Publicaciones - NeuroMente

contador de visitas com
Nuestras Visitas
Vaya al Contenido

Menu Principal:

Outcome-oriented moral evaluation in terrorists

Assessment of moral judgements and social-cognitive profiles of Colombian paramilitary terrorists  by Baez et al. reveals a moral code abnormally guided by outcomes, rather than the integration of intentions and outcomes.


Sandra Baez, Eduar Herrera, Adolfo M. García, Facundo Manes, Liane Young & Agustín Ibáñez
Nature Human Behaviour 1, Article number: 0118 (2017)


As  shown by the Office of the United Nations High Commissioner for Human  Rights, terrorism is one of the most pernicious threats to contemporary  societies1.  In addition to obliterating the freedom and physical integrity of  victims, terrorist practices can destabilize governments, undermine  civil harmony and threaten economic development1.  This is tragically corroborated by the recent history of Colombia, a  country marked by escalations of paramilitary terrorist violence2.  Although multiple disciplines are struggling to understand these  atrocities, the contributions from cognitive science have been limited.  Social cognition abilities3,4,5,6,7  have been proposed as important variables in relation to criminal and  violent profiles. Against this background, this study aimed to assess  the moral judgements and social-cognitive profiles of 66 ex-combatants  from a paramilitary terrorist group. We found that moral judgement in  terrorists is abnormally guided by outcomes rather than by the  integration of intentions and outcomes. This pattern was partially  related to emotion recognition and proactive aggression scores but  independent from other cognitive domains. In addition, moral judgement  was the measure that best discriminated between terrorists and  non-criminals.
With extreme violence escalating for more than 60 years, Colombia features one of the greatest insurgency rates in the world8.  In particular, terrorism has become the main political and economic  tool of paramilitary groups—illegal right-wing armed organizations first  formed by state and landowners in response to guerrilla movements2. Their violent practices have grown so steeply that Colombia now has one of the highest levels of terrorism worldwide8. International Amnesty9  estimates that, in the past two decades, at least 70,000 people have  been killed by terrorists in this country. Thousands more have been  victims of enforced disappearance, kidnapping and torture, mostly at the  hands of paramilitary groups2. Paradoxically, the terrorists typically justify their actions in terms of moral imperatives10,11.  Indeed, they have invoked the need for ‘social cleansing’, killing  thousands of drug addicts, criminals, prostitutes, homosexuals and  homeless individuals as part of a ‘moral purification’ campaign12. Thus, this population constitutes an invaluable model to assess the links between violent experiences and moral cognition.
Within a given social group, moral norms emerge from conventionally accepted values that guide adaptive behaviour13.  In civilized social niches, individuals tend to attach greater  importance to intentions than to outcomes in judging the morality of an  action. Actions aiming to induce harm, regardless of their success, are  typically deemed less morally permissible than those in which harm was  neither intended nor inflicted, or merely accidental14,15.  Although no study has assessed moral cognition in extreme terrorists,  recent evidence shows that criminal psychopaths feature abnormally  utilitarian personal moral judgements16 and judge accidental harm as more permissible than non-psychopaths17. Moreover, when faced with moral dilemmas, psychopaths show reduced activity in brain regions associated to moral judgement18.  By the same token, extreme terrorists could be distinctively  characterized by deviant forms of moral cognition, arguably shaped by  their particular cultural milieus. Specifically, if terrorists deem it  morally appropriate to do whatever it takes in the pursuit of an aim,  their moral judgements may be critically rooted in the success of an  action rather than the probity of its underlying intention.
To  address this issue, we evaluated 66 incarcerated members of an illegal  armed paramilitary group, designated as a terrorist organization by  multiple countries and organizations. All of them were convicted of  murder, with a mean of 33 victims per subject. We also assessed 66  sociodemographically matched controls (non-criminals). Participants  performed a well-characterized moral judgement task14,15  (see Methods section) that disentangles the contributions of intentions  and outcomes to moral judgement. The task included two conditions in  which intentions and outcomes matched (‘no harm intended or inflicted’  and ‘successfully attempted harm’) and two in which these variables  mismatched (‘unsuccessfully attempted harm’ and ‘accidental harm’) (see Fig. 1).  Additional tasks were administered to assess relevant  cognitive-affective domains (intellectual level, executive functions,  aggressive behaviours and emotion recognition). Between-group  comparisons were performed to determine the domains in which terrorists  exhibited atypical performance. Results showed no significant  differences in fluid intelligence, verbal intelligence quotient (IQ) or  executive functions. However, scores on the Motives for Aggression  Inventory (MAI) and the Situation and Aggressive Behavior Inventory  (SABI) revealed a higher frequency of aggressive behaviours in  terrorists than in non-criminals (Table 1).  Terrorists also showed higher levels of proactive aggression than  non-criminals as measured by the Reactive–Proactive Aggression  Questionnaire (RPQ). No differences between groups were observed in the  levels of reactive aggression (Table 1).  In addition, terrorists exhibited lower scores in emotion recognition,  with specific difficulties in detecting anger, sadness and disgust (Table 1).

