Dysfunctional personality, Dark Triad and moral disengagement in incarcerated offenders: implications for recidivism and violence

Abstract

It is particularly important to study the underlying processes of the dysfunctional personality patterns and of antisocial behaviour in the prison population, to identify them and to analyse the functioning of the psychological mechanisms involved in these constructs. The main goal of this study was to analyse dysfunctional personality patterns, Dark Triad, moral disengagement mechanisms and their relationship with violence and recidivism. Participants were 63 incarcerated offenders in two prisons. The study found a higher degree of moral disengagement in the participants convicted for crimes involving the use of violence. The results indicated that antisocial and aggressive-sadistic tendencies were the dysfunctional personality traits most strongly associated with moral disengagement. The binary logistic regression analysis showed that the variables, which influenced recidivism in criminal behaviour, were the use of violence in the crime committed, antisocial personality traits, and advantageous comparison and dehumanisation as mechanisms of moral disengagement.

Introduction

Much of the research conducted to date into antisocial behaviour has focused on analysing individual differences (Black, 2015; Docherty et al., 2019; Frick, 2016), the characteristics of this pattern of behaviour (Glenn et al., 2013; Morales et al., 2019), associated personality traits (Alcázar-Córcoles et al., 2017; Decuyper et al., 2013) and the typical underlying psychological mechanisms (Mayer et al., 2018; Raine, 2019). In particular, a substantial number of research studies have investigated moral reasoning in individuals with antisocial behaviour, analysing which socio-cognitive mechanisms facilitate this group of people to morally disengage in order to commit criminal or immoral acts (Bandura et al., 1996). Furthermore, some research has attempted to describe the dysfunctional patterns of personality that explain criminal or antisocial behaviour. One of the patterns that has been associated with criminal behaviour is the so-called Dark Triad of personality (Paulhus & Williams, 2002). Most of these lines of research have attempted to describe antisocial behaviour, but few research studies have analysed the two concepts together in the prison population. Nevertheless, in a large number of countries, there is a growing professional and social interest in examining the prevalence, psychological characteristics and underlying psychological processes of persons in the prison population with dysfunctional personality disorder traits (Andersen, 2004; Arboleda-Flórez, 2009; Fazel & Seewald, 2012).

In order to gain a greater understanding of the moral reasoning that leads people to act in a way that is harmful to others, the main goal of this study is to analyse the relationship between the dysfunctional patterns of personality and the mechanisms of moral disengagement in members of the prison population who have committed and have been criminal sentenced for one or several criminal acts. A better understanding of the mechanisms that can lead to criminal and antisocial behaviours can promote the design and implementation of prevention programmes that take into account socio-cognitive processes and those related to moral reasoning. In addition, we consider it is important to know which variables can explain recidivist behaviour, considering the variables historically described as predictive of antisocial behaviour, but also introducing socio-cognitive and moral mechanisms that have not been previously considered. This knowledge may in the future benefit treatment programmes inside and outside correctional institutions.

Dysfunctional personality patterns and antisocial behaviour

Personality traits are persistent patterns in terms of how one perceives, thinks about and relates to one’s environment and to oneself. These characteristics can be observed in a wide variety of settings of both a social and a personal nature. When these personality traits are inflexible and dysfunctional, causing significant functional deterioration and subjective distress, they constitute a personality disorder (American Psychiatric Association, APA, 2013). These dysfunctional patterns of personality entail a great many problems in the individual’s life, on both a personal and a social level, which is why it has become an important area of study for mental health researchers today.

We know that some of these dysfunctional patterns of personality are strongly related to antisocial behaviour and frequently associated with poor social functioning, unemployment and difficulties in the management of interpersonal relations, leading to the manifestation of behavioural problems and substance abuse, among other risky behaviours, generally resulting in criminal conduct (Gunderson et al., 2011).

Much of the research into the antisocial personality pattern has been devoted to studying the individual differences between male and female offenders. Generally, the research conducted to date has found that men are more likely than women to display antisocial behaviours (Archer, 2009; Fagan & Lindsey, 2014; Thomson et al., 2019). Furthermore, crime and conviction rates among women for violent crimes are lower than those for men (Churcher & Nesca, 2013; Logan, 2008; Strand & McEwan, 2012). Studies that analyse the risk factors and characteristics of antisocial behaviour have generally focused on the male population. In relation to the prevalence of psychopathology in the prison population, research studies have shown that when compared with the wider community, the members of this population group, in particular its female members, are more prone to depressive and anxiety disorders, personality disorders and substance abuse disorders (Ducat et al., 2017; Fazel et al., 2006; Maden et al., 2006). The understanding of these differences has a significant bearing on the analysis of antisocial behaviour and on the identification of separate treatment strategies specifically aimed at male and female offenders.

Dark Triad of personality

The Dark Triad of personality, first described by Paulhus and Williams (2002), is composed of the personality traits of Machiavellianism, narcissism and psychopathy. Machiavellianism is characterised by falseness or deceitfulness, externalisation of guilt, emotional coldness and manipulation of interpersonal relations for self-gain (Ali et al., 2009). Narcissism is characterised by exaggerated self-esteem and grandiosity, egoism, arrogance and exploitation of interpersonal relations, and seeing others as a means to bolster self-perception even further (Campbell, Reeder, Sedkides, & Elliot, 2000; Campbell, Rudich, & Sedikides, 2002). Last of all, psychopathy is characterised by lack of empathy, difficulties in having satisfactory relationships with others, superficial affection, lack of remorse and absence of feelings of guilt, inability to learn from experience, impulsiveness and presence of antisocial behaviour that tends to begin in childhood (Cleckley, 1976; Hare, 1991; Levenson et al., 1995). This construct has become increasingly important in the study of its association with underlying psychological processes, above all with empathy (Jonason & Krause, 2013; Wai & Tiliopoulos, 2012).

These three personality traits have been associated with antisocial and criminal behaviour (Barry et al., 2007; Muris et al., 2017), including aggression (Barlett, 2016; D. N. Jones & Neria, 2015). Research suggests that these associations differ for the three personality traits and antisocial behaviour in adults (Klimstra et al., 2014; Lau & Marsee, 2013). Some studies report that psychopathy is the most socially undesirable trait of the three, more strongly related to behavioural dysregulation and aggression (Jonason et al., 2015; Paulhus & Williams, 2002). Narcissistic personality tends to be more firmly linked to ego-threat and reactive aggression (Campbell et al., 2004; Czarna et al., 2016). Last of all, Machiavellianism was the factor most strongly associated with certain risky behaviours in the adult population (Maneiro et al., 2019).

More recently, sadistic personality has been added to this dark constellation, deriving to the currently named Dark Tetrad of personality (Chabrol et al., 2009; Paulhus & Jones, 2015). This new construct has been gaining importance since its association with antisocial behaviour has been confirmed (Buckels et al., 2013; Chabrol et al., 2009) but also because of its relation with other types of aversive behaviours such as online trolling (Buckels et al., 2013) or the delight or satisfaction towards the harm of others (Porter et al., 2014). It should be emphasised that sadistic personality traits have been shown to contribute significantly to the perpetuation of antisocial behaviours in students over Dark Triad personality traits (Chabrol et al., 2015).

