The dark triad and normal personality traits

https://doi.org/10.1016/j.paid.2005.07.006Get rights and content

Abstract

Machiavellianism, Narcissism and Psychopathy are often referred to as the ‘dark triad’ of personality. We examined the degree to which these constructs could be identified in 82 persons recruited from the general population, predicting that the dark triad would emerge as a single dimension denoting the cardinal interpersonal elements of primary psychopathy. We expected the primary psychopathy dimension to correlate negatively with Agreeableness (A) and Conscientiousness (C), whereas secondary psychopathy would be associated with Neuroticism (N). The negative correlation was found between primary psychopathy and A, but not with C. While the predicted correlation between secondary psychopathy and N was found, N was also positively associated with primary psychopathy and Machiavellianism. Factor analysis revealed that all measures of the dark triad loaded positively on the same factor, upon which A loaded negatively. Secondary psychopathy loaded positively on a second factor, together with N and (negatively) with C. These findings reiterate the distinguishing properties of secondary psychopathy, impulsivity and anti-social behaviour relative to primary psychopathy. Thus, even in the general population, the dark dimension of personality can be described in terms of low A, whereas much of the anti-social behaviour in normal persons appears underpinned by high N and low C.

Introduction

The term ‘Dark Triad of Personality’ refers to three interrelated higher-order personality constructs—psychopathy, narcissism and Machiavellianism (Paulhus & Williams, 2002). The majority of work conducted on psychopathy builds upon observations by Cleckley (1941/1988), operationalised in Hare’s revised Psychopathy Checklist (PCL-R; Hare, 1991). The PCL-R and similar measures (e.g., Levenson’s self-report measure of psychopathy (LSRP; Levenson, Kiehl, & Fitzpatrick, 1995)) measure two facets of psychopathy. Factor 1 reflects primary psychopathy (e.g., selfishness, callousness, lack of interpersonal affect, superficial charm and remorselessness), factor 2 measures anti-social lifestyle and behaviours, and is akin to secondary psychopathy. It should be noted that researchers now propose three facets to core psychopathy; an arrogant and deceitful interpersonal style, deficient affective experience, and an impulsive and irresponsible behavioural style (Cooke & Michie, 2001).

Most research into psychopathy involves forensic populations such as prisoners and mentally disordered offenders. However, not all persons with primary and secondary psychopathy are in custody, and a literature has gradually emerged examining psychopathy-like traits in the general population (Board and Fritzon, 2005, Ross et al., 2004). This suggests that if psychopathy is a trait, it should be apparent in non-offenders, and that it may even confer some kind of social advantage (Levenson, 1992). Examining psychopathy in the general population overcomes the sample bias of only seeing persons from prison settings, who are essentially homogenous regarding socio-economic background and who are competitively disadvantaged intellectually, socially, and interpersonally. It is probable that the majority of institutionalised offenders are more inclined to secondary than primary psychopathy, with different dispositional mechanisms driving their transgressive behaviour (Lykken, 1995, McHoskey et al., 1998).

Machiavellianism (MACH) refers to interpersonal strategies that advocate self-interest, deception and manipulation. Christie and Geis (1970) examined the extent to which people use qualities such as deceit, flattery and emotional detachment to manipulate social and interpersonal interactions. While high MACHS are perceived to be more intelligent and attractive by their peers (Cherulnik, Way, Ames, & Hutto, 1981), MACH does not correlate with intelligence or measures of success in modern life such as income or status (Ames and Kidd, 1979, Hunt and Chonto, 1984). In experimental settings high MACHS frequently outperform low MACHS, whether this be bargaining and alliance forming (Christie & Geis, 1970), or assuming leadership in group situations (Cherulnik et al., 1981). As persons high in MACH are likely to exploit others and less likely to be concerned about other people beyond their own self-interest, MACH is predictably negatively correlated with empathy (Barnett & Thompson, 1985). Given these findings, one would expect a relationship between MACH and primary psychopathy.

The concept of narcissism derives from the psychodynamic formulations such as a pathological form of self-love (Freud, 1914), or personality development, whereby “narcissistic wounds” sustained in childhood may lead to an arrest in development and increased shame-driven rage (Kohut, 1977). It has been argued that the construct of narcissism is compromised by the contrast between vague psychoanalytic terminology and theory, and more observable elements of the concept (Bradlee and Emmons, 1991, Watson and Morris, 1991). However as a means of encapsulating the behavioural grandiosity and perceived entitlement of an individual, the concept of narcissism is a very useful concept. One commonly used scale to assess narcissism is the Narcisssistic Personality Inventory (NPI; Raskin & Hall, 1979). The NPI measures persistent attention seeking, extreme vanity, excessive self-focus, and exploitativeness in interpersonal relationships (Millon & Davis, 1996), and comprises four factors: Exploitativeness/Entitlement, Leadership/Authority, Superiority/Arrogance and Self-Absorption/Self-Admiration (Emmons, 1984).

