Introduction: Although various theoretical models of radicalization mention the interaction between dispositional and (perception of) contextual factors as predictors of radicalized attitudes and behaviors, such interactions are rarely empirically tested. Simultaneously, results of studies focused on studying individual roles of dispositions or (perception of) contextual factors explain only a minor portion of the variation in radicalized attitudes and behaviors. This gap between the theoretical models and their empirical evaluation motivated us to conduct this study. The study was focused on testing the individual and interactive contribution of dispositions and perceptions of contextual factors – more precisely, the general dark personality trait (i.e., the dark core of personality) and relative deprivation (with attributed blame) - to the prediction of radicalized outcomes. In this study, based on quantitative research methodology, we focused on two types of relative deprivation that are often mentioned in radicalization research: egoistic relative deprivation, or relative deprivation one experiences based on the comparison of their status, belongings or opportunities with those of similar individuals, and fraternalistic relative deprivation, or relative deprivation individuals experience based on the comparison of their ingroup’s status, belongings or opportunities with those of some outgroup. In line with the contemporary distinction between cognitive and behavioral radicalization and the general absence of political violence in the context in which the study was conducted, we focused on radicalized attitudes and intentions as criteria. Additionally, this study tested whether blame attribution determines the relationship between relative deprivation and radicalized attitudes and behavioral intentions, which at the moment of writing this manuscript, represented an unanswered question stemming from meta-analyses. Methodology: This study was based on a cross-sectional research design. Analyses conducted in this study were based on two convenience samples of youth from Croatia: a student (N = 716) and a nonstudent (N = 140) sample. Potential participants were invited to take part in the study via social media and mailing lists, with invitations containing the link to the online questionnaire. The online questionnaire contained multiple self-report scales necessary to measure the central constructs of this study, along with several items measuring common socio-demographic factors (e.g., age, sex, socio-economic status). The order of questionnaires was fixed – participants first completed the questionnaire measuring the dark core of personality, followed by scales measuring egoistic relative deprivation, fraternalistic relative deprivation, attributed blame, support for political violence and non-violent political actions, intentions to participate in activities of a non-violent and violent organization and socio-demographic characteristics. The analytical approach adopted in this study consisted of multiple steps conducted in R. The first step after data cleaning included testing the psychometric properties of the applied multiitem measures. After ascertaining that all the measures exhibited acceptable psychometric properties in both groups, we tested our hypotheses based on the principles of structural equation modelling. Testing of first-order latent interactions was conducted following the principles of the product-indicator approach. Unfortunately, this approach exhibited suboptimal performance when testing more complex models that included multiple interaction terms. Therefore, latent profile analysis was conducted to identify profiles of participants with relatively consistent results on the dark core trait, egoistic and fraternalistic relative deprivation and attributed blame. The analysis provided robust evidence of the existence of four profiles, and the participants’ probability of belonging to each profile was correlated with support for political violence and intentions to participate in activities of a violent organization. In all these analyses, we applied two models of intentions to participate in activities of a violent organization. In line with a more traditional approach, we firstly conceptualized them as a common factor of four belonging factors. Additionally, and in line with the more contemporary approaches, we used bifactor (S-1) models to analyze the variance of intentions to participate in activities of a violent organization independently of the variance of general (normative) political mobilization. Results: Firstly, we tested the individual contribution of the dark core of personality, egoistic and fraternalistic relative deprivation in the prediction of support for political violence and intentions to participate in activities of a violent organization. The results revealed that participants scoring higher on the dark core and fraternalistic relative deprivation, respectively, exhibited more support for political violence and stronger intentions to participate in activities of a violent organization. On the other hand, stronger egoistic relative deprivation was related to more support for political violence. Simultaneously, it was unrelated to the intentions to participate in activities of a violent organization. Secondly, we tested the interactive contribution of attributed blame and the two types of relative deprivation in the prediction of support for political violence and intentions to participate in activities of a violent organization. The analyses exhibited weak evidence in favor of the interactive contribution of attributed blame and fraternalistic relative contribution to the prediction of intentions to participate in activities of a violent organization. All other interactions, although similar in magnitude and direction, failed to reach the threshold of significance, which could, at least in part, be attributed to a somewhat high correlation between fraternalistic relative deprivation and attributed blame, as well as the low number of participants not blaming the outgroup for experienced inequality. Thirdly, we tested the interactive contribution of the dark core of personality, egoistic relative deprivation, fraternalistic relative deprivation and attributed blame in the prediction of support for political violence and intentions to participate in activities of a violent organization. The applied latent profile analysis consistently indicated the existence of four profiles: 1) a profile of non-deprived individuals who did not blame the outgroup for the experienced injustice, 2) a profile of dominantly egoistically deprived individuals, 3) a profile of dominantly fraternalistically deprived individuals, and 4) a profile of dominantly fraternalistically deprived individuals who achieved high scores on the dark core of personality. While the probability of belonging to the first profile was associated with less support for political violence and weaker intentions to participate in activities of a violent organization, only the probability of belonging to the fourth profile was consistently related to stronger support for political violence and stronger intentions to participate in activities of a violent organization. Additionally, the profile of dominantly egoistically deprived individuals was related to the higher support for political violence. Simultaneously, it was unrelated to the intentions to participate in activities of a violent organization. Therefore, next to confirming the findings of earlier studies, outcomes of conducted analyses resulted in additional insights. Namely, fraternalistic relative deprivation (and attributed blame) was not related to radicalized outcomes among individuals scoring lowest on the dark core of personality. On the other hand, in the absence of fraternalistic relative deprivation (and attributed blame), no relationship between the dark core of personality and radicalized attitudes and intentions was established. Conclusion: Altogether, these findings suggest that dispositional factors and factors emerging from the perception of the environment complement each other in determining radicalized attitudes and intentions, which offers a new perspective on the interplay between the hypothesized factors that lead to radicalized cognitions and intentions. However, it is also important to keep in mind that contemporary models of radicalization are more complex than the models studied in this manuscript. Therefore, the outcomes of this study should not be treated as final answers but as steppingstones that should encourage researchers to consider both individual and interactive contributions of the existing models of cognitive and behavioral radicalization in order to gain deeper insights and, eventually, result in more effective detection, prevention and treatment of individuals at risk of involvement in political violence.