As such, evidence exists of the relationship between the Dark Triad of personality and antisocial behaviour in subclinical samples, although few studies have been conducted involving members of the prison population. The identification of the potential underlying factors of these personality traits will help us to understand antisocial behaviour and to analyse which underlying processes may help in preventing criminal acts and in creating intervention programmes.

Mechanisms of moral disengagement

Some socio-cognitive theories on morality argue that moral reasoning is mediated by a series of mechanisms of self-regulation, based on social norms and one’s own values, which ensure the regulation of moral conduct (Bandura et al., 1996). The voluntary inactivation of these self-monitoring processes is known as moral disengagement. It involves the selective use of certain socio-cognitive mechanisms that entail the breaking of rules, the disinhibition of impulses and the carrying out of immoral actions (Bandura, 1990).

These mechanisms involve the reinterpretation of harmful behaviour, the minimisation of the notion that the individual is doing harm, the falsification or distortion of the consequences of violent or immoral behaviour, and the blaming and dehumanisation of the victim (Bandura et al., 1996; Osofsky, Bandura, & Zimbardo, 2005). Accordingly, Bandura (1990) describes eight mechanisms of moral disengagement: (a) moral justification: injurious behaviour is portrayed as if it has a moral purpose in order to make it acceptable; (b) euphemistic labelling: destructive behaviour is portrayed as innocuous by means of using sanitised and convoluted language; (c) advantageous comparison: one’s own harmful behaviour is made to seem harmless by comparing it against that of other individuals; (d) displacement of responsibility: self-censuring reactions are spared because people believe they are not the actual agent of their actions; (e) diffusion of responsibility: one’s own responsibility can be diffused if an entire group is engaged in the same behaviour; (f) distortion of consequences: the individual minimises or avoids facing harm when the behaviour that causes it is overlooked, minimised, distorted or questioned; (g) dehumanisation: the self-regulation of conduct is deactivated by denying other people their human qualities; and (h) attribution of blame: the individual sees their harmful behaviour as forced by circumstances and not as a personal decision.

According to some authors, these socio-cognitive processes of justification and moral reasoning can act as a linking factor between the dysfunctional traits of personality and antisocial behaviour (Caprara et al., 2014; Sijtsema et al., 2019). Moreover, these socio-cognitive patterns of moral reasoning may help to justify antisocial behaviour, and their continued use may increase the likelihood of reoccurrence of this type of behaviour.

Most of the research in this area focuses on children and adolescents (Gini et al., 2014; Hyde et al., 2010; Hymel & Perren, 2015; Paciello et al., 2008; Sijtsema et al., 2019), finding positive relations between individuals who exhibit mechanisms of moral disengagement and antisocial behaviour, and also concluding that these mechanisms may act as predictors of aggressive behaviour (Obermann, 2011; Risser & Eckert, 2016). Research in recent years has highlighted the importance of the mechanisms of moral disengagement more specifically in cyberbullying (Pornari & Wood, 2010), in prison bullying (South & Wood, 2006), in situations of terrorism (Bandura, 2004), in war and military operations (McAlister et al., 2006) and, last of all, in the legal destruction of the environment and harmful industrial activities (White et al., 2009).

Dark Triad personality and mechanisms of moral disengagement

As far as morality is concerned, some studies have found a positive relation between a more utilitarian moral judgement and the psychopathic and Machiavellian personality traits (Djeriouat & Trémolière, 2014; Glenn et al., 2009). A research study by Arvan (2013) found a positive relation between the personality traits of the Dark Triad and more conservative moral judgements. A recent study of the relationship between the Dark Triad of personality and the mechanisms of moral disengagement suggested that the characteristics of the three personality traits that comprise the Dark Triad of personality are associated in a different way with the mechanisms of moral disengagement and, therefore, with antisocial behaviour in a sample of adolescents (Sijtsema et al., 2019).

Most of the specific studies on the relationship between the mechanisms of moral disengagement and the Dark Triad have focused on the analysis of the psychopathic personality trait (Hyde et al., 2010; Risser & Eckert, 2016; Shulman et al., 2011). DeLisi et al. (2014) argue that adolescents with psychopathic personality traits tend to feel fewer moral emotions, such as guilt or shame. Meanwhile, Shulman et al. (2011) concludes that the participants with psychopathic personality traits have a greater tendency to justify their antisocial behaviour. By contrast, a recent meta-analysis by Marshal et al. (2018) has found a weak association between the psychopathic personality traits and moral reasoning difficulties in adults, inferring that these individuals may have difficulty in distinguishing between right and wrong and, therefore, are not in a position to employ mechanisms of moral disengagement. Last of all, the findings of a research study by Risser and Eckert (2016) indicate that both antisocial personality traits and moral disengagement mechanisms serve as predictors of non-violent antisocial behaviour but that only the lack of remorse is a predictor of violent conduct.

In respect of Machiavellian traits, the findings are similar to those of studies on psychopathic personality traits due to their overlapping nature (Miller et al., 2017). In this respect, a research study by Egan et al. (2015) found evidence that moral disengagement mechanisms in adults tend to intervene in the association between the Machiavellian personality trait and antisocial behaviour. Meanwhile, in a study on the relationship between the Dark Triad and moral disengagement mechanisms, Sijtsema et al. (2019) found a positive relation between the Machiavellian personality trait, moral disengagement mechanisms and antisocial behaviour. However, they do not specify which of these mechanisms are most strongly associated with the Machiavellian personality trait.

The literature shows that the narcissistic personality trait is distinguished from the psychopathic and Machiavellian traits in respect of its relationship with moral disengagement mechanisms. In this respect, some studies have shown the association between narcissistic personality trait, moral disengagement and antisocial behaviour in adult population in different ways. A study by Locke (2009), with a university sample, found that the narcissistic personality trait was associated with the moral disengagement mechanism of dehumanisation. In another study with a sample of sportspeople, B. D. Jones et al. (2017) found that moral disengagement mechanisms acted as mediators between antisocial behaviour and narcissistic personality traits. Therefore, there is evidence of a relation between narcissistic personality traits and moral disengagement. However, there is no information about which are the specific mechanisms involved in these personality traits and how they are associated with antisocial behaviour.

Studies on this association in the prison population are limited, but we have found recent studies that investigate this relationship specifically with psychopathic personality traits. A study by Petruccelli et al. (2016) showed the difference between penalised participants (non-sex and sex offenders) and controls, finding that penalised participants exhibited higher levels of moral disengagement. Particularly, those participants whose offense was related to deprivation of sexual liberty seemed to use more moral disengagement mechanisms than the rest of the participants with convictions for other offenses. Subsequently, the same authors found a positive association between moral disengagement and psychopathic personality (Petruccelli et al., 2017), and, more particularly, they found that this personality was more related to those mechanisms involving individual responsibility, such as attribution of blame, advantageous comparison and displacement of responsibility.

