Detection of deception: Event‐related potential markers of attention and cognitive control during intentional false responses. (11th December 2017)

• Record Type: Journal Article
• Authors: Gibbons, Henning; Schnuerch, Robert; Wittinghofer, Christina; Armbrecht, Anne‐Simone; Stahl, Jutta
• Is Part Of: Psychophysiology. Volume 55:Number 6(2018)
• Page Start: n/a
• Page End: n/a
• Publication Date: 2017-12-11
• Subjects: cognitive control -- detection of deception -- ERPs -- instructed deception -- medial‐frontal negativity -- P3a -- P3b; Psychophysiology -- Periodicals; 612.8
• DOI: 10.1111/psyp.13047
• Languages: English
• ISSNs: 0048-5772

Abstract

Abstract: Successful deception requires the coordination of multiple mental processes, such as attention, conflict monitoring, and the regulation of emotion. We employed a simple classification task, assessing ERPs to further investigate the attentional and cognitive control components of (instructed) deception. In Experiment 1, 20 participants repeatedly categorized visually presented names of five animals and five plants. Prior to the experiment, however, each participant covertly selected one animal and one plant for deliberate misclassification. For these critical items, we observed significantly increased response times (RTs), error rates, and amplitudes of three ERP components: anterior P3a indicating the processing of task relevance, medial‐frontal negativity reflecting conflict monitoring, and posterior P3b indicating sustained visual attention. In a blind identification of the individual critical words based on a priori defined criteria, an algorithm using two behavioral and two ERP measures combined showed a sensitivity of 0.73 and a specificity of 0.95, thus performing far above chance (0.2/0.2). Experiment 2 used five clothing and five furniture names and successfully replicated the findings of Experiment 1 in 25 new participants. For detection of the critical words, the algorithm from Experiment 1 was reused with only slight adjustments of the ERP time windows. This resulted in a very high detection performance (sensitivity 0.88, specificity 0.94) and significantly outperformed an algorithm based on RT alone. Thus, at least under controlled laboratory conditions, a highly accurate detection of instructed lies via the attentional and cognitive control components is feasible, and benefits strongly from combined behavioral and ERP measures.