Introduction

Antisocial behaviors encompass a wide range of socially questionable and aversive tendencies that challenge established cultural and ethical norms, often at a social cost. Although many individuals engage in some form of antisocial behavior (e.g. lying, cheating, or manipulating) they do not necessarily become criminals. Criminal behavior, reflected in disruptive actions such as theft, assault, fraud, vandalism, and murder, inflicts significant harm on society by negatively affecting social order and causing extensive damage. To understand why some individuals cross the line into criminality, researchers discovered that a repertoire of coexisting "bright" and "dark" personality traits play a crucial role1,2. Personality traits are relatively stable patterns of emotion, motivation, cognition, and behavior that emerge from the dynamic interaction between individuals and their environments, influencing and shaping future behavior3. In stable and predictable environments, individuals tend to develop “bright” traits, such as the motivation to maintain positive relationships and emotional stability4. Conversely, in harsh and unsupportive environments, individuals may develop over time “darker” traits, including dominance, deceitfulness, and disregard for moral norms, which increase the risk of engaging in criminal activities5.

The bright triad of personality

When relating typical personality to criminal behavior, most studies relied on the seminal Five-Factor Model (FFM)6 comprising five main dimensions: Openness, Conscientiousness, Extraversion, Agreeableness, and Neuroticism. A combination of high Agreeableness, high Conscientiousness, and low Neuroticism was recently deemed to constitute the “Bright triad” of personality, defined as a combination of basic traits predicting both well-being components of personality and a prosocial value orientation2. Both incarcerated individuals and those from the general population who exhibit low levels of this protective profile were more likely to engage in criminal activities7,8,9. For instance, high Neuroticism, coupled with low Agreeableness and low Conscientiousness was associated with higher aggression in inmates10,11. However, there are also mixed findings revealing higher levels of Conscientiousness among prisoners and individuals involved in criminal activities12,13. Looking at the other two typical personality dimensions, some studies found no significant differences in Extraversion between offenders and non-offenders14. Further complicating the picture, some studies support an association between low Openness and Extraversion and criminal behavior in juvenile offenders15,16, while others revealed higher levels of these traits in adult inmates17 or no association with criminal behavior18. Regarding recidivism, while some studies revealed that officially recorded recidivism cases were associated with higher levels of Agreeableness and Openness compared to nonrecidivists19, others showed that ex-prisoners who possess lower Agreeableness and Conscientiousness were more prone to re-offend20.

To summarize, one perspective is to see individual differences in prosocial and antisocial behaviors as potentially predicted by FFM dimensions alone, thus seeing the dark personality traits as simply the opposite pole to cooperative, prosocial tendencies (e.g. honesty-humility, agreeableness), both placed on a broader agreeableness-antagonism continuum21. An alternative view is that the dark personality brings a distinct contribution to predicting antisocial outcomes, with specific traits being differentially associated with criminal behavior22.

The dark triad of personality

The Dark Triad of Personality consists of three interrelated traits: Narcissism (grandiosity, entitlement, selfishness), Machiavellianism (duplicity, social manipulation), and Psychopathy (impulsivity, diminished empathy, and callous affect)23. Among these traits, Narcissism is considered the most benign dimension, probably due to the co-existence of positive attributes, such as social desirability, and higher achievement motivation, contributing to social adaptation24. Psychopathy is deemed the darkest and thus particularly studied in relation to criminality25 and was shown to increase the probability of criminal recidivism, sexual recidivism, and instrumental violence in adults26. In addition, individuals high in psychopathic traits begin their criminal activities at a young age; tend to commit a wider variety of crimes, and recidivate faster than non-psychopathic criminals25,27.

Dark personality traits can provide crucial insights into predispositions towards immoral behaviors, but their predictive value regarding criminality is limited, possibly due to the presence of both non-aversive and adaptive aspects (e.g. vulnerability in the case of Narcissism, or disinhibition in the case of Psychopathy)28. Although the dark traits vary in their dark saturation, their aversive essence is potentially better captured in models identifying a shared latent variance between the traits, a “dark core” or a “dark factor” of personality29,30.