Recidivism in antisocial behaviour

Previous research related to recidivism in antisocial behaviour has found multiple variables explaining this conduct. One of the most noteworthy risk factors of recidivism is the previous record of violent behaviour of the prison inmate (Coid et al., 2007; Eher et al., 2016; Lund et al., 2013). Furthermore, it has been shown that the more severe the psychopathology presented by an inmate, the greater the risk of recidivism (Bengtson et al., 2019; Fazel et al., 2016; Fazel & Wolf, 2015). Dysfunctional patterns of personality and, more specifically, psychopathy also increase the risk of recidivism in prison inmates (Harris et al., 2017; Olver et al., 2015). Research into the sociodemographic factors involved in the risk of recidivism indicates that being male and living in a disadvantaged social context are associated with the repetition of antisocial behaviour (Bonta et al., 2014; Collins, 2010; Wright et al., 2014). The identification of the individual factors involved in the risk of recidivism in prison inmates may facilitate greater knowledge of the mechanisms of risk evaluation, as well as their management with specific intervention measures for the prison population.

Summarising, empirical evidence of the association between moral disengagement and Dark Triad personality traits exists separately but basically in adolescent and community populations. Nevertheless, as far as we know, no research exists that analyses the relationship between dysfunctional personality traits and the eight mechanisms of moral disengagement specifically in the adult prison population. The main goal of this study is to examine the prevalence of and the association between the dysfunctional patterns of personality and moral disengagement in the prison population. More specifically, it aims (a) to investigate gender differences in all study variables; (b) to analyse differences in participants convicted for crimes involving the use of violence in the investigation variables; (c) to analyse the association between the dysfunctional personality patterns, Dark Triad and moral disengagement mechanisms in this sample of prison inmates; (d) to examine the differences between recidivist and non-recidivist participants in the study variables; and (e) to analyse which variables of the study have an influence on the risk of reoffending.

Method

Participants

Participants were 63 offenders (69.8% men and 30.2% women), inmates of two prisons in Catalonia (Spain), who were serving sentences for a wide variety of crimes. The age range of the participants varied from 25 to 80 years old (M = 40.29, SD = 10.64). 68.3% of the participants were Spanish nationals by birth, while 31.7% were born in foreign countries (Table 1).

Table 1.

Sociodemographic characteristics of the participants.

Participants sociodemographics
Sex N = 63	Men (N = 44) 69.8% Women (N = 19) 30.2%
Origin N = 63	Spain (N = 43) 68.3% France (N = 2) 3.2% East Europe (N = 1) 1.6% Latin America (N = 9) 14.3% North Africa (N = 5) 7.9% Western Africa (N = 3) 4.8%
Profession N = 57	Administrative activity (N = 8) 12.7% Agrarian-forestry activity (N = 1) 1.6% Artistic activity (N = 2) 3.2% Communication (N = 2) 3.2% Economy and business (N = 3) 4.8% Education and orientation (N = 3) 4.8% Aesthetics (N = 1) 1.6% Health service (N = 2) 3.2% Applied technique (N = 23) 36.5% Hospitality sector (N = 12) 19% No information (N = 6) 9.5%
Education N = 58	Primary education (N = 11) 17.5% Secondary school (N = 21) 33.3% High school (N = 3) 4.8% Professional training (N = 16) 25.4% University (N = 7) 11.1% No information (N = 5) 7.9%
Religion N = 60	Catholicism (N = 16) 25.4% Evangelism (N = 6) 9.5% Orthodoxism (N = 1) 1.6% Islam (N = 4) 6.3% No religion practiced (N = 33) 52.4% No information (N = 3) 4.8%
Current crime N = 59	Robbery with violence (N = 13) 20.6% Crimes of homicide and murder (N = 6) 9.5% Gender and domestic violence (N = 7) 11.1% Sexual assault (N = 2) 3.2% Crimes against public health and public treasury (N = 8) 12.7% Fraud and forgery (N = 5) 7.9% Robbery without violence (N = 4) 6.3% Criminal gang and drug trafficking (N = 10) 15.9% Traffic offences (N = 4) 6.3% No information (N = 4) 6.3%

Participants with a poor understanding of Spanish and participants who were on remand awaiting sentencing were excluded from the study. The participants took part in the study on a voluntary basis, having been duly informed of its characteristics. Loss of participants in some of the statistical analyses was due to two factors: some participants decided to drop out before completing the test or because psychometric instruments were excluded where reliability and validity were compromised.

Procedure

A research protocol was designed and submitted to the Penitentiary Services of the Department of Justice of the Government of Catalonia for approval. It was approved by the Rehabilitation Service on 12 February 2018, with record number 0310/836/2018. Periodical meetings were held with the multidisciplinary team of the prisons in order to design the procedure to be followed. In both prisons, all the security, confidentiality and anonymity protocols of the Penitentiary Services of the Department of Justice of the Government of Catalonia were strictly followed.

At the start of the study, the entirety of prison modules of the first prison were visited in order to explain the characteristics of the study and to ask for volunteers to take part in it. In the second prison only the women's module was visited; since in the first institution the number of female offenders was reduced, an attempt was made to achieve greater female participation in the study. The questionnaires were administered in two sessions – to prevent fatigue and to improve concentration and performance – and in a group setting in the educational facilities of each module. Personality disorders were assessed in Session 1, and Dark Triad and moral disengagement mechanisms in Session 2. The data were collected from March to June 2018. The main researcher – first signatory of this paper – was present throughout the administration of the psychometric instruments in order to clear up any doubts that might have arisen.

In order to analyse the results, groups of participants were created on the basis of the sociodemographic variables of the study. These classifications were made to study two components closely linked to antisocial behaviour, to obtain more qualitative information (whether the crime used violence or not) and more quantitative information (number of previous institutionalised crimes). First, the types of crime for which the participants had been convicted were classified as follows: (a) crimes involving the use of violence: robbery with violence, homicide and murder, gender-based violence and sexual assault; and (b) crimes not involving the use of violence: crimes against public health (trade and processing of substances harmful to society and related to drug trafficking) and against the public treasury (fraud and evasion of public finance and social security taxes), fraud and forgery, robbery without violence, forming part of criminal and drug gangs, and traffic offences. Second, considering that having a criminal history has been identified as one of the strongest predictors of reoffending (Andrews & Bonta, 2010, Leung et al., 2021; Maden et al., 2006) the participants were also classified on the following basis: (a) recidivists: inmates who had already served time for one or several other crimes in prison; and (b) non-recidivists: prisoners serving their first sentence in prison. The latter classification was made on the basis of previous studies where prior criminal history was used to assess recidivism (Fazel et al., 2016; Leung et al., 2021; Pechorro et al., 2019). In this study, to objectify the variable, a measure of institutionalised recidivism was used, in other words, how many times the participant has had a conviction and prison sentence for committing a crime.

