A test of the construct validity of the Triarchic Psychopathy Measure in an Italian community sample

Highlights

• The Triarchic model tries to reconcile differing conceptions of psychopathy.

• This study examined reliability as well validity of the TriPM on an Italian community.

• Disinhibition was associated with indices of the behavioral deviance and distress.

• Boldness was associated with indices of adaptive function as well as maladjustment.

• Meanness was associated with indices of coldheartedness and antagonism.

Abstract

The Triarchic Psychopathy Measure (TriPM) is a relatively new self-report measure of the basic traits thought to be associated with psychopathy: Boldness, Disinhibition and Meanness. The TRiPM was administered to 286 Italian individuals along with the Psychopathy Personality Inventory-Revised as well as measures of negative affect (anxiety, depression, stress, and hopelessness) and normal-range personality traits. The Italian TriPM showed excellent reliability and was minimally influenced by age and education. Some interesting differences emerged between males and females even though the overall pattern of correlations for TriPM scales with many criterion measures was fairly consistent across gender. TriPM Boldness was associated with indices of adaptive function as well as maladjustment. TriPM Disinhibition was most strongly associated with indices of the behavioral deviance, as well as internalizing problems. Scores on TriPM Meanness were most strongly associated with scales indexing coldheartedness, Machiavellianism, and antagonism. These findings broaden the nomological network of the Triarchic model of psychopathy.

Introduction

In the last few years, unresolved issues in the study of psychopathy have been revisited. First, evidence supporting a dimensional as opposed to typological view of psychopathy raised questions about the unitary versus configural nature of the construct. Second, the role of adaptive versus maladaptive features in psychopathy were discussed. Third, it was debated whether antisocial or criminal behavior should be considered a defining feature of psychopathy or instead as a potential consequence or expression of it. Other discussion points included the need to separate symptom indicators from clinical outcomes in order to avoid criterion contamination and limits on the applicability of crime-oriented diagnostic criteria in non-forensic settings (Lilienfeld et al., in press, Miller and Lynam, in press, Patrick and Drislane, in press).

Patrick, Fowles, and Krueger (2009) advanced a Triarchic model of psychopathy as a point of reference for reconciling differing historic conceptions of psychopathy (and alternative approaches for assessing it), and for addressing the aforementioned points of contention. According this model, psychopathy encompasses these three distinct but intersecting phenotypic tendencies: disinhibition, boldness, and meanness.

Disinhibition is defined as the general proneness toward externalizing problems and comprises traits such as impulsivity, irresponsibility, and hostility. It represents the nexus of impulsivity and negative emotionality (e.g., Krueger, Markon, Patrick, Benning, & Kramer, 2007). Boldness entails traits of emotional stability, dominance, self-assurance, and social efficacy (Benning et al., 2003, Kramer et al., 2012). Lastly, Meanness (for which referents include the “callous-unemotional” factor of child psychopathy and the “callous-aggression” subdomain of the externalizing spectrum in adults (Frick and Hare, 2001, Krueger et al., 2007)) encompasses tendencies toward manipulativeness, lack of empathy, inability to form close attachments with others, and cruelty.

Patrick’s model allows for understanding psychopathy in its different manifestations (i.e., criminal and noncriminal, unsuccessful and successful, primary and secondary) and therefore may serve as a vehicle for linking conceptions of psychopathy per se to broader dimensional models of normal personality and psychopathology (Patrick & Drislane, in press).

The Triarchic Psychopathy Measure, a 58-item self-report inventory that assesses Boldness (19 items), Disinhibition (20 items) and Meanness (19 items) through separate targeted subscales, represents the most common way to operationalize phenotypic constructs of the Triarchic model. In studies conducted so far, the disinhibition scale was associated with measures of impulsivity and fun seeking, low levels of self-discipline, and higher degrees of emotional instability. The boldness scale has typically correlated in a positive direction with traits of narcissism, thrill seeking, extraversion, fearlessness, stress immunity and in a negative direction with measures of neuroticism. Lastly, TriPM meanness was associated with measures of Machiavellianism, low empathy, and aggressiveness (for details on development of the TriPM, as well as a review of the main results in literature, please refer to Patrick & Drislane, in press).