The more important findings,  however, concerned moral judgement. A mixed-effects 2 (group: terrorist  versus non-criminals) × 2 (intention: neutral versus negative) × 2  (outcome: neutral versus negative) ANOVA revealed that both groups  judged actions with neutral intentions and neutral outcomes as more  permissible than actions with negative intentions and negative outcomes  (main effects of intention (F(1, 130) = 621.56, p < 0.01, η2 = 0.82) and outcome (F(1, 130) = 468.2, p < 0.01, η2 = 0.78)). Furthermore, accidental harm was judged as more permissible than intentional harm (intention × outcome interaction (F(1, 130) = 54.61, p < 0.01, η2 = 0.3)).
In addition, a significant three-way interaction was detected among intention, outcome and group (F(1, 130) = 9.62, p < 0.01, η2 = 0.09).  A post-hoc analysis (Tukey HSD, mean square error = 0.66,  d.f. = 235.70) revealed that terrorists judged accidental harm as less  permissible (p < 0.01) and attempted harm as more permissible (p < 0.01)  than non-criminals. Neither of these conditions was affected by  executive skills (accidental harm judgement: backward digit span (p = 0.16), verbal working memory (p = 0.51), abstraction capacity (p = 0.89); attempted harm judgement: backward digit span (p = 0.3), verbal working memory (p = 0.5), abstraction capacity (p = 0.17)). Also, no significant differences emerged in judgements of non-harm (p = 0.14) or successful attempt to harm (p = 0.46) (see Fig. 2a). The terrorists’ moral judgements for no harm (r = −0.014, p = 0.23), accidental harm (r = .016, p = 0.92), unsuccessfully attempted harm (r = −0.23, p = 0.17) or successfully attempted harm (r = 0.15, p = 0.21) were not significantly associated to the time spent in prison.

Figure 2: Significant  differences between groups, associations between moral judgement and  other relevant factors, and ROC curve analyses.
a, Moral judgements of terrorists (N = 66) and non-criminals (N = 66). Asterisks indicate significant differences (p < 0.05). Error bars represent standard deviations. b,  Regression analysis with global moral score as the dependent variable  and RPQ proactive aggression score as the significant predictor. c, Regression analysis with global moral score as the dependent variable and total TASIT score as the significant predictor. d,  ROC curves for the global moral score and the SVM model. The global  moral score yielded the best discrimination accuracy (area under the  curve (AUC): 0.91, CI: 0.85–0.96; p < 0.01), followed by the SVM model (AUC: 0.78, CI: 0.70–0.86; p < 0.01). e,  ROC curves for the attributes that did not show good discrimination  accuracies. Motives for aggressive behaviours (AUC: 0.74, CI: 0.66–0.83;  p < 0.01); RPQ proactive aggression score (AUC: 0.72, CI: 0.63–0.81; p < 0.01); frequency of aggressive behaviours (AUC: 0.69, CI: 0.60–0.78; p < 0.01); emotion recognition (AUC: 0.66, CI: 0.56–0.86; p < 0.01); executive functions (AUC: 0.54, CI: 0.44–0.64; p = 0.39); fluid intelligence (AUC: 0.54, CI: 0.44–0.64; p = 0.37).