The constructs of the dark triad correlate with each another singularly and in combination; (Hare, 1991, Skinner, 1988, Smith and Griffith, 1978, McHoskey, 1995). Such findings led McHoskey et al. (1998) to argue that for the general population MACH is a global measure of psychopathy comparable to primary and secondary psychopathy, this comparability being confounded due to the different paradigms and theoretical orientations of clinical and differential psychology. McHoskey et al. found that when secondary psychopathy was controlled for, primary psychopathy remained associated with narcissism (r = 0.46). Paulhus and Williams (2002) also investigated the dark triad in the general population and found considerable overlap between the constructs, although the scale used to measure psychopathy (the SRP-III; Hare, 1985) did not distinguish between primary and secondary psychopathy. The scales of the dark triad were correlated with Agreeableness, and revealed negative correlations of −0.36, −0.47 and −0.25 for narcissism, psychopathy and MACH, respectively. No other dimension of the Big Five captured the constructs of the dark triad. This is perhaps surprising, as MACH is typically negatively correlated with anxiety (Wiggins & Pincus, 1989), as is primary psychopathy (Fehr, Samson, & Paulhus, 1992), so one would expect Neuroticism (N) to be a negative associate of the construct.

The current study investigated to what extent MACH, primary psychopathy, secondary psychopathy and narcissism reflect the same underlying construct, and to examine the extent to which normal dimensions of personality indexed by a brief measure of the Big Five could capture the constructs of the dark triad. We expected to find low A and low C associated with higher scores on each of the dark triad dimensions, and to load on a single dark triad factor. We expected N to be unrelated to the core traits of psychopathy, but to be associated with secondary psychopathy.

Section snippets

Sample and procedure

The sample in this study were recruited opportunistically from the general population using a ‘snowball’ system, whereby a starter sample were further asked recruit people from their environment who would be willing to take part in this study. Although this meant that not all participants were in direct contact with the researcher, this form of recruitment ensured that a diverse selection of the general population, was enlisted. Eighty-two persons were recruited, their mean age being 29; of the

The revised NEO Five Factor Inventory (NEO-FFI-R; McCrae & Costa, 2004)

The NEO-FFI-R is a revised version of the NEO-FFI, in which 14 items of the NEO-FFI were changed. This was done to increase the correlation between the shortened version of the scale and overall NEO-PI-R scores, diversify item content by selecting items from underrepresented facets, and to increase the intelligibility of the items. This modification was required following the discovery that not all scales of the NEO-FFI had equally stable structures (e.g. Egan, Deary, & Austin, 2000). The

Statistical analysis

Correlations were calculated between scores on the dark triad measures and the NEO-FFI-R. The reliability of these measures was calculated using Cronbach’s alpha. To simplify the relationships between measures of personality and the dark triad, and to examine whether all three scales of the dark triad reflected the same underlying construct, principal components analysis with Varimax rotation of the derived factors was calculated.

Results

Table 1 presents means, standard deviations and internal reliabilities for the measures used in the study. Reliabilities were adequate to good, but some measures clearly had greater internal reliability than others (e.g. N and narcissism vs. secondary psychopathy). While not a main hypothesis, independent sample t-tests examined whether there were any sex differences for the measures. Results showed that males were higher than women for C (t(79) = 2.79; p < 0.05), whereas females were significantly

Discussion

The current study examined the relationship between the constructs of the dark triad and how they fitted into the five factor space of personality. Previous studies indicated that there is considerable overlap between MACH-IV, primary psychopathy, secondary psychopathy and narcissism, and that more specifically that MACH-IV is a global measure of psychopathy (McHoskey et al., 1998). This could reflect the varied nature of the MACH-IV, which contains items measuring dispositions as well as

References (41)

  • P.M. Bradlee et al.

    Locating narcissism within the personal circumplex and the five factor model

    Personality and Individual Differences

    (1991)
  • P.D. Cherulnik et al.

    Impressions of high and low Machiavellian men

    Journal of Personality

    (1981)
  • R. Christie et al.

    Studies in Machiavellianism

    (1970)
  • H. Cleckley

    The mask of sanity

    (1988)
  • D.J. Cooke et al.

    Refining the construct of psychopath: towards a hierarchical model

    Psychological Assessment

    (2001)
  • V. Egan et al.

    Personality traits, personality disorders and sensational interests in mentally disordered offenders

    Legal and Criminological Psychology

    (2003)
  • R.A. Emmons

    Factor analysis and construct validity of the Narcissistic personality inventory

    Journal of Personality Assessment

    (1984)
  • B. Fehr et al.

    The construct of Machiavellianism: twenty years later

  • S. Freud

    On Narcissism: an introduction

  • M.D. Gupta

    Effects of age and family structure on Machiavellianism

    Indian Journal of Current Psychological Research

    (1986)
  • Cited by (0)

    View full text