Measures

Sociodemographic data

An ad hoc questionnaire of sociodemographic data was created in order to describe the characteristics of the participants in the study. Questions were included on age, nationality, completed schooling, religion practised, psychological background and record of consumption of addictive substances – including currently. The participants were also asked about the type of crime(s) for which they had been convicted, the length of their sentence and their criminal record.

Dysfunctional personality patterns

The dysfunctional personality traits were evaluated using the third edition of the Millon Clinical Multiaxial Inventory (MCMI–III; Millon, Davis, & Millon, 2007). This is a self-administered clinical questionnaire used to evaluate personality disorders and clinical syndromes. It is composed of 175 true–false questions that provide information on four validity scales and 24 clinical scales grouped according to the level of severity: (a) personality scales (schizoid, avoidant, depressive, dependent, histrionic, narcissistic, antisocial, sadistic, compulsive, negativistic and masochistic); (b) severe personality pathology (schizotypal, borderline and paranoid); (c) clinical syndromes (anxiety, somatoform, bipolar, dysthymia, alcohol dependence, drug dependence, post-traumatic stress disorder); and (d) severe clinical syndromes (thought disorder, major depression, delusional disorder). Its Spanish adaptation has good psychometric properties, with a test–retest reliability of between .84 and .96, and an internal consistency of more than .80 (Cardenal & Sánchez, 2007).

Given that the use of the instrument in this study is aimed at detecting the dysfunctional patterns of personality, the following results were presented in respect of the 14 scales of personality disorders.

Dark Triad

Dark Triad of personality was evaluated by means of the Short Dark Triad (SD3; D. N. Jones & Paulhus, 2014; Spanish adaptation by Nohales, 2015). This is a 27-item inventory split into three areas to measure the three personality traits that comprise the Dark Triad: Machiavellianism (α = .70), narcissism (α = .69) and psychopathy (α = .75). Each area contains nine items, which through a five-point Likert scale are used to evaluate the various facets of the three constructs. Given that this instrument is a very recent creation, few studies have analysed its psychometric characteristics, although three studies have found satisfactory evidence regarding its reliability and validity (D. N. Jones & Paulhus, 2014; Maples et al., 2014; Pineda et al., 2020).

Moral disengagement

The mechanisms of moral disengagement were evaluated by means of the Mechanisms of Moral Disengagement Scale (MMDS; Bandura et al., 1996; Spanish validation by Rubio-Garay et al., 2017). This is a 32-item inventory that examines the eight mechanisms of moral disengagement: moral justification (α = .72), euphemistic labelling (α = .44), advantageous comparison (α = .57), displacement of responsibility (α = .51), diffusion of responsibility (α = .45), distortion of consequences (α = .58), attribution of blame (α = .40), dehumanisation (α = .73). A reliable total score is obtained by the sum of the 32 items as an overall indicator of moral disengagement (α = .87). On a four-point Likert scale, the participants had to respond to statements justifying behaviour considered socially negative. High scores in this evaluation indicate a high degree of moral disengagement. Previous studies that have used the MMDS to evaluate adults have indicated an acceptable internal validity of the instrument (Paciello et al., 2008).

Data analysis

Different analyses were conducted with the IBM SPSS Statistics 25 program. First, we did three group comparisons to analyse differences in dysfunctional personality, Dark Triad and moral disengagement between (a) men and women, (b) violent and non-violent offenders and (c) recidivist and non-recidivist offenders. For this purpose, we carried out a t-Student test for independent samples. Second, we carried out Pearson correlations analysis with all the study variables. Then, to analyse the predictive accuracy of the specified predictor’s variables of the present investigation, a binary logistic regression was performed.

Results

Table 1 shows the sociodemographic data of the 63 participating offenders in relation to the sociodemographic variables of the study. The work activity of most of the participants has been in the sector of applied technical activities, followed by the hospitality sector and the administrative sector. As regards schooling, 50.8% of the participants say that they have completed compulsory education, while 41.3% say they have completed some form of higher education or vocational training. As far as religion is concerned, more than half the sample (52.4%) state that they do not practise any religion, while 42.8% say that they do practise one. Last of all, in respect of the crime for which they were convicted, 47.5% of the participants have been convicted for crimes involving the use of violence while 52.6% have been convicted for crimes not involving the use of violence.

Gender differences in dysfunctional personality patterns, Dark Triad and moral disengagement

Table 2 shows the differences according to gender for the means and standard deviations of the variables of the study (dysfunctional personality patterns, Dark Triad and mechanisms of moral disengagement). The results show that in our sample of offenders, there were significant differences according to gender in most personality disorders (except in the narcissistic, antisocial and dependent personality disorders). Female participants obtained higher mean scores than their male counterparts in the majority of the scales. The size of the effect of these differences could be considered between medium (95% confidence interval, CI [0.50, 0.70]) and large (95% CI [0.80, 1.00]). No significant differences were found between men and women in any of the constructs of the Dark Triad of personality or in the mechanisms of moral disengagement, although it must be pointed out that in most of the subscales, the mean scores of the female participants were higher than those of their male counterparts.

Table 2.

Descriptive statistics and gender differences for all study variables.

	N = 62		d	T	p
	Men (N = 44)	Women (N = 18)
	M (SD)	M (SD)
Dysfunctional personality patterns
Schizoid	7.11 (3.72)	10.33 (4.37)	>−0.82	−2.90	.005
Avoidant	6.07 (4.49)	8.94 (4.75)	−0.63	−2.25	.028
Depressive	7.14 (5.86)	12.78 (5.77)	−0.98	−3.45	.001
Dependent	6.70 (4.42)	8.33 (6.08)	−0.29	−1.02	.313
Histrionic	16.05 (4.02)	12.50 (3.27)	>0.94	3.31	.002
Narcissistic	15.27 (3.48)	14.33 (4.70)	>0.24	0.86	.389
Antisocial	9.70 (4.88)	10.67 (5.29)	−0.19	−0.68	.494
Sadistic	7.86 (3.75)	11.89 (5.10)	−0.97	−3.43	.001
Compulsive	18.86 (3.78)	16.33 (3.83)	>0.67	2.37	.021
Negativistic	8.09 (4.65)	13.72 (5.75)	−1.14	−4.03	.000
Masochistic	4.07 (4.41)	7.22 (4.85)	−0.70	−2.48	.016
Schizotypal	5.77 (4.72)	10.83 (7.22)	−0.77	−2.74	.012
Borderline	6.43 (4.68)	12.28 (6.31)	−1.14	−4.02	.000
Paranoid	8.48 (5.03)	12.56 (6.74)	−0.74	−2.61	.011
Dark Triad
Machiavellianism	2.91 (0.73)	3.14 (0.80)	−0.30	−1.07	.286
Narcissism	2,54 (0.57)	2.39 (0.42)	>0.29	0.96	.336
Psychopathy	2.24 (0.53)	2.33 (0.50)	−0.17	−0.59	.544