In addition, to control for the  effect of aggression, fluid intelligence, executive functions and  emotion recognition on moral judgement, we reanalysed the data  considering the following covariates: Raven’s matrices, total scores  from the INECO Frontal Screening battery (IFS), MAI and The Awareness of  Social Inference Test (TASIT), and the SABI and RPQ subscales. Results  showed that the three-way interaction among intention, outcome and group  remained significant (F(1, 122) = 4.01, p < 0.05, η2 = 0.03).
As in previous reports19, paired-sample t-tests  were used to compare within-group performance on the moral conditions  in which terrorists differed from non-criminals. These comparisons  revealed that non-criminals judged accidental harm as more permissible  than attempted harm (t(65) = 15.52, p < 0.01). The opposite difference was observed in terrorists (t(65) = −6.27, p < 0.01), who judged attempted harm as more permissible than accidental harm.
To  establish how specific these patterns were to terrorists, as opposed to  other criminals, we administered the moral judgement task to a second  control group of incarcerated murderers with no terrorist background  (see Supplementary Information 1).  This second group was matched for years of imprisonment, age, sex and  education. Relative to both this sample and the first control group,  terrorists judged accidental harm as less permissible and attempted harm  as more permissible. No significant differences were observed between  the two control groups in any condition (see Supplementary Information 2 and Supplementary Fig. 1).
Considering  the marked distortions observed in terrorists’ moral cognition, we  conducted a multiple regression analysis to explore whether moral  judgement was associated with performance in the other relevant domains.  For this analysis, we estimated their overall moral judgement profile  by calculating a global moral score. This score was represented by the  average of the difference between raw scores for accidental and  attempted harm and the maximum expected rating for each condition (7 and  1, respectively). Thus, we subtracted the accidental harm score from 7  and the attempted harm score from 1, and then we averaged both results.  The higher this global score, the worse the sample’s performance. We  estimated a model in which the above global score was considered as the  dependent variable, while group, Raven’s matrices, IFS, MAI and TASIT  total scores, and SABI and RPQ subscales were the predictors. This model  (F(9, 122) = 21.67, p < 0.01) showed that RPQ  proactive aggression score (beta = 0.23), total TASIT score  (beta = −0.24) and group (beta = −0.50) predicted moral judgement,  explaining 58% of the variance (Fig. 2b,c).  Fluid intelligence, executive functions, reactive aggression and  frequency of aggressive behaviours did not predict moral judgement  performance. Standardized coefficients and significance levels are shown  in Table 2.

