Cleckley (1941) described psychopathy in terms of chronic behavioral deviance (e.g., inadequately motivated antisocial behavior, unreliability), emotional-interpersonal deficits (e.g., untruthfulness and insincerity, lack of remorse or shame), as well as features of positive adjustment (e.g., superficial charm and good intelligence, absence of “nervousness” or psychoneurotic manifestations). Psychopathy has since become one of the most widely researched personality constructs (Patrick, 2006), particularly in offender populations, due primarily to the development of Hare’s psychopathy checklist – revised (PCL-R; Hare, 1991, 2003), which was based in part on Cleckley’s criteria.
Although the Cleckley and Hare conceptualizations portray psychopathy as a unitary construct, a substantial literature asserts that psychopathy is a heterogeneous construct (for a review, see Skeem, Poythress, Edens, Lilienfeld, & Cale, 2003). Working within a psychodynamic framework, Karpman (1949) asserted that subtypes of psychopathy may be distinguished in part in terms of susceptibility to negative affectivity. He described the primary psychopath as essentially unable to experience anxiety, whereas secondary psychopathy is marked by features consistent with negative affectivity such as depression, neuroticism, and guilt. Lykken’s (1995) more modern theory employed constructs from Gray’s (1982) reinforcement sensitivity theory and asserted that primary psychopathy is distinguished by an innate fearless temperament (i.e., low sensitivity to cues of potential harm or non reward), whereas secondary psychopaths’ problem behavior may often result from over-sensitivity to cues of reward/reinforcement. Still, in Lykken’s theory secondary, but not primary, psychopaths are prone to experience anxiety and other negative emotions.
The PCL-R is the best validated measure for assessing psychopathy in offenders, and most of the PCL-R research uses a two-factor structure identified by Harpur, Hakstian, and Hare (1988) (for views of alternative three- and four-factor structures, see Hare & Neumann, 2006, and Cooke, Michie, & Hart, 2006). Factor 1 captures the core interpersonal (e.g., superficial charm, failure to accept responsibility) and affective (e.g., lack of remorse, shallow affect) features of psychopathy, whereas factor 2 captures features of a chronic socially deviant lifestyle (e.g., juvenile delinquency, irresponsibility).
Because PCL-R factors are usually correlated (about .50), it is important to control for covariance between them in examining associations with external variables due to possible suppressor effects (Hicks & Patrick, 2006). When covariance between the factors is controlled for, the PCL-R factors display divergent associations with external variables such that “the descriptive features of primary and secondary psychopathy mirror the psychometric correlates of PCL–R factors 1 and 2” (Hicks, Markon, Patrick, Krueger, & Newman, 2004, p. 277). Factor 1 is negatively associated with measures of negative emotionality, whereas factor 2 has positive associations. Factor 2 correlates positively with impulsivity and anger, and negatively with conscientiousness and constraint, whereas factor 1 tends to be unrelated to these traits (Hicks et al., 2004). Factor 2 has been associated with risk for suicidal behavior (Douglas et al., in press, Verona et al., 2005 and Verona et al., 2001), whereas factor 1 has been shown to be either unrelated (Douglas et al., in press and Verona et al., 2001) or negatively associated with suicidality (Verona et al., 2005). Factor 2 is also commonly associated positively, and more strongly than factor 1, with self-report measures of substance use problems (Hare, 1991/2003).
An alternative measure of psychopathic features is the psychopathic personality inventory (PPI; Lilienfeld & Andrews, 1996), a self-report measure whose content scales (see Measures, below) also load on two higher order factors (Benning, Patrick, Hicks, Blonigen, & Krueger, 2003; cf. Neumann, Malterer, & Newman, 2008), PPI-I (Fearless Dominance; FD), and PPI-II (Impulsive Antisociality; IA). The PPI differs in content from the PCL-R in two important ways. First, the PPI places greater emphasis on features of positive adjustment such as absence of neurotic symptoms and effective social functioning (Patrick, in press); second, the PPI lacks items that tap explicitly criminal behavior (e.g., delinquent juvenile or adult criminal behavior). Also, in contrast to the PCL-R factors, in most samples the PPI factors are uncorrelated and represent a model of psychopathy based on complementary, rather than correlated, higher order facets. For these (and other) reasons, PPI-I and PPI-II are imperfect analogues of PCL-R factors 1 and 2.
Nevertheless, in studies with offenders the external correlates of FD and IA, respectively, are similar to those described above for PCL-R factors 1 and 2. For example, associations with measures of negative emotionality have been negative for PPI-I and positive for PPI-II, (Benning et al., 2005 and Patrick et al., 2006), although Uzieblo, Verschuere, and Crombez (2007) reported that PPI-II was unrelated to scores on the behavioral inhibition scale (Carver & White, 1994), which largely indexes anxiety or negative emotionality (Poythress et al., 2008). Similarly, PPI-II correlated positively, and PPI-I negatively, with indices of suicidality (Douglas et al., in press), and PPI-II correlated positively with self-reported substance use problems, whereas PPI-I has been negatively related, or unrelated, to these indices (Benning et al., 2005 and Patrick et al., 2006).
Based on this literature, some regard psychopathy measures’ factor scores as proxy indicators for primary and secondary psychopathy (e.g., Hundt et al., 2008 and Ross et al., 2007). It is in this context that the present study examined associations between psychopathy and indices of impulsivity that may be helpful in distinguishing psychopathic subtypes.