The dark factor of personality

The diverse dark traits previously discussed emerge as expressions of a purported “dark core”, named the Dark factor of personality30,31,32, defined as the "general tendency to maximize one’s individual utility—disregarding, accepting, or malevolently provoking disutility for others, accompanied by beliefs that serve as justifications". Recent studies document the stability of the Dark factor in predicting various forms of antisocial behavior, such as deception or self-reported aggression in the general population33. However, research on the Dark factor within prison populations remains limited to a single unpublished study34, indicating no significant difference between the two populations in terms of the general Dark Factor score, except for an increased Moral Disengagement in prisoners.

The next step was to go beyond a network of multiple traits to explore the internal structure of the Dark Factor and identify five dark themes underlying the dark core of personality: Callousness, Sadism, Vindictiveness, Deceitfulness, and Narcissistic Entitlement35. Conceptually, Callousness has been described as the lack of empathy towards others and indifference towards the distress of others. Deceitfulness implies an inclination towards adopting deceptive, unlawful, and illicit behaviors, disobeying the norms of civic cooperation in the pursuit of personal goals. Sadism refers to the tendency to provoke disutility in others and enjoyment from destroying the proprieties of others, while Vindictiveness entails a strong desire for vengeance to restore a perceived sense of equity. Narcissistic Entitlement shows a pronounced sense of deservingness and disproportionate claim to resources that can support any kind of aversive behavior.

The authors advocate the use of the five themes rather than the twelve aversive traits, as they allow for a more parsimonious and empirically valid description of the Dark factor35. However, we have not yet identified any empirical investigation analyzing the five Dark Factor themes within a prison population.

Criminal history

While individuals from the general population are motivated to display their socially desirable behaviors, when placed within the confines of a prison environment, individuals may feel justified to use unethical strategies, such as deception, violence, or vengeance, to address perceived or potential harm36,37 or develop underlying cognitive justifications for the expression of aversive or undesirable behaviors. Comparing prisoners to the general population, Balafoutas et al.,38 found a negative association between the time spent in prison and cooperation, benevolent behaviors towards others, sincere responses, and reciprocity, thereby supporting the notion that the prison environment provides the opportunity to acquire antisocial attitudes and criminal skills from the other inmates39. Unlike the social world, where prosocial behaviors are highly adaptable, the prison environment involves a more constant aversive atmosphere characterized by manipulation, deception, and violence40. The need to operate in this dangerous environment may affect personality and influence the expression of dark and bright tendencies, which themselves might predispose an individual to re-offend.

However, there is a paucity of research tapping into the relationship between dark and bright personality dimensions in prison settings41. Numerous questions persist, such as whether there is a blend of bright and dark personality traits in prisoners, and to what degree the dark traits overshadow the bright(er) ones, potentially found in the general population. Finally, to what extent personality traits can enable us to distinguish occasional lawbreakers from those with a consistent pattern of offending (recidivism)? Thus, a direct comparison between prisoners and individuals from the general population, through the lenses of the five Bright and Dark factors of personality may shed light on potential variations within and across these populations and reveal key characteristics associated with crime and recidivism.

The current study

We sought to examine for the first time in a unitary design the mix of bright and dark core personality tendencies to identify the differentiating elements between individuals from the general and prison population (first-time offenders, recidivists), while controlling for potential confounding factors, such as age, sex and socially desirable presentation tendencies.

Firstly, we aimed to reveal the predictive value of the bright and dark core of personality for criminal history, based on established findings concerning Agreeableness, Conscientiousness37, and Neuroticism41 and to explore the potential added contribution of Openness and Extraversion, where evidence was more mixed. We also expected that the Dark Factor of personality would predict criminal history above the contributions of the FFM dimensions (Aim 1). A second aim was to explore whether replacing the Dark Factor total score with its five expressed dark themes (Callousness, Deceitfulness, Narcissistic Entitlement, Sadism, and Vindictiveness) would better predict criminal behavior over the contribution of the FFM personality dimensions than the Dark Factor (Aim 2).