In addition, a logistic regression  was conducted to determine domains associated with group membership.  This model included group as dependent variable and Raven’s matrices,  moral global score, IFS, MAI and TASIT total score, and SABI and RPQ  subscales as covariates. The model showed that moral judgement global  score was the only significant predictor of group membership  (beta = −1.57, p < 0.001). Moreover, this model correctly predicted group membership for 89% of the terrorists and 83% of the non-criminals.
Receiver  operating characteristic (ROC) curves were calculated to test (i)  whether any of the assessed domains successfully discriminated  terrorists from non-criminals, and (ii) which of these domains yielded  the best discrimination accuracy. The best discrimination between groups  was afforded by global moral score (area under the curve 0.91,  confidence interval CI: 0.85–0.96; p < 0.01). At a cut-off of 2.5 on the global moral score, sensitivity and specificity were 0.86 and 0.84, respectively (Fig. 2e).  Fluid intelligence, executive functions, RPQ proactive aggression  score, MAI, SABI and TASIT total scores did not accurately discriminate  terrorists from non-criminals (Fig. 2e).
Given  the high discrimination accuracy obtained by the moral global score, we  tested whether moral judgement offered a better group classification  than the combination of measures yielding group differences. For this  purpose, the measures revealing differences between groups (RPQ  proactive aggression score, SABI situations subscale, MAI and TASIT  total scores), except for global moral score, were combined into a  support vector machine (SVM) model (see details in Data analysis). This  model achieved an average classification accuracy of 75%, a sensitivity  of 0.79 and a specificity of 0.73. The area under the ROC curve  calculated from the decision values produced by the SVM model was 0.71,  indicating that the model has acceptable discrimination accuracy.  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 = −2.27, p = 0.01) (Fig. 2d).
In  sum, our results provide evidence of distorted moral cognition in  extreme terrorists. Whereas previous psychological examinations have  used projective and self-report tests to characterize terrorist samples20,  this study used experimental tasks to assess moral cognition and other  social-cognitive domains in paramilitary terrorists. The finding that  moral judgement was the measure that best discriminated between groups,  whereas other measures showed mild or null differences, suggests that  distortion in this domain is a hallmark of the terrorist mindset. This  approach to understanding terrorists’ social-cognitive profiles has  important legal and forensic implications.
Adult  moral judgement typically depends on the capacity to represent and  integrate information about the intentions and consequences of actions21.  In many cases, moral judgement is determined primarily by intention;  however, when intention and outcome are in conflict, moral judgements  are normally construed by considering both factors22.  Here, terrorists exhibited the opposite pattern. Unlike non-criminals,  they judged attempted harm by focusing on the neutral outcome rather  than on the protagonist’s negative intention. Similarly, they judged  accidental harm by focusing on the negative outcome without considering  the neutral intention. Surprisingly, this moral judgement pattern  resembles that observed at early developmental stages23,24 (see Supplementary Information 3  for a further discussion). Thus, our results suggest that the  terrorists’ moral judgement is characterized by an overreliance on  outcomes rather than by the integration of intentions and outcomes.
This pattern opposes the widely described ‘harm magnification effect’25,  which shows that people overestimate the damage caused by intentional  harm compared with accidental harm, assigning more punishment and moral  condemnation25.  Indeed, terrorists judged attempted harm as more permissible and  accidental harm as less permissible than did non-criminals. Moreover,  unlike the latter, terrorists considered accidental harm to be more  morally wrong than attempted harm. Previous studies have suggested that  terrorists’ behaviour is goal-directed11.  The distorted moral judgement pattern observed here may be one of the  factors related to such a tendency. This does not mean that terrorists  are committed to a single focal goal. Instead, it suggests that our  sample is characterized by a general tendency to focus more on the  outcomes of actions than on the actions’ underlying intentions. Also,  when applied to terrorism, rational choice theory would assume that  terrorist acts usually emanate from rational, conscious outcome-oriented  decisions26,27.  A rational choice is one that maximizes one’s outcomes by choosing  means that favour the attainment of prime goals with the least sacrifice  of outcomes in terms of possible alternative goals28.  This does not imply that the decision is morally permissible or that  the means to achieve an outcome is behaviourally normal. Indeed, in our  target population, violent behaviour would seem to be the principal  means to pursue specific goals. In addition, although rational choice  theory may be applied to the interpretation of organizational terrorist  behaviour27,  it may also be used to explain individual behaviour. In fact, this  theory has been previously considered to explain criminal actions that  do not imply an organizational structure. For example, it has been  adopted as a framework to interpret individual criminal conducts, such  as sexual assault29,30 and theft31. According to this theory, sexual assault is the product of a rational decision29.  In this case, the offenders’ decisions would be based on the evaluation  of the situational factors, legal consequences, and the perceived costs  and benefits of the crime. Sexual assault cannot be objectively  considered as a normal response to maximize an outcome. Thus, although  our results are not odds with rational-choice theory, further studies  should explore other factors related to violent behaviour as a means to  achieve a specific goal.
Moreover, our results support the proposal27  that terrorists can suppress instinctive and learned moral constraints  against harming innocents, such as empathy, fairness and prosociality.  This could be caused by intrinsic or acquired factors, and by individual  or group forces. In addition, the profile observed in the terrorists  may reflect their fixation on utopian visions whereby only (idealized)  ends matter32. That is, their outcome-based moral judgements may be related to the belief that any action can be justified
insofar  as it favours the accomplishment of a utopian aim. Although these  speculations exceed the scope of our study, they open interesting  avenues for future research.
In  addition, it has been suggested that ideology may direct violence and  terrorism against a well-defined enemy to achieve a specific goal. This  view, however, does not fully apply to the terrorist group assessed  here. A crucial point is that most ex-combatants joined paramilitary  groups for economic reasons33,34,35,36. Only about 13% of ex-combatants had an ideological motivation for joining the paramilitary group24.  These points suggest that, in our sample, the use of violence is not  fully justified by ideology. Thus, it is unlikely that terrorist and  other criminal acts (massacres, murder, theft, kidnapping and fraud)  committed by these individuals were guided purely by their ideological  convictions. Moreover, it is unlikely that outcome-based moral judgement  observed in this terrorist group is related to its ideology. Future  studies should investigate whether there is a relationship between the  strength of terrorists’ ideological convictions and their moral  judgement patterns (see Supplementary Information 4).
The  distorted outcome-based moral judgement pattern observed in terrorists  may seem paradoxical if we consider that members of this paramilitary  group typically justify their actions in terms of moral imperatives10,11.  They have invoked the need, based on moral values, for so-called  ‘social cleansing’, killing thousands of people. Although this is  apparently contradictory, note that invoking an argument to justify an  action does not necessarily mean believing in that argument. Indeed,  moral justification may be a post-hoc strategy to save face or reduce  personal responsibility. Evidence for this has been provided by numerous  previous cognitive science studies37,38,39,40.
The  pattern of outcome-based moral judgement observed in terrorists does  not seem related to language comprehension impairments, as all  conditions involved similar verbal demands. Furthermore, as revealed by  covariance analyses, it was not related to working memory or abstraction  capacity impairments. In addition, the task employed has been  previously used in Latin-American populations19,41, which supports the cross-cultural validity of the moral scenarios considered.