	N = 63
	(N = 44)	(N = 19)
Moral disengagement			>
Moral justification	8.29 (3.30)	8.00 (2.98)	>0.09	0.33	.739
Euphemistic language	6.20 (2.47)	6.21 (2.04)	>−0.00	−0.00	.993
Advantageous comparison	7.68 (3.35)	9.05 (4.15)	>−0.36	−1.38	.172
Displacement of responsibility	8.77 (3.80)	8.84 (3.13)	>−0.02	−0.70	.945
Diffusion of responsibility	11.59 (2.86)	13.00 (3.36)	>−0.44	−1.69	.094
Distortion of consequences	8.11 (3.22)	6.78 (2.32)	>0.47	0.12	.112
Attribution of blame	8.63 (3.23)	8.89 (3.29)	>−0.07	−0.28	.773
Dehumanisation	8.06 (3.63)	9.10 (4.22)	>−0.26	−0.99	.332
Total MMDS	67.36 (18.15)	69.89 (14.48)	>−0.15	−0.53	.593

Note: MMDS = Mechanisms of Moral Disengagement Scale.

Differences between participants whose crimes involved the use of violence and those whose crimes involved no violence in dysfunctional personality patterns, Dark Triad and moral disengagement

Table 3 shows the descriptive results and differences between the participants whose crimes involved the use of violence and those whose crimes involved no violence, in all the variables of the study. In respect of dysfunctional personality traits, in our sample we observed significant differences in the schizoid personality scale, where participants whose crime involved the use of violence obtained a higher score, and on the narcissistic and compulsive personality scales, where participants whose crimes did not involve the use of violence obtained higher scores. The size of the effect of these differences could be considered as medium (95% CI [0.50, 0.70]). As regards the three constructs of the Dark Triad of personality, significant differences were found in respect of the narcissistic personality trait, where participants whose crimes did not involve the use of violence obtained higher scores. The size of the effect of this difference was medium (95% CI [0.50, 0.70]). No significant differences were found in the variables that comprise the mechanisms of moral disengagement.

Table 3.

Descriptive statistics and differences between participants with violence or without violence in the perpetration of crime.

	N = 59		d	t	p
	Violence (N = 27)	No violence (N = 32)
	M (SD)	M (SD)
Dysfunctional personality patterns
Schizoid	9.41 (4.41)	6.91 (3.87)	0.60	2.31	.024
Avoidant	7.48 (5.33)	6.31 (4.14)	0.24	0.94	.348
Depressive	9.67 (7.24)	7.91 (5.69)	0.27	1.04	.301
Dependent	7.81 (4.94)	6.91 (5.14)	0.17	0.68	.494
Histrionic	14.22 (4.51)	15.81 (3.65)	−0.38	−1.49	.141
Narcissistic	13.59 (4.07)	15.94 (3.38)	−0.62	−2.41	.019
Antisocial	11.15 (5.24)	9.09 (4.65)	0.41	1.59	.116
Sadistic	9.81 (5.16)	8.41 (3.99)	0.30	1.18	.243
Compulsive	17.04 (4.49)	19.16 (3.34)	−0.53	−2.07	.043
Negativistic	10.26 (6.23)	9.41 (5.24)	0.14	0.57	.570
Masochistic	5.89 (5.35)	4.31 (4.10)	0.33	1.25	.217
Schizotypal	8.00 (6.36)	6.28 (5.54)	0.28	1.10	.273
Borderline	9.00 (6.40)	7.50 (5.45)	0.25	0.97	.335
Paranoid	9.37 (6.38)	9.75 (5.54)	−0.06	−0.24	.808

	Violence (N = 26)	No violence (N = 33)
Dark Triad
Machiavellianism	3.10 (0.73)	2.91 (0.76)	0.25	0.95	.343
Narcissism	2.33 (0.55)	2.65 (0.50)	−0.60	−2.32	.024
Psychopathy	2.42 (0.45)	2.16 (0.56)	0.51	1.90	.062

	N = 60
	Violence (N = 27)	No violence (N = 33)
Moral disengagement
Moral justification	8.77 (3.23)	7.84 (3.19)	0.28	1.11	.270
Euphemistic language	6.22 (2.50)	6.15 (2.26)	0.02	0.11	.909
Advantageous comparison	7.70 (3.42)	8.48 (3.86)	−0.21	−0.81	.416
Displacement of responsibility	8.33 (3.55)	9.36 (3.65)	−0.28	−1.10	.276
Diffusion of responsibility	12.25 (3.52)	12.06 (2.46)	0.06	0.25	.799
Distortion of consequences	7.51 (2.83)	8.06 (3.19)	−0.18	−0.68	.494
Attribution of blame	8.92 (3.89)	8.81 (2.61)	0.03	0.12	.899
Dehumanisation	9.00 (3.70)	7.78 (3.77)	0.32	1.24	.217
Total MMDS	68.74 (17.74)	68.57 (16.48)	0.00	0.03	.970

Note: MMDS = Mechanisms of Moral Disengagement Scale.

Associations among dysfunctional personality patterns and moral disengagement

In order to examine the relationship between variables, correlations were calculated between the variables of the study. Table 4 shows the correlations between the dysfunctional patterns of personality and the mechanisms of moral disengagement. In this respect, a positive relation was found between the moral justification mechanism and the antisocial, sadistic and negativistic personality traits. Meanwhile, a negative relation was found with the compulsive personality trait. A positive relation was found between the advantageous comparison mechanism and the schizoid, depressive, dependent, antisocial, sadistic, negativistic, masochistic, schizotypal, borderline and paranoid personality traits. The only positive relation found in respect of the displacement of responsibility mechanism was with the narcissistic personality trait. Furthermore, a negative relation was found between the diffusion of responsibility mechanism and the avoidant personality trait. A positive relation was found between the attribution of blame mechanism and the narcissistic, antisocial, sadistic, negativistic and paranoid personality traits. A positive relation was found between the dehumanisation mechanism and the depressive, antisocial, sadistic, negativistic, borderline and paranoid personality traits, while a negative relation was found between this mechanism and the compulsive personality trait. Last of all, a positive relation was found between the total score for moral disengagement and the dependent, narcissistic, antisocial, sadistic, negativistic, masochistic, borderline and paranoid personality traits. No significant associations were found in our sample of offenders between the euphemistic labelling and the distortion of consequences mechanisms and dysfunctional personality traits. It should be pointed out that the dysfunctional personality traits most strongly associated with moral disengagement were the antisocial, sadistic and negativistic traits, in respect of which a positive relation was found with five of the eight mechanisms of moral disengagement.

Table 4.

Correlations among dysfunctional personality patterns, Dark Triad and mechanisms of moral disengagement.

