Observers’ Real-Time Sensitivity to Deception in Naturalistic Interviews

This study tested the ability of experienced interviewers and novice observers to detect deception while watching mock interviews featuring experimental or control questioning methods and different detainee languages. The protocol featured a complex, realistic critical event and naturalistic interviews in which mock detainees could report unconstrained. Experimenters recorded these interviews and presented them to observers who judged veracity in real time. In general, experienced interviewers were no more sensitive to deception than were novices and both groups set conservative response criteria. Observers were more sensitive to deception when viewing control versus experimental questioning methods. Observers were more sensitive to deception when viewing Arabic speakers interviewed through an interpreter. Results imply that not all trained interviewers exhibit a lie bias; additional research should examine how best to transition lab-tested interview methods into the field, and language and interpreter factors may impact the ability to assess veracity in multiple ways.