Importantly,  the specific moral cognition profile of terrorists differs from those  previously observed in multiple populations via the same task15,17,42. For example, a previous study17  using the same scenarios showed that criminal psychopaths judged  accidental harm as more morally permissible than criminal  non-psychopaths. This pattern is notably different from the one shown by  terrorists. Although some members of paramilitary groups may exhibit  psychopathic traits, their moral judgements are not explained by such a  factor (see below). In this sense, an inconsistency has been noted  between psychopathic personalities and the mutual commitment and  cooperation evident within terrorist groups43. Moreover, there is little evidence that terrorists suffer from psychopathy27, and it has been shown that criminal behaviour is a correlate, not a component, of psychopathy44.  Thus, criminal psychopaths and terrorists seem to constitute  cognitively different populations possessing distinctive moral judgement  tendencies.
On the other  hand, the pattern observed in terrorists resembles moral judgement  impairments in patients with neurological disorders19,45 (frontotemporal dementia), who present high levels of impulsivity as well as sociopathic and criminal traits46,47.  This condition, too, is characterized by reduced reliance on  information about a person’s innocent or negative intentions and, hence,  overreliance on the action’s outcome for both attempted and accidental  harm. Thus, moral judgement seems comparable, to some degree, between  terrorists and subjects with damage in frontal and temporal regions  involved in moral cognition. This loose comparison suggests a need for  further research on the structural and functional brain correlates of  moral judgement in terrorists.
It  is also noteworthy that moral judgement in our target group was not  associated with fluid intelligence or executive functions. This finding  aligns with two strands of evidence. First, intellectual level does not  necessarily correlate with moral reasoning abilities48. Second, deviant moral judgements may be present even in individuals with normal or above-average IQ42.  Moreover, our results suggest a dissociation between fluid  intelligence, executive functions and moral judgement, in line with a  previous study49  showing that intelligence is not associated with external aggression.  Taken together, these data weaken the view that low fluid intelligence  and executive functions are key factors related to aggression and  offensive behaviour. However, given that the complexity of executive  functions makes it impossible for a single test to evaluate this  cognitive domain in its entirety, future studies should assess the  terrorists’ executive profile through an exhaustive neuropsychological  battery.
We also found  that scores on aggression scales measuring motives for aggression,  frequency of aggressive behaviours or levels or reactive aggression were  not related to moral judgement. However, levels of proactive aggression  were significantly associated with moral judgement performance. This  finding is consistent with research showing that participants who score  higher on aggression measures (that is, total RPQ score) are more likely  to favour utilitarian responses in moral dilemmas50. Also, our results align with previous evidence51  revealing that moral judgement is related to proactive aggression but  not to reactive aggression. Furthermore, although emotion recognition  abilities were partially associated with moral judgement, they did not  predict the terrorists’ moral judgement pattern or discriminate between  terrorists and non-criminals. In this sense, our findings corroborate  behavioural research15,52  suggesting that emotional processing deficits are associated with moral  judgement impairments. Our results are also in line with neuroimaging  studies18,53,54  showing that key brain regions involved in emotion processing (for  example the amygdala and ventromedial prefrontal cortex) are also  relevant for this domain. Moreover, the amygdala seems to be crucial in  supporting early detection of intentional harm54.
Crucially,  the logistic regression model showed that moral judgement was the only  domain significantly associated with group membership. Furthermore, ROC  curve analyses revealed that moral judgement was the measure with the  best sensitivity and specificity to distinguish between terrorists and  non-criminals. By the same token, the comparison with the SVM  classification confirmed that moral judgement performance, by itself,  was the best measure to classify the groups, even when compared with the  combination of those domains revealing distorted performance in  terrorists. Thus, deviant moral judgement seems to constitute the most  prominent attribute of our terrorist sample. In brief, this result  highlights the importance of evaluating moral judgement to characterize  terrorist groups and to understand the socio-cognitive processes  underlying their brutal acts.
From  a translational perspective, our findings have legal and forensic  implications. Sensitive instruments tapping socio-cognitive profiles  could eventually contribute to characterize terrorist behaviour.  Although our results do not suffice to determine whether moral judgement  tasks could be used to identify those terrorists more likely to relapse  or to predict who will become a terrorist, they do open the door to  future research on moral cognition in terrorist groups. Future studies  should test the predictive value of moral judgement and other  social-cognitive tasks to identify dangerous insurgent individuals. In  this sense, further cross-sectional and longitudinal studies are needed  to test the predictive value of moral cognition tasks in the assessment  of future aggressive behaviour and social adaptation. Moreover,  terrorism is undeniably a dynamic phenomenon in which group processes,  culture and socio-psychological factors are important to radicalization55. Therefore, future research on the topic should explore the relationship between these factors and social-cognitive domains.
More  particularly, our results may have regional implications for the  Colombian legal system, because most of the terrorists who participated  in this study were freed last year. Given that no significant  associations were found between the terrorists’ moral judgement  performance and the time spent in prison, their moral judgement pattern  could hardly be explained by proximity to release. Indeed, if such a  factor were biasing their performance, one would expect an effort  towards more socially acceptable responses. In addition, moral reasoning  is essential for proper social functioning and for preventing  delinquent behaviour56.  Because the terrorists in our sample exhibited skewed moral judgements,  emotion recognition impairments and high levels of aggression, especial  attention should be paid to them on release, especially in light of the  high levels of relapse reported among demobilized paramilitaries28,30.  Psychological and social-cognitive interventions may be beneficial for  these individuals. Also, further research should explore whether the  terrorists’ moral judgement changes during imprisonment or after  release.
Finally, although  psychopathy may be present to some degree in any group of delinquents,  we cannot confirm whether these terrorists are psychopaths. However, two  key aspects suggest that their socio-cognitive profile is not explained  by psychopathy. First, not all psychopaths are involved in criminal  behaviours44,57, and no conclusive evidence exists linking such a trait with terrorism27,58,59. Second, our results showed that moral judgement distortions differ between terrorists and criminal psychopaths17. Still, future research should explore the prevalence of psychopathy in terrorist paramilitary groups.
In  conclusion, this study provides evidence about the socio-cognitive  profile of terrorists, showing that moral judgement is the measure that  best distinguished between terrorists and non-criminals. In legal and  cognitive settings, intentions are assessed and often used to evaluate  others’ actions. The capacity to represent and reason about intentions  is crucial in judging whether others’ actions are right or wrong,  harmless or harmful, punishable or unpunishable21.  However, our results reveal that terrorists judge others’ actions by  focusing on the outcomes, suggesting that their moral code prioritizes  ends over means. Thus, impairments in processing intentions and in  integrating them with actions’ outcomes may be one of the key social  cognitive factors underlying the cruel acts committed by terrorist  paramilitary groups.