Results

Data analysis was performed on three groups, differentiated by their criminal history (recidivists, first-time offenders, and individuals from the general population, with no criminal history), in terms of FFM dimensions (Neuroticism, Extraversion, Openness to experience, Agreeableness, and Conscientiousness), as well as the Dark Factor and five specific dark themes: Callousness, Deceitfulness, Narcissistic Entitlement, Sadism, and Vindictiveness35.

Descriptive data regarding means and standard deviations, as well as differences in all FFM dimensions, Dark Factor, and the Dark Themes depending on criminal history, can be found in the Supplementary information. A general overview of typical and dark personality dimensions across the three groups is reflected in Fig. 1.

Figure 1
figure 1

Group means (with 95% confidence intervals) for the three Criminal History Groups (None, First offense, Recidivism) across the FFM dimensions (Neuroticism, Extraversion, Openness, Agreeableness, Conscientiousness), the Dark factor, and the Dark Themes (Callousness, Deceitfulness, Narcissistic Entitlement, Sadism, Vindictiveness). Note: The Left panel = Typical personality dimensions (from left to right, images represent Neuroticism, Extraversion, Openness, Agreeableness, and Conscientiousness), and the Right panel = Dark personality dimensions (from left to right, images represent The Dark Factor, Callousness, Narcissistic Entitlement, Deceitfulness, Sadism, and Vindictiveness).

Full size image

Preliminary analysis

As a preliminary analysis, we performed a two-step procedure using classical structural equation modeling (SEM) approach to replicate the previously found bifactorial structure of the Dark personality with one general Dark factor and five specific themes or factors: Callousness, Deceitfulness, Narcissistic Entitlement, Sadism, and Vindictiveness35. The findings (resulting models, congruence coefficients, RMSEA, and SRMR for all specific factors) from the sample of prisoners (N = 296) and matched community participants (N = 282) (manuscript in preparation) largely confirmed the five themes previously supported by Bader et al.35.

Across the five models and in both samples (community and prison) the Dark factor emerged as a general factor (Dark factor) and the five themes emerged as specific factors. All models exhibited comparable goodness of fit indices for both community and prison samples, RMSEA values being situated slightly below the 0.05 threshold. Also, before conducting any statistical analyses employing the five themes, we estimated the Variance Inflation Factors. VIFs ranged between 1.62 for Vindictiveness to 2.40 for Callousness, suggesting a moderate degree of overlap between the factors, but not sufficiently large as to require the elimination of any of the five dimensions. The VIFs for the remaining dimensions were: 1.89 Narcissistic Entitlement, 2.28 Deceitfulness, and 2.22 Sadism.

To address our two research aims, we employed a multinomial logistic regression to test two different models of bright and dark personality dimensions predicting perseverance in criminal behavior. To facilitate interpretation, all the variables were standardized. For the parameter estimates, a value below p = 0.05 was considered significant. As an initial note, Nagelkerke’s Pseudo R-Square values for the two models were relatively similar and the difference was not particularly striking: Nagelkerke’s Pseudo R-Square for the first model (Dark Factor and the FFM personality traits) was 0.24 and for the second model (the Dark themes and the FFM personality traits) was 0.27.

Aim 1: First, we estimated the relationships between the overall Dark Factor score, the FFM personality traits, and membership in the three groups (control group – no offense, first offense, and criminal history), while controlling for age, sex, and impression management.

Table 1 outlines the unique contributions of each predictor included in the first model (Dark Factor and the FFM personality traits) in differentiating between each criminal history type and the no offense condition. Although we controlled for the effect of age, sex, and impression management strategies, Neuroticism, Extraversion, Openness, and the Dark Factor still displayed significant contributions to accounting for group membership (Table 2).