Mechanisms of moral disengagement	Moral justification	Euphemistic language	Advantageous comparison	Displacement of responsibility	Diffusion of responsibility	Distortion of consequences	Attribution of blame	Dehumanisation	Total MMDS
Dysfunctional personality patterns
Schizoid	.014	−.033	.305*	−.157	−.207	−.041	.012	.235	.040
Avoidant	.099	.062	.191	.073	−.2 56*	.088	−.039	.156	.077
Depressive	.088	.111	.434**	.019	−.066	.020	.068	.258*	.192
Dependent	.192	.146	.318*	.108	.053	.232	.120	.192	.261*
Histrionic	.251	−.063	−.139	.125	.228	.108	.147	−.021	.111
Narcissistic	.052	−.009	.222	.318*	.109	.177	.315*	.164	.269*
Antisocial	.340**	.188	.367**	−.033	−.082	.142	.263*	.389**	.305*
Sadistic	.291*	.185	.487**	.081	−.035	.113	.273*	.477**	.370**
Compulsive	−.292*	−.243	−.244	.106	.037	−.036	−.131	−.376**	−.204
Negativistic	.288*	.230	.536**	.133	.100	.078	.369**	.439**	.418**
Masochistic	.191	.091	.305*	.043	−.136	.054	.020	.217	.149*
Schizotypal	−.153	.172	.305*	.022	−.190	.003	.097	.167	.150
Borderline	.191	.179	.496**	.013	−.085	.039	.248	.288*	.269*
Paranoid	.220	.137	.470**	.157	.072	.074	.317*	.379**	.356**
Dark Triad
Machiavellianism	.170	.244	.330**	.083	.039	.142	.242	.368**	.314*
Narcissism	.167	.130	.199	.407**	.062	.368**	.075	.062	.282*
Psychopathy	.395**	.412**	.398**	.055	−.011	.158	.295*	.494**	.420**

Note: MMDS = Mechanisms of Moral Disengagement Scale.

*p< .05. **p< .01.

In respect of the associations between the three constructs of the Dark Triad of personality and the mechanisms of moral disengagement, a positive relation was found between Machiavellianism and the advantageous comparison and dehumanisation mechanisms. A positive relation was found between narcissistic personality trait and two mechanisms, namely displacement of responsibility and distortion of consequences. Psychopathic personality trait was the one with the largest number of positive relations – five of the eight mechanisms – with moral disengagement: moral justification, euphemistic labelling, advantageous comparison, attribution of blame and dehumanisation. Last of all, a significant association was found between the three traits that comprise the Dark Triad of personality and the total score for moral disengagement, the strongest relation being with the psychopathic trait.

Differences between recidivist offenders and non-recidivist offenders in dysfunctional personality patterns, Dark Triad and moral disengagement

Table 5 shows the differences between recidivist and non-recidivist offenders. As regards the dysfunctional patterns of personality, a significant difference was found in relation to the antisocial trait, with higher scores for the recidivist group. The size of the effect was medium (95% CI [0.50, 0.70]). No significant differences were found in respect of the variables of the Dark Triad of personality, although higher scores were observed in the psychopathic personality trait for the group of recidivists. In respect of the mechanisms of moral disengagement, higher scores were observed in all the mechanisms and in the total score for the recidivists, although the only significant difference was in the advantageous comparison mechanism. The size of the effect of this difference could be considered medium (95% CI [0.50, 0.70]).

Table 5.

Descriptive statistics and differences between recidivist and non-recidivist participants.

	N = 60		d	t	p
	Recidivists (N = 27)	Non-recidivists (N = 32)
	M (SD)	M (SD)
Dysfunctional personality patterns
Schizoid	7.88 (4.50)	8.09 (4.14)	−0.04	−0.18	.856
Avoidant	5.92 (5.09)	7.37 (4.41)	−0.30	−1.17	.224
Depressive	9.32 (7.20)	8.20 (5.84)	0.17	0.66	.509
Dependent	7.36 (5.36)	7.14 (4.87)	0.04	0.16	.871
Histrionic	15.68 (4.21)	14.77 (4.05)	0.22	0.84	.403
Narcissistic	14.92 (3.97)	14.89 (3.81)	0.00	0.03	.973
Antisocial	11.92 (4.59)	8.49 (4.90)	0.70	2.74	.008
Sadistic	9.88 (4.82)	8.31 (4.36)	0.34	1.31	.195
Compulsive	17.28 (3.87)	18.86 (3.98)	−0.40	−1.52	.132
Negativistic	10.52 (6.07)	9.14 (5.38)	0.24	0.92	.358
Masochistic	5.00 (5.03)	4.91 (4.61)	0.01	0.06	.946
Schizotypal	6.72 (6.58)	7.20 (4.48)	−0.08	−0.30	.760
Borderline	8.92 (6.60)	7.51 (5.43)	0.23	0.91	.367
Paranoid	8.80 (6.42)	10.06 (5.44)	−0.21	−0.81	.417

	N = 60
	Recidivists (N = 25)	Non-recidivists (N = 35)
	M ( SD )	M(SD)
Dark Triad
Machiavellianism	2.94 (0.72)	3.03 (0.76)	−0.12	−0.49	.626
Narcissism	2.40 (0.52)	2.58 (0.54)	−0.33	−1.33	.188
Psychopathy	2.38 (0.46)	2.18 (0.57)	0.38	1.39	.167

	N = 61
	Recidivists	Non-recidivists
	(N = 25)	(N = 36)
	M (SD)	M (SD)	d	t	p
Moral disengagement
Moral justification	8.92 (3.08)	7.75 (3.24)	0.37	1.41	.163
Euphemistic language	6.52 (2.29)	5.88 (2.38)	0.27	1.03	.306
Advantageous comparison	9.28 (3.97)	7.22 (3.24)	0.56	2.22	.030
Displacement of responsibility	9.36 (3.16)	8.44 (3.93)	0.25	0.96	.338
Diffusion of responsibility	12.36 (2.94)	12.00 (2.96)	0.12	0.46	.642
Distortion of consequences	8.36 (3.06)	7.33 (2.98)	0.34	1.30	.197
Attribution of blame	9.48 (2.94)	8.30 (3.14)	0.38	1.39	.168
Dehumanisation	8.64 (3.32)	8.11 (4.01)	0.14	0.54	.590
Total MMDS	72.92 (15.76)	65.05 (17.32)	0.47	1.80	.76

Note: MMDS = Mechanisms of Moral Disengagement Scale.

Risk of recidivism or non-recidivism

A binary logistic regression analysis was carried out in order to evaluate the prediction accuracy of the specified predictor variables of the present study in respect of recidivism. The results of the regression are shown in Table 6. The specified combination of predictor variables was found to be significant (χ² = 26.24, p = .02, Nagelkerke R² = .48). Results indicated that the predictive power of this model was driven by four variables: the use of violence in the crime for which the individual has been convicted, the antisocial personality trait and two moral disengagement mechanisms: advantageous comparison and dehumanisation. On the one hand, the analysis indicated that individuals who have been convicted for a crime involving the use of violence are 0.09 times more likely to reoffend. Furthermore, it was observed that the higher the score in the antisocial personality trait, the greater the likelihood of recidivism. In respect of moral disengagement mechanisms, advantageous comparison mechanism also increases the likelihood of recidivism. By contrast, the higher the score in the dehumanisation mechanism, the lower the likelihood of recidivism.