Our  sample included 66 incarcerated paramilitary terrorists who  participated in a collective demobilization from 2003 to 2006. The  demobilization process is formally supported by Colombian statutory law  975/05 (Ley de Justicia y Paz, Justice and Peace Law)60,  which promotes social reinsertion of armed group members who contribute  to national peace. The law offers reduced punishment by means of an  ‘alternative’ sentence (suspension of existing sentences, to be replaced  with imprisonment of no less than 5 and no more than 8 years) for  beneficiaries who comply with basic demobilizing requirements. These  individuals have not participated in any rehabilitation or reinsertion  programme (see Supplementary Information 5).
All  66 terrorists declared having participated in illegal armed right-wing  paramilitary groups and gave a full, voluntary deposition and confession  of crimes involving terrorist acts. This unique sample is characterized  by high levels of terrorism and insurgency as well as aggressive and  disruptive behaviours. Indeed, the paramilitary group to which they  belonged was designated as a terrorist organization by multiple  countries — for example the United States and Canada — and organizations  such as the European Union. All participants in this group were  convicted of murder, with a mean of 33 victims per subject (most of them  were accountable for several massacres, with death tolls sometimes  exceeding 600 victims). They had also engaged in other crimes, such as  theft, kidnapping and fraud. Each paramilitary was screened to exclude  neurological disorders, axis I psychiatric conditions and drug  consumption habits that might affect any of the target variables.
We  also formed a control group comprising 66 healthy individuals from the  same geographical region and with no terrorist background. These  participants were matched in age, sex and years of education with the  terrorists group (see Table 1).  A neuropsychiatric interview confirmed that control subjects had no  history of alcohol/drug abuse or neurologic or psychiatric disorders.  All participants provided written informed consent in agreement with the  Helsinki declaration. The Ethics Committee of the Autonomous University  of the Caribe approved the study.