Table 1 Predictors’ Unique Contributions in Explaining Criminal History—Multinomial Logistic Regression (N Offenders = 296, N Control = 282).
Full size table
Table 2 Predictors’ Unique Contributions in Explaining Criminal History—Multinomial Logistic Regression (N Offenders = 296, N Control = 282).
Full size table

For a more detailed analysis, we calculated the parameter estimates having as a reference group the sample of participants with no criminal history (control). Consequently, each predictor has two parameters corresponding to the first offense and criminal history outcomes. Out of the five typical personality traits, Neuroticism and Extraversion had unique and significant contributions to differentiating between both the prison (first offense and criminal history) and the control group (no criminal history). Specifically, Neuroticism significantly increased the chances to be included in both first offense and criminal history categories, while Extraversion revealed a contrasting result, higher scores increasing the odds for an individual to be included in the control group rather than the first offense group or the criminal history group.

Regarding Openness and the Dark Factor, the two personality dimensions had unique contributions only in differentiating between the first offense group and the control one, with higher scores increasing the odds of being included in the first offense group (by a factor of 2.43 for Openness and 2.36 for the Dark Factor), but had a negligible effect on the chances to be included in the criminal history group. Additional information regarding parameter estimates for the first model (Dark Factor and the FFM personality traits) can be found in Supplementary information.

Aim 2: Next, we employed the same approach to the relationships between the distinct five dark themes and the typical personality dimensions against the same criterion. For the second model, we employed the same analytical approach to estimate the separate contributions of the five dark themes. Similarly, Neuroticism, Extraversion, and Openness had unique contributions to accounting for group membership. Out of the five dark themes, only Deceitfulness and Sadism had significant contributions to account for the focal outcome.

The parameter estimates for this model, having as a reference group the sample of participants with no criminal history (control), indicated that three out of the FFM personality traits significantly accounted for increased chances to belong to either one or both of the criminal offense groups: Neuroticism, Extraversion, and Openness. Similar to the first model (Dark Factor and the FFM personality traits), higher Neuroticism scores increased the odds of being included in both first offense and criminal history categories, and Openness differentiated only between the first offense group and the control one, with higher scores increasing the odds to be included in the first offense group, by a factor of 2.53. Regarding Extraversion, for every one standard deviation increase, the chances to belong to the first offense group or the criminal history group rather than the control one decreased by 0.38, respectively 0.46.

For the dark themes, higher Sadism scores increased the odds of being included in the first offense group by a factor of 2.20 but had a negligible effect on the chances of being included in the criminal history group. Deceitfulness significantly increased the odds of being included in both first offense and criminal record categories by a factor of approximately 1.70 (1.77 – 1.79). Additional information regarding parameter estimates for the second model (Dark themes and the FFM personality traits) can be found in Supplementary information.

Discussion

We aimed to deepen the understanding of how bright and dark personality profiles contribute to the occurrence and perseverance of criminal behavior. We highlighted the differential predictive power of the bright profile (the five dimensions of the Five-Factor Model (FFM) and the dark profile (the Dark Factor with the five dark themes) in explaining both offending and recidivism.

Beyond the bright personality

Firstly, we sought to confirm the additive contribution of the FFM dimensions and the Dark Factor in predicting criminal history while controlling for age, sex, and impression management.

