one referring to the cognitive strain of inventing a complex false declaration, the other being related to self presentation as a competent, reliable and truthful person by avoiding potentially “damaging” contents. When evalutating statements, forensic experts take into account that cognitive capacity - and thus potential content quality of a false declaration - increases with age. On the other hand there is no empirical evidence for an improvement of deceptive self-presentation strategies so far. Recent studies on content-related deceptive strategies with children and adults support the assumption that children as well as adults do apply specific deception strategies. Due to methodological differences, however, a direct comparison between these studies to obtain results on possible age effects has not been possible. The aim of this study was to investigate whether children, adolescents and adults differ in the content-related deceptive strategies they apply. Children (n = 48), adolescents (n = 73) and adults (n = 50) were given an age appropriate standardized questionnaire about their content-related deception strategies. Results show age differences in the strategic value assigned to certain content characteristics.