Intellectual level and executive function measures

Intelligence was evaluated with the Wechsler Abbreviated Scale of Intelligence61  (WASI), which includes vocabulary and similarities subtests and  provides a verbal estimated IQ. Fluid intelligence was assessed via  Raven’s standard progressive matrices62. The maximum score on Raven’s test is 60 points.
Executive functions were evaluated through the IFS battery63, a brief and well-validated63,64,65,66,67  instrument which includes several subtests tapping into various  executive functions. This battery includes eight subtests: (i) motor  programming (Luria series, ‘fist, edge, palm’); (ii) inhibition  (subjects are asked to hit the table once when the administrator hits it  twice, or vice versa); (iii) motor inhibitory control; (iv) numerical  working memory (backward digit span); (v) verbal working memory (months  backwards); (vi) spatial working memory (modified Corsi tapping test);  (vii) abstraction capacity (inferring the meaning of proverbs); and  (viii) verbal inhibitory control (modified Hayling test). The maximum  global score on the IFS is 30 points. This measure was selected given  the limited time available to evaluate participants and its utility for  detecting executive function impairments in different populations63,64,65,66,68 (see Supplementary Information 6).

Aggression scales

All participants completed the MAI69,  consisting of 26 items for which participants are asked to choose  between three options (‘never or almost never’, ‘sometimes’ or ‘often’)  to indicate the frequency with which different motives (such as “You  have to defend your ideas,” or “You cannot control yourself”) trigger  aggressive behaviours. Scores on this inventory range between 26 and 78  points. Moreover, participants completed the SABI69.  This questionnaire has two subscales. The aggressive behaviours  subscale measures the frequency (‘never or almost never’, ‘sometimes’ or  ‘often’) of different types of verbal and physical aggression (for  example beatings, threats, attitudes or gestures of anger). This  subscale has scores between 9 and 27 points. The situations subscale  evaluates the frequency (‘never or almost never’, ‘sometimes’ or  ‘often’) of aggressive behaviours in response to specific situations  (for example, family problems, economic difficulties or health  problems). Scores on these subscales are between 13 and 39 points.
In addition, participants completed the RPQ70,  a self-report scale developed to distinguish between reactive and  proactive aggression. The scale consists of 23 items rated on a  three-point scale (0 = never, 1 = sometimes, and 2 = often). A total of  11 items assess reactive aggression (for example “Reacted angrily when  provoked by others”), and 12 items assess proactive aggression (for  example “Hurt others to win a game”). We calculated scores of reactive  or proactive aggression together with an overall score of total  aggression (the sum of reactive and proactive aggression scores).

Emotion recognition assessment

Emotion  recognition was evaluated through TASIT, a measure of social perception  based on videotaped vignettes of everyday social interactions71. This task, which has been used previously with incarcerated samples68,  introduces contextual cues (such as prosody, facial movement or  gestures) and additional processing demands (such as adequate speed of  information processing, selective attention or social reasoning) that  are not taxed when viewing static displays. We used a modified version  of part 1, called the Emotion Evaluation Test (EET). The EET assesses  recognition of spontaneous emotional expressions (fearful, surprised,  sad, angry, and disgusted) as conveyed by professional actors  interacting in everyday situations. Emotional meaning is indicated by  speaker demeanour (voice, facial expression and gesture) together with  the social situation. Some scenes involve a single actor who talks  either on the phone or to the camera. Other scenes depict two actors, in  which case participants are instructed to focus on one of them. All  scripts are neutral in content and do not lend themselves to any  particular emotion. The brief EET comprises 10 short (15–60 s)  videotaped vignettes. After viewing each scene, the participant must  choose from a forced-choice list the emotion expressed by the focused  actor. The maximum global score is 10 points.