Contrary to our initial expectations, the Bright Triad dimensions did not have the expected significance in predicting crime. Of the three dimensions, only Neuroticism emerged as a risk factor for both first-time offending and recidivism, confirming previous findings suggesting that an increased tendency to perceive the world as problematic and threatening, coupled with negative affect can predispose to criminal behavior and reoffending18,42,43. The other two dimensions, Agreeableness and Conscientiousness did not turn out to be significant predictors. While previous articles indicate that lower levels of Agreeableness are associated with offending10,12 and re-offending20,44, our study does not support this association, highlighting that lower Agreeableness alone may not be a sufficient predictor of criminal history, but a possible byproduct of a criminal track. An alternative explanation may be that the self-reported Agreeableness is already contaminated by a desire to present oneself in a favorable light (social desirability)45, which also contaminates the relationship between Agreeableness and criminal behavior. In our sample, Agreeableness was strongly correlated with Impression Management (r = 0.40). This is probably why when we controlled for social desirability, the differences in Agreeableness which differentiated at least the recidivist group from the two other groups (see Fig. 1 and the ANOVAS in the Supplementary material) disappeared. Interestingly, a study on a large prison population showed that inmates with high Impression Management scores were less likely to report antisocial attitudes, and more likely than those scoring low to be convicted of the most morally reprehensible crimes (homicide, sexual assault, pedophilia, and incest), plus they received longer sentences46. The authors conclude that high levels of this trait might not be indicators of genuine higher agreeableness, but instead of a manipulative desire to be seen as good, despite committing the most morally repugnant crimes (similar to 76% of our sample being convicted for murder, sex-related crimes and robbery). This indicates that previous findings that did not control for social desirability might have overestimated the differences in Agreeableness between prisoners and the general population.

Going beyond the Bright Triad to look at the remaining FFM dimensions, our results confirmed that both first-time offenders and recidivists scored lower than non-offenders in Extraversion. This confirms previous studies that show offenders as being generally less talkative, sociable, and confident and displaying fewer assertive behaviors in social interactions12. This negative association between Extraversion and criminal behavior could also be a consequence of adapting to the strict regulations of prison, which limits opportunities for assertive social interaction. The prison is a dangerous and aggressive environment, and high levels of manifest Extraversion, resulting in increased talkativeness, initiating social contact, and high excitability may generate higher conflict risk, so inmates can learn that it is not a desirable trait. In addition to that, Openness was higher in first-time offenders compared to non-offenders. Our findings suggest that people who describe themselves as curious, sensitive, and open to new experiences and adventures, are more likely to engage in risky behavior, but not necessarily to recidivate. People with high levels of Openness may be more likely to break the law due to their willingness to take risks. This conflicts with studies showing a link between low Openness and criminal behavior12,47, or revealing no meaningful differences43,48,49, but confirms studies indicating higher Openness among inmates compared with participants from the general population17.

As a significant addition to the existing literature on the contribution of dark personality traits above typical personality dimensions, we found that the Dark Factor predicted criminal behavior beyond the five dimensions of the FFM (which includes the core of bright personality, the Bright triad). This supports previous studies showing that individuals who weigh their utility over others and declare antisocial values and justifications to support their actions are more likely to break the law22. Interestingly, results revealed that the Dark Factor had no additional predictive power for recidivism, suggesting its limited value in differentiating between individuals without a criminal record and those with a criminal history. Other dynamic and static factors (time spent in prison, education, lack of social support, financial difficulties, antisocial acquaintances) might be more responsible for recidivism than simply the dark personality core50.

Dark themes of personality and criminal behavior

Next, we explored the themes that comprise specific characteristics of the dark core of personality, more or less aversive, thus differing in their capacity to predict offending and/or reoffending. Lack of empathy and compassion, indifference towards the distress of others (reflected by the Callousness theme), the strong desire for vengeance to restore a perceived sense of equity (Vindictiveness), and a pronounced sense of deservingness and disproportionate claim to resources (Narcissistic Entitlement) were shown in other studies to predict criminal behavior51,52,53. However, in the Dark Factor framework used in our paper, they were not significantly associated with either criminal behavior or criminal history. Although some aspects of Callousness, Vindictiveness, and Narcissistic Entitlement may equip individuals with the necessary traits to pursue selfish goals through aversive behaviors (including criminal acts) with little or no regard for the pain or disutility caused to others, these may not be sufficient to predict the risk of crime.

Sadism, reflecting the tendency to malevolently provoke disutility in others and derive pleasure from such acts, predicted offending, but not recidivism. This confirms that sadistic tendencies are a risk factor only for crime, not for reoffending54. A possible explanation may be that recidivism is influenced by a wide variety of factors, such as the lack of social support after prison release, antisocial associates, limited employment opportunities, or even social stigmatization. Faced with these challenges, ex-offenders might repeat this behavior because of their inability to cope with their environment and not so much because they simply derive pleasure from harming others50.