Table 6.

Results of logistic regression analysis predicting recidivism.

	B	SE	Wald’s χ²	p	e b	95% CI
The crime involves violence	2.36	0.86	7.47	.00	0.09	[0.01, 0.51]
Sex	−1.35	1.04	1.67	.19	0.25	[0.03, 2.00]
Age	0.06	0.03	3.30	.06	1.06	[0.99, 1.37]
Dysfunctional personality patterns
Antisocial	0.25	0.12	4.45	.03	1.29	[1.01, 1.64]
Dark Triad personality
Machiavellianism	−0.08	0.05	2.17	.14	0.91	[0.82, 1.02]
Psychopathy	0.00	0.10	0.00	.95	1.00	[0.82, 1.22]
Moral disengagement
Moral justification	−0.05	0.15	0.10	.74	0.95	[0.70, 1.28]
Euphemistic language	0.05	0.21	0.06	.79	1.05	[0.68, 1.62]
Advantageous comparison	0.38	0.18	4.41	.03	1.47	[1.02, 2.11]
Displacement of responsibility	0.11	0.13	0.65	.41	1.11	[0.85, 1.46]
Diffusion of responsibility	−0.12	0.15	0.03	.85	0.97	[0.71, 1.31]
Distortion of consequences	0.10	0.14	0.50	.47	1.10	[0.83, 1.46]
Attribution of blame	0.02	0.15	0.02	.87	1.02	[0.76, 1.37]
Dehumanisation	−0.35	0.17	4.02	.04	0.70	[0.49, 0.99]
Constant	−4.89	3.96	1.53	.216	0.007

Note: CI = confidence interval.

^bindicates the strength of the relationship. The further away from 1 it is, the stronger the relationship.

Discussion

The primary aim of this research study is to examine the relationship between dysfunctional personality traits – as a whole and specifically those that comprise the Dark Triad – and moral disengagement mechanisms in a prison population. In most of the research carried out in this area the association between the Dark Triad of personality and moral disengagement mechanisms has only focused on the wider community (Sijtsema et al., 2019), adolescents (Gini et al., 2014; Hymel & Perren, 2015) or the university population (Risser & Eckert, 2016). To date, no specific research study has been carried out involving prison inmates.

Descriptive analysis of the study indicates a greater prevalence of dysfunctional personality traits in the female participants of the study than in their male counterparts. In this respect, and taking into account that fewer women than men took part in the study, significant differences were found when comparing the means of the two groups in most of the dysfunctional personality traits, except for the dependent, antisocial and narcissistic traits. As regards the constructs of the Dark Triad of personality and the mechanisms of moral disengagement, no significant differences were found between the female and male participants of the sample. Very few research studies have analysed the differences between men and women in the prison population in relation to morality, which makes it difficult to compare the results in this respect.

Concerning the differences between participants convicted for crimes involving the use of violence and those convicted for crimes not involving the use of violence, we found higher scores in the schizoid personality disorder in those participants convicted for crimes involving the use of violence. The schizoid personality disorder is described by the APA (2013) as a personality trait characterised by interpersonal difficulties, low self-esteem, detachment, flat affect and emotional coldness. Overall, this personality trait is not associated with violent conduct. Although there is no research on the subject, a possible explanatory hypothesis could be that due to its asocial component and the fact that the individual may feel overwhelmed or rejected by society, it can lead to violent outbursts that may be even more extreme than in other personality disorders (Esbec & Echeburúa, 2010). Moreover, narcissistic and compulsive personality traits were significantly higher in the participants convicted for crimes not involving the use of violence. A possible explanation for that is hence both the narcissistic and the compulsive trait would appear to show greater adaptability in our sample of offenders, while the remaining personality traits are presented in a more dysfunctional way, being more present in those participants with violent offences. Our results seem to point in the same direction as other studies on personality disorders, where these traits conceptualised as dysfunctional increasingly show a greatly promoted and accepted pattern in Western society (Caparrós & Villar, 2013). In these studies, these traits have shown a more adaptive pattern, either in undergraduate students (Caparrós & Villar, 2013) or in a substance use disorders sample (Masferrer & Caparrós, 2017). It must be highlighted that in our study, the participants who have been convicted for crimes involving the use of violence scored higher, although not significantly, in terms of the antisocial personality trait, in line with the findings of previous research projects (Gunderson et al., 2011). No differences were found between both groups regarding the constructs of the Dark Triad of personality in relation to the mechanisms of moral disengagement, although it was observed that participants convicted for crimes involving the use of violence scored higher in most of the mechanisms of moral disengagement, in accordance with previous findings that have associated antisocial conduct with moral disengagement (Sijtsema et al., 2019). Furthermore, we observed higher scores analysing the narcissistic personality of the Dark Triad in participants convicted for crimes not involving the use of violence, offering more evidence that narcissistic personality could be set apart from the so-called dark personalities so strongly associated with antisocial behaviour, shown to be more adapted in the current sample of offenders.

When examining the relationship between recidivism and dysfunctional personality traits, only the antisocial personality disorder was found significant in those recidivist participants. These results empirically support the research studies carried out by Obermann (2011) and Risser & Eckert (2016), in which reoffenders scored higher in all the mechanisms of moral disengagement. Furthermore, this difference is justified by and matches with the definition of the antisocial personality disorder formulated by the APA (2013), which refers to the frequent breach of social norms, manifested in actions that repeatedly lead to arrest. Concerning the constructs of Dark Triad of personality, there were no differences between recidivist and non-recidivist offenders. Nevertheless, the recidivist group scored higher in the psychopathic personality construct, matching both the definition formulated by the APA (2013), which highlights the lack of remorse and the rationalisation of the harmful behaviour carried out, and with the more classic definitions of psychopathy formulated by Cleckley (1976) and Hare (1991).

Research studies have been carried out that analyse the relationship between Dark Triad of personality and the mechanisms of moral disengagement (Egan et al., 2015; Risser & Eckert, 2016; Sijtsema et al., 2019). However, these studies have not analysed their association with the entire set of the dysfunctional personality traits, and, furthermore, they have not focused on the prison population, which is one of the main objectives of this study. In the present study, advantageous comparison and dehumanisation were the moral disengagement mechanisms most strongly associated with the dysfunctional personality patterns. The advantageous comparison mechanism involves viewing oneself more favourably than others, perceiving one’s own criminal behaviour and actions as less morally reprehensible than those of others. Dehumanisation is also related to dysfunctional perception and cognition processes that lead individuals to deny the human attributes of others in order to justify carrying out harmful or immoral actions against them. In addition, the dysfunctional personality traits most significantly associated with moral disengagement mechanisms are the antisocial, sadistic and negativistic traits. The negativistic personality has not been associated with dark personalities. Nonetheless, an analysis of its characteristics as conceived by Millon et al. (2007) and Millon (2011) shows that some of them, such as hostility, argumentativeness, criticism of and an irrational disdain for figures of authority, pessimism, irritability and impulsiveness, may be related to and match some of the characteristics of dark personalities. Previous research studies have found an association between the antisocial personality trait and moral disengagement (Obermann, 2011; Risser & Eckert, 2016). However, in our research study, in addition to this relation we also found that the sadistic personality trait was strongly associated with moral disengagement. This enables us to support the recent proposal to add a fourth personality to the so-called Dark Triad of personalities. Accordingly, the triad would become a tetrad, as proposed by Chabrol et al. (2015). In congruence with the existing literature, our results indicate that these socio-cognitive processes of reasoning and moral justification may act as a link between dysfunctional personality traits and antisocial behaviour (Caprara et al., 2014; Sijtsema et al., 2019). Moreover, these thought patterns may help to justify aggressive and antisocial behaviour in the prison population, increasing the likelihood of recidivism, in line with the results of the binary logistic regression analysis carried out in this research.