However, further studies are needed to extend what is known about the psychometric properties and correlates of the TriPM scales, including the potential moderating role of gender or the association between TriPM scores and educational attainment. Additionally, research is needed to establish the potential utility of the TriPM in cultural contexts different from the USA, since cultural values play a crucial role in defining the way people interpret and express their emotions and clinical symptoms (e.g., Sica et al., 2002, Sica et al., 2006).

In particular, this study was designed to examine the reliability as well construct validity of the TriPM in relation to a psychopathy-specific measure (i.e., Psychopathic Personality Inventory-Revised) and other conceptually-relevant criteria including normal-range personality traits and psychological distress (anxiety, depression, stress, hopelessness).

Since the PPI-R includes a broad factor labeled fearless-dominance (FD) directly reflecting boldness and extracted via factor analysis on the PPI-R subscales (Benning et al., 2003), we hypothesized that that scores on FD of the PPI-R (and its constituent subscales) would correlate strongly and preferentially with TriPM Boldness. On the other hand, positive relations were predicted for TriPM Disinhibition with the self-centered impulsive factor of the PPI-R (PPI-R-SCI) and its constituent subscales, given evidence that this factor indexes externalizing proneness. We further hypothesized that scores on PPI-R Coldheartedness (a measure of callousness and low empathy) would be strongly and uniquely associated with TriPM Meanness. Lastly, it was expected that also PPI-R Machiavellian Egocentricity (which indexes callous-aggressiveness) would be related somewhat to TriPM Meanness,

In addition, we predicted that anxiety, depression, stress, and hopelessness would be mainly related (in a positive direction) to disinhibition, because one of its core features is negative emotionality. On the contrary, an inverse association between psychological distress and boldness was expected since this TriPM facet is commonly related in a negative direction with measures of neuroticism.

Regarding the associations between the TriPM and basic personality traits (i.e. the Five Factor Model), we predicted that: (1) disinhibition would be strongly associated with low Conscientiousness, low Agreeableness, and high Neuroticism, since high impulsivity and emotional instability are the core features of this facet; (2) boldness would be associated with low Neuroticism and high Extraversion, and to some extent low Agreeableness, since it represents the nexus of stress immunity, thrill-adventure seeking/fearlessness, and social dominance; (3) meanness would correlate primarily with low Agreeableness and Conscientiousness, given previous findings linking meanness with lack of empathy, callousness, inability to form close attachments, and antagonism.

For all previous hypothesized associations we were also interested in determining unique associations of scores on the three Triarchic domains with the various criterion variables. For this purpose, we also performed multiple regression analyses in which scores for TriPM Boldness, Meanness, and Disinhibition scales were included as concurrent predictors of each criterion variable. Regressions coefficients from these analyses provided an index of the distinct contribution of each TriPM scale to prediction.

Previous studies showed that males obtained higher scores than females on each of the TriPM facet scales as well TriPM total scores (e.g., Sellbom and Phillips, 2013, Poy et al., 2014); as such, we expected to find similar results in the current study (see also, Grieve & Panebianco, 2013). At the same time, we speculated that the overall pattern of correlations for TriPM scales with many criterion measures would be fairly consistent across gender, as showed by previous research (e.g., Marion et al., 2013).

To explore the impact of gender on the predicted associations, a series of hierarchical linear regression models were computed using each criterion measure as the dependent variable. Therefore, scores on the three TriPM facets and gender were entered as predictors at Step 1, and the respective Gender × TriPM Domain interactions were entered at Step 2. A significant increase in R² on the second step would indicate gender differences in the relation between the criterion and the TriPM facets (see also, Poy et al., 2014).

Lastly, associations of three TriPM facets with age and education were also examined. We expected to find a negative association between age and TriPM facets, since social deviance is commonly inversely related to age (e.g., Olver & Wong, in press). On the contrary we predicted null associations between the TriPM facets and education as shown in previous studies (e.g., Lilienfeld & Widows, 2005).