Lastly, Deceitfulness emerged as a pivotal marker of first-time offending, being also the only dark theme explaining recidivism. Defined as a tendency to use deceptive, unlawful, and illicit behaviors for personal gain, this theme not only distinguishes between the individuals who behave unethically or immorally and those who are willing to cross the boundary to illegality but adds value to the prediction of crime, above the Dark Factor. This presents the prospect that people with high levels of this theme are more likely to “bend the truth” and the law, for selfish purposes, not only once, but repeatedly.

Conclusions

Comparing first-time offenders or recidivists to individuals with no criminal history, we observed no differences in their declared motivation (or lack thereof) to cooperate with others (Agreeableness). However, individuals engaging in criminal behaviors described themselves as less able to maintain emotional stability in the face of negative affect, including sadness, anxiety, and anger (high Neuroticism), less likely to express positive emotions, assertive behavior, and less sensitive to social attention (low Extraversion), and reporting more curiosity and proneness to new experiences and risky behaviors (Openness). Additionally, recidivists, probably due to the time spent in prison, may develop a preference for task-oriented coping, and become less distractible and better organized (tendency to show higher Conscientiousness). Looking at the “dark side of the moon”, our results support the notion that individuals with a criminal record have a higher tendency to provoke disutility in others (high Dark Factor), to deceive others in the pursuit of their interests (Deceitfulness), and even derive satisfaction from doing so (Sadism).

Limitations

Our study has the distinct advantage of exploring for the first time the themes within the Dark Factor and relating them to FFM dimensions in a large sample of inmates from three high-security prisons. However, several limitations to this ambitious endeavor must be acknowledged. First, although both the sample from prisons and the general community had their education and cognitive abilities measured, they were not fully matched on these dimensions, generating caution in interpreting the findings. Despite utilizing large samples and controlling for potential confounding effects in age, sex, and impression management, the comparisons with matched participants without criminal records may yield slightly different results. Another limitation is the exclusive reliance on self-report measures for assessing personality traits, potentially influenced by inmate’s tendency to present a favorable image [55] due to privacy concerns, fear of negative consequences, or attempts to avoid stigma and personal shame (although the Impression Management scale partially accounts for this favorable self-presentation tendency). Finally, even though the increased size of our sample is one of the assets of this investigation (compared to other research on dark traits in prisoners which involved lower sample sizes, e.g. N = 131 in Eriksson’s research) [12], it did not allow for a nuanced analysis according to offense type, which was deemed relevant by previous literature. For instance, it was shown that homicide offenders displayed a distinct dark personality profile (lower psychopathy and sadism) compared to other offenders56. In our sample, although the percentage of homicide is significantly higher than other offense types (57.4%), we also had white-collar crimes (10.5%), sexual offenses (6.4%%), and crimes involving lower levels of violence, like theft or robbery (12.2%), which suggests the need for future studies focused on potential specificities according to a type of offense or criminal versatility56.

Evaluating individual propensity towards criminal and antisocial behaviors through the lens of Dark personality can be important in developing effective prevention and rehabilitation policies. Firstly, understanding how everyday behavior is influenced by Dark dimensions is crucial for designing tailored training programs for prison staff. A model example is the guide "Working with offenders with personality disorder – A practitioner's guide"57, offering practical insights to help staff better communicate with challenging inmates, handle conflicts, enhance overall security, and foster an environment conducive to change and rehabilitation within the prison. Secondly, specific interventions should target self-serving justifications supporting a criminal lifestyle. Further research should explore the possibility of employing the Dark Factor inventory as a screening measure for antisocial behavior, as the total Dark Factor score predicted the extent to which individuals decide to maximize their utility while disregarding the negative consequences of their actions for others33. Teaching individuals to recognize and challenge their antisocial beliefs could foster greater cognitive flexibility, assertive problem-solving skills, and resilience. Caution should be exercised when designing intervention programs to reduce reoffending, as an increase in social skills may also enhance the use of Dark dimensions and better equip offenders to manipulate and take advantage of their victims' vulnerabilities.