Moral judgement task

Moral judgement was assessed through a previously reported protocol15,19.  Participants were presented with 24 scenarios involving two  individuals. Each scenario was presented as a written story and featured  four variations following a 2 × 2 design: (1) the protagonist either  harmed another person (negative outcome) or did no harm to him/her  (neutral outcome); (2) the protagonists either believed that they would  cause harm (negative intention) or believed that they would cause no  harm (neutral intention). Each possible belief was true for one outcome  and false for the other outcome. Thus, the four scenarios were (1) no  harm, (2) accidental harm, (3) unsuccessfully attempted harm, and (4)  successfully attempted harm (see Fig. 1).  This experimental manipulation allows one to dissociate intentions and  outcomes, so that some of the combinations of variables do not entail  bad intentions (no harm and accidental harm), while others do (attempted  harm and successfully attempted harm), and some of them have pernicious  outcomes (accidental harm and successfully attempted harm), but others  do not (no harm and attempted harm). Thus, the attempted harm and  successfully attempted harm clearly involve moral violations based on  unjustified aggressions.
After reading each story, the  participants rated the scenario on a Likert scale ranging from totally  forbidden (1) to totally permissible (7). In total, six trials of each  of the four conditions were presented. The stimuli were presented in  pseudorandom order, and the conditions were counterbalanced across  participants. To decrease working memory load, the whole text remained  visible until the trial was completed. This instrument has been widely  used on both neurotypical and patient populations17,19,42,45,72,​73,​74,​75,​76,​77, and it has also been administered to incarcerated psychopaths17 (see more information on the task’s validity in Supplementary Information 7).

Data analysis

Demographic,  neuropsychological, and experimental data were compared between groups  with ANOVA and Tukey’s HSD post-hoc tests (when appropriate). The  assumption of normality was verified using the Shapiro–Wilk test. Data  also met the assumption of homogeneity of variance, assessed with  Levene’s test. Following the procedure reported elsewhere17,19,42,45,73,​74,​75,​76,​77,  moral judgement data was analysed via a 2 (intention: neutral,  negative) × 2 (outcome: neutral, negative) × 2 (group: terrorists,  non-criminals) repeated-measures ANOVA. Considering that terrorists  showed a non-significant trend (p = 0.09) for lower working  memory and abstraction capacity, we applied ANCOVA tests adjusted for  IFS scores on these domains to control for their influence on moral  judgements. We reported only effects that were still significant after  covariation.
Paired-sample t-tests were used to compare  intra-group performance on the moral conditions in which terrorists  differed from non-criminals. We estimated overall moral judgement  impairment by calculating a global moral score. Also, we conducted a  multiple regression analysis to explore whether moral judgement was  associated with other relevant domains. We estimated a model in which  the above global score was considered as the dependent variable. The  following variables were included as predictors: group, Raven’s  matrices, IFS, MAI, SABI and RPQ subscales, and TASIT total scores.
We  conducted ROC curve analyses, which are useful to assess the  effectiveness of a given test in classifying individuals as belonging  within one group or the other78, and it allows comparing the discrimination accuracy of two or more tests79.  Specifically, we employed these analyses to test (i) whether any of the  assessed domains successfully discriminated terrorists from  non-criminals, and (ii) which of these domains yielded the best  discrimination accuracy. The variables included in ROC curve analyses  were Raven’s matrices, IFS, MAI, SABI and TASIT total scores, the RPQ  proactive aggression score, and the moral judgement global score. The  area under the ROC curve was used as a measure of discriminatory  accuracy.
In addition, we tested whether moral judgement offers a  better group classification than the combination of measures yielding  group differences. Using the latter measures, we implemented a SVM to  classify terrorists and non-criminals. SVM is a supervised  classification algorithm rooted in statistical learning theory80,  in which input data are classified into two classes (for example  terrorists and non-criminals). Conceptually, input vectors are mapped to  a higher-dimensional feature space using kernel special functions.  Classification is performed by constructing a hyperplane in the feature  space that optimally discriminates between the two classes of the  training data by maximizing the margin between the two data clusters80.  The variables (sensitive measures which discriminate between groups)  included in the SVM analysis were MAI, SABI and TASIT total scores, and  the RPQ proactive aggression score (see Results section for the  measures’ selection criteria). The SVM analysis was implemented through  the WEKA software package81,  with 10-fold cross validation and a radial basis kernel function. An  additional ROC curve was then calculated from the decision values  produced by the SVM model. Finally, using a Mann–Whitney test, we  assessed the statistical difference between the area under the ROC curve  for the moral global score and that for the SVM model.
Regreso al contenido | Regreso al menu principal