Aberrant moral judgment in extremist terrorists

BAEZ, SANDRA; EDUAR HERRERA; NATALIA TRUJILLO; FACUNDO MANES; AGUSTÍN IBÁÑEZ

Congreso; 18th IOP World Congress; 2016

Terrorism jeopardizes its victims? freedom and physical integrity while compromising governmental stability. This has become clear in Colombia, where terrorist violence by paramilitary groups has shockingly escalated in the last decades. An examination of the social-cognitive profiles characterizing terrorist groups is urgently needed. We assessed the moral judgments and social-cognitive profiles of 66 ex-combatants from a paramilitary terrorist group, relative to a control group matched in age, gender, years of education, and verbal and fluid intellectual capacities. Demographic, neuropsychological, and experimental data were compared between groups with ANOVA and Tukey?s HSD post-hoc tests. We conducted a multiple regression analysis to explore whether moral judgment was associated with other domains. Moreover, receiver-operating char- acteristic (ROC) curves were calculated to test (a) whether moral judgment or any of the other domains successfully discriminates between terrorists and controls, and (b) which of these domains yields the best discrimination accuracy. In addition, we tested if moral judgment offers an adequate group classification by implementing a support vector machine (SVM) to classify terrorists and controls. Results showed a significant three-way interaction among intention, outcome, and group (F (1, 130) = 9.62, p b 0.01, η2 = 0.09). A post-hoc analysis revealed that terrorists judged accidental harm as less permissible (p b 0.01) and attempted harm as more permissible than controls (p b 0.01). A multiple regression model (F (7, 124) = 23.45, p b 0.01) showed that emotion recogni- tion (beta = -0.25) and group (beta = -0.60) predicted moral judgment, explaining 57% of the variance. Fluid intelligence, executive functions, and frequency of aggressive behaviors did not predict moral judgment performance. In addition, moral judgment was the measure that best discriminated between groups, even when compared with other cognitive-affective variables in which terrorists exhibited atypical or impaired performance (AUC: 0.91, CI: 0.85 -0.96; p b 0.01; specificity = 0.86; sensitivity = 0.84). Other measures did not accurately discriminate terrorists from controls. The SVM model achieved an average classification accuracy of 72%, a sensitivity of 0.67, and a specificity of 0.79. However, a statistical comparison between the area under the ROC curves for the global moral score and the SVM model revealed that the former, by itself, showed better discrimination accuracy (z = 2.69, p = 0.007). This approach to understanding terrorists? cognitive profiles has impor- tant legal and forensic implications.