In general, our results corroborate the findings of previous research studies, according to which the three personalities that comprise the Dark Triad of personality are related in different ways to moral disengagement mechanisms (Egan et al., 2015; Shulman et al., 2011; Sijtsema et al., 2019). More specifically, we have found that the psychopathic personality is the one that has the strongest positive relation with moral disengagement mechanisms. Our findings show significant relations with five of the eight mechanisms: moral justification, euphemistic labelling, advantageous comparison, attribution of blame and dehumanisation. The findings of previous research studies are contradictory in respect of whether or not the psychopathic personality disconnects from the mechanisms used for moral reasoning (DeLisi et al., 2014; Marshal et al., 2018; Shulman et al., 2011). However, according to these findings and on the basis of an analysis of the affective nature characterised by the lack of remorse and the absence of feelings of guilt, along with the inability to accept responsibility for one’s actions, characteristic of this personality pattern (Hare, 1991), it can be argued that morality is compromised in our sample of prison inmates and in the current configuration of the psychopathic personality. The Machiavellian personality is positively associated with two of the eight mechanisms of moral disengagement that are also related to the psychopathic personality: the advantageous comparison and, very significantly, the dehumanisation mechanism. Machiavellian personality traits include, among others, the externalisation of guilt and the use of manipulative interpersonal strategies. Conforming to the analyses, these moral disengagement mechanisms may be used in relationships for the purpose of self-protection, leading to a form of reasoning aimed at safeguarding one’s identity and justifying the means for carrying out harmful actions. Last of all, the narcissistic personality is associated with two of the eight mechanisms of moral disengagement that are not related to either the psychopathic personality or the Machiavellian personality. In this respect, positive associations were observed with two mechanisms of moral disengagement: displacement of responsibility and distortion of consequences. We observed some differences in the direction of the narcissistic personality compared to the Machiavellian and psychopathic personalities, insofar as the moral disengagement mechanisms related to narcissism are aimed at preserving one’s moral integrity rather than justifying harmful behaviour carried out against others. These findings are in line with Hoffman et al. (2011), stating that people with narcissistic traits tend to shift responsibility, attributing successes to oneself and instead failures to the group in general. Overall, the results on the Dark Triad and moral disconnection tell us that narcissism follows a unique pattern; in future research, to obtain a more precise understanding, we suggest further study on the multiple facets of narcissism.

In the predictive analysis of recidivism, our findings indicate that a background of violent antisocial behaviour is the most powerful variable when it comes to predicting the risk of reoffending. Furthermore, the antisocial personality trait is also shown to be a predictive variable of recidivism. These findings match those of previous research studies (Coid et al., 2007; Eher et al., 2016; Lund et al., 2013), in which both variables have been studied as high-risk factors for recidivism. In relation to the mechanisms of moral disengagement, our study found that the advantageous comparison mechanism was a significant predictor of recidivism, showing that the individual’s perception of their crime(s) being less serious than those of others may constitute a risk factor in terms of reoffending. Nonetheless, our analysis found that the moral disengagement mechanism of dehumanisation is a significant factor in predicting recidivism, but in an inverse manner. This result may be due to the fact that the dehumanisation mechanism involves a more socio-cognitive process of perceiving others and is less behavioural in terms of antisocial conduct. In future research studies, it will be necessary to identify the underlying mechanisms involved in each of the mechanisms of moral disengagement in order to better understand the construct of moral disengagement in an individual’s reasoning of their antisocial behaviour.

Our study has some limitations. First, the sample size is small. In this respect, it is important not to generalise but rather to restrict our results to the present sample. An analysis of the problem with a larger sample of participants is highly advisable. However, it must be pointed out that our sample size is similar to that in other research studies in the same area involving offenders. Furthermore, while most previous research studies concerning the prison population only include male prisoners, our sample does include women, although there are significantly fewer of them. Another limitation is the evaluation of the problem by means of self-report tools. In future research, it would be advisable to use mixed study methodology – with the inclusion of more qualitative methods – rather than exclusively quantitative methods. Nonetheless, it is necessary to consider that the complexity of the organisational system (sample access restrictions and protocols as well as the time cost for each evaluation) will make it difficult to obtain these qualitative results.

In conclusion, our research contributes to the existing literature by incorporating new results regarding the relationship between the entire set of dysfunctional personality traits, specifically those of the Dark Triad, and the mechanisms of moral disengagement in a prison population sample. Specifically, it finds both positive and negative connections between most of the dysfunctional personality traits and the mechanisms of moral disengagement, highlighting the important connection with the antisocial and sadistic personality traits. Moreover, in respect of the Dark Triad, it is important to note the special importance of the psychopathic personality and its positive relation with five of the eight mechanisms of moral disengagement. This study contributes to obtaining tools and knowledge for analysing the risk of recidivism, as well for implementing specific intervention programmes that address dysfunctional personality traits and the moral reasoning of the prison population. Understanding the configuration of dysfunctional personality traits and the specific socio-cognitive mechanisms that influence antisocial behaviour will enable us to cover the need to establish specific treatments for the prevention of recidivist behaviour as well as for the facilitation of an adequate social reintegration. Additionally, this study represents an important contribution as it highlights the importance of socio-cognitive and moral processes in the explanation of recidivist behaviour. These aspects should be a key factor in treatment programmes inside and outside penitentiary institutions. In future research, it will be necessary to conduct further research in order to understand the multiple nuances of the darkest personality traits and in order to determine the associated personalities.

Ethical standards

Declaration of conflicts of interest

Glòria Brugués Català has declared no conflicts of interest.

Beatriz Caparrós Caparrós has declared no conflicts of interest.

Ethical approval

All procedures performed in studies involving human participants were in accordance with the ethical standards of the institutional and/or national research committee (Penitentiary Services of the Department of Justice of the Government of Catalonia, Rehabilitation Service on 12/02/2018, with record number 0310/836/2018) and with the 1964 Helsinki declaration and its later amendments or comparable ethical standards.

Informed consent

Informed consent was obtained from all individual participants included in the study.