Methods

Participants

The prison sample included 296 incarcerated offenders, 209 men (70.6%) and 87 women (29.4%), recruited from two maximum security prisons (from Northwestern Romania), with an age range from 21 to 79 years (M = 39.65, SD = 11.28). More detailed socio-demographic data for the prison and community samples can be found in the Supplementary information as Supplementary Table S1. The control sample comprised 282 participants, 156 men (55.3%) and 126 women (44.7%), with an age range from 19 to 64 years (M = 36.08, SD = 10.25).

Measures

All instruments were administered in Romanian. Items within each scale were answered on a 5-point Likert scale, ranging from 1 = strongly disagree to 5 = strongly agree. Social desirability was measured using Paulhus Deception Scale (PDS)58.

The Dark Factor Inventory (D70)31 is the self-report scale measuring the Dark Factor of Personality, and the 12 dark traits: Egoism, Machiavellianism, Narcissism, Moral Disengagement, Psychological Entitlement, Psychopathy, Sadism, Self-Centeredness, Spitefulness, Cruelty, Frustration, and Greed. The internal consistency coefficient for the English version is α = 0.97 31 and α = 0.95 for the Romanian version (previously validated in a large Romanian-speaking sample). Correlations among the scales ranged from 0.72 to 0.87.

The NEO Five-Factor Inventory (NEO-FFI)59 is a 60-item self-report questionnaire used to measure the basic five dimensions of personality, with 12 items for each factor: Neuroticism, Extraversion, Openness to experience, Agreeableness, and Conscientiousness. Higher values of mean scores indicate higher levels of a specific personality dimension. Internal consistency reliability as measured with Cronbach’s alpha ranged from 0.73 to 0.84 for the Romanian version60.

Procedure

The current study is part of a larger research on dark and bright tendencies and behaviors in the prison environment and was conducted following local regulatory and legal frameworks that govern research with vulnerable populations, including inmates. To guarantee compliance with ethical and deontological research principles, all methods used for data collection respected both national and international standards, including the APA’s Responsible Conduct of Research guidelines and the recommendations of the European Committee for the Prevention of Torture and Inhuman or Degrading Treatment or Punishment.

The research protocol received the ethics approval of the Scientific Council of the Babeș-Bolyai of Cluj-Napoca no. 23553/17.12.2018. Also, due to specific regulations applicable to the prison system, the research protocol has been approved by the management boards of all institutions included in the study (penitentiaries, public institutions, and private companies), before data collection.

Before being included in the study, all participants were informed about the research purpose, type of data collected, target sample, procedures taken for participant data protection, possible risks, rewards, the guarantee of confidentiality and anonymity of their personal information, as well as the possibility of consent withdrawal at any time. Participation in the study was voluntary and conditioned by the signed informed consent of all subjects (both prison and community participants). Each participant received a unique code of registration for anonymity reasons. In return for their participation, individuals from the general community received a small gift and time off work, while the prisoners received three credits to be exchanged for specific rewards. No monetary compensation has been provided. Participants were required to have adequate Romanian reading–writing skills and those with a known history of mental illness or receiving psychiatric medication were excluded.

Data collection was conducted in two sessions. First, each individual completed an individual interview assessing a variety of personal information such as age, sex, family and educational background, employment, marital status, alcohol and drug use, criminal history, and mental health issues. For the inmate sample, additional information has been gathered from the prison files, such as offense type, sentence length, prison conduct, and recidivism rate. In addition, all participants completed the Raven's Standard Progressive Matrices- Classic, within a time limit of 60 min. In the second session, all the participants individually completed the Dark Factor and the Five Factor dimensions of personality, organized in groups of 6 to 10 individuals, while making sure their responses remained confidential.