Thermal Imaging as a Lie Detection Tool at Airports

We tested the accuracy of thermal imaging as a lie detection tool in airport screening. Fifty-one passengers in an international airport departure hall told the truth or lied about their forthcoming trip in an interview. Their skin temperature was recorded via a thermal imaging camera. Liars’ skin temperature rose significantly during the interview, whereas truth tellers’ skin temperature remained constant. On the basis of these different patterns, 64% of truth tellers and 69% of liars were classified correctly. The interviewers made veracity judgements independently from the thermal recordings. The interviewers outperformed the thermal recordings and classified 72% of truth tellers and 77% of liars correctly. Accuracy rates based on the combination of thermal imaging scores and interviewers’ judgements were the same as accuracy rates based on interviewers’ judgements alone. Implications of the findings for the suitability of thermal imaging as a lie detection tool in airports are discussed.     

Field reliability of the SAVRY with juvenile probation officers: Implications for training

Two complimentary studies were conducted to investigate the inter-rater reliability and performance of juvenile justice personnel when conducting the Structured Assessment of Violence Risk for Youth (SAVRY). Study 1 reports the performance on four standardized vignettes of 408 juvenile probation officers (JPOs) and social workers rating the SAVRY as part of their training. JPOs had high agreement with the expert consensus on the SAVRY rating of overall risk and total scores, but those trained by a peer master trainer outperformed those trained by an expert. Study 2 examined the field reliability of the SAVRY on 80 young offender cases rated by a JPO and a trained research assistant. In the field, intra-class correlation coefficients were 'excellent' for SAVRY total and most domain scores, and were 'good' for overall risk ratings. Results suggest that the SAVRY and structured professional judgment can be used reliably in the field by juvenile justice personnel and is comparable to reliability indices reported in more lab-like research studies; however, replication is essential.     

Interpersonal Relationships and the Development of Behavior Problems in Adolescents in Urban Schools: A Longitudinal Study

The purpose of this study was to investigate the trajectories of behavioral problems for a sample of predominately minority adolescents (n = 212, 91% African-American and/or Hispanic, 45% boys, 55% girls) in a large, urban school district and to determine the impact of parental and peer relationships, gender, and risk status on their development during middle and high school. Multi-level growth modeling was the primary statistical procedure used to track internalizing and externalizing behavioral problems across time. Results indicated that behavioral problems as rated by students’ teachers declined significantly for both boys and girls, a finding that is in direct contrast to previous studies of adolescent behavior. The quality of parental relationships was a strong predictor of both types of behavior whereas the quality of peer relationships predicted only internalizing behavioral symptoms. These findings suggest that behavioral trajectories may be somewhat unique for this population underscoring the need for additional research in this area. The findings also have implications for intervening with children and youth who display behavioral problems during critical developmental periods.     

An Empirical Test of the Behaviour Analysis Interview

The present experiment is the first empirical test of the Behaviour Analysis Interview (BAI), an interview technique developed by F. E. Inbau, J. E. Reid, J. P. Buckley, & B. C. Jayne (2001) designed to evoke different verbal and non-verbal responses from liars and truth-tellers. Inbau et al. expect liars to be less helpful than truth-tellers in investigations and to exhibit more nervous behaviours. Just the opposite predictions, however, follow from the deception literature, which notes that liars take their credibility less for granted and are therefore more aware of their responses and their impact on others. This suggests that liars' answers should be more helpful than truth-tellers' answers, and liars' non-verbal responses should appear more relaxed than truth-tellers' non-verbal responses. In the present experiment, 40 participants (undergraduate students) lied or told the truth about an event during a BAI interview. The interviews were coded according to Inbau et al.'s guidelines. The results showed that, compared to liars, truth-tellers (a) were more naive and evasive when explaining the purpose of the interview, and (b) were less likely to name someone who they felt certain did not commit the crime. Truth-tellers also exhibited more nervous behaviours. The results were consistent with the predictions of the deception literature, and directly opposed to the predictions of BAI.