Gray’s two-factor model represents motivation in terms of a behavioral inhibition (BIS) and a behavioral activation system (BAS). Although this model has theoretical links to psychopathy, few studies have examined this relationship. In a sample of 326 noninstitutionalized young adults, we examined the relationship of the BIS/BAS dimensions to multiple indices of primary and secondary psychopathy. Across measures of psychopathy, primary but not secondary was related to BIS standing, whereas indices of both psychopathic dimensions demonstrated robust, positive relationships to the BAS. Generally, results support Newman, MacCoon, Vaughn, and Sadeh (2005) distinction between primary and secondary psychopathy on the basis of the BIS. However, positive associations of all psychopathy measures with BAS indices emphasize the role of a common BAS in psychopathy.
Based on the psychobiology of animal learning, Gray, 1970 and Gray, 1987 proposed a theoretical model encompassing two primary motivational systems: the behavioral inhibition system (BIS) and the behavioral activation system (BAS). The BIS represents apprehensive motivation and is sensitive to conditioned signals of punishment, frustrative nonreward and novelty. In contrast, the BAS is appetitive and sensitive to conditioned signals of reward and nonpunishment. At the trait level, Gray’s (1987) two factors of BIS and BAS translate into anxiety and impulsivity, respectively. A theory of normal adaptation, Gray’s model is drawing increased interest from psychopathy researchers. A personality disorder, psychopathy has been the subject of considerable theorizing vis-à-vis Gray’s original model.1
Although three-factor (Cooke & Michie, 2001) and four-facet (Hare, 2003) models of psychopathy have been recently advanced, factor analytic studies of psychopathy measures have generally supported a two-factor model (Benning et al., 2003, Hare, 1991, Harpur et al., 1989 and Lynam et al., 1999). Broadly, these dimensions represent the theoretical distinction between primary and secondary psychopathy (Levenson et al., 1995 and Brinkley et al., 2001). Though primary and secondary psychopathy remain open constructs, primary psychopathy represents the core emotional deficits and interpersonal manipulation in psychopathy, whereas secondary psychopathy may represent nonessential but associated characteristics, including antisociality and neurotic tendencies (Karpman, 1941). Using multiple measures of each construct, we examine the relationship of Gray’s two-factor model of BIS and BAS in relation to primary and secondary psychopathy.
2. Gray’s original BIS/BAS and psychopathy
Fowles (1988), building on Gray’s initial formulation, hypothesized that both an overactive BAS and an underactive BIS are related to “behavioural excess, in the sense of doing things that potentially lead to trouble” (p. 421). An overactive BAS would lead to more frequent approach responses irrespective of potential punishments. Similarly, an underactive BIS would result in more frequent approach responses due to a person’s lower sensitivity to punishment that, in turn, leads to reduced inhibition. Thus, high-impulsive and low-anxious personalities are both prone to risky and disinhibited behavior in approach-avoidance conflicts. Characterized by disinhibition and risk-taking, psychopathy has been hypothesized by Lykken (1957) to result from low fear, as a core emotional marker. Although fear and anxiety represent different constructs (Gray, 1987), Lykken (1995) used Gray’s model for underpinning differences between primary and secondary psychopathy. Lykken suggests that primary psychopathy is associated with a hyporeactive, weak BIS and normal (i.e., average) BAS. Fowles (1980) similarly points to the role of the BIS in primary psychopathy. In contrast, Lykken (1995) views secondary psychopathy as resulting from an overactive BAS, but normal BIS.
Following these theorists, Newman et al. (2005) proposed that long-standing operationalizations of primary and secondary psychopathy using the Psychopathy Checklist-Revised (PCL-R; Hare, 1991) combined with scores on the Welsh Anxiety Scale (WAS) might be supported by differential associations between psychopathic groups and Gray’s BIS and BAS dimensions. Using a median split on the WAS, Newman et al. (2005) classified high PCL-R and low WAS scorers as primary psychopaths. In contrast, high PCL-R and high WAS scorers were classified as secondary psychopaths. Compared with other offenders, primary psychopaths exhibited an underactive BIS and normal BAS; secondary psychopaths exhibited an overactive BAS and modestly overactive BIS. However, unpredicted findings for a weakened BIS in the secondary psychopathy control group and overactive BIS in the primary psychopathy comparison group may point to a methodological weakness using anxiety, in part, to determine psychopathic groups. Nonetheless, we know of no other studies specifically addressing the role of the BIS and BAS in psychopathy.
3. Current study
Although the PCL-R remains the gold-standard in psychopathy measurement, researchers have pointed to the “… strong need for empirical research to address the validity of alternative methods for identifying primary and secondary psychopathy” (p. 320; Newman et al., 2005). Because any single measure of a construct is contaminated by measurement error (Cronbach & Meehl, 1955), we used a multi-measure approach to assess the constructs of primary and secondary psychopathy. Similarly, we also used a multi-measure approach in determining individual differences in standing on the BIS/BAS constructs. For both sets of measures (e.g., psychopathy and the BIS/BAS), we used factor analysis to determine the latent constructs of interest. In an undergraduate sample, Carver and White’s BIS and BAS scales and scales from the Sensitivity to Punishment and Sensitivity to Reward Questionnaire (SPSRQ) were included along with the State-Trait Anxiety Inventory (STAI) to represent Gray’s BIS/BAS model. Rather than representing differences in primary and secondary psychopathy by separating high-psychopathy individuals using the WAS (or another measure of anxiety and neuroticism), we included only measures of psychopathy designed to assess either primary or secondary psychopathy. For each factor, we included four indices of primary and secondary psychopathy, embodied in four self-report scales of psychopathy: the Levenson Self-Report Psychopathy scales (LSRP; Levenson et al., 1995), the Antisocial Process Screening Device (APSD; Frick & Hare, 2001); Hare’s Self-Report Psychopathy Scale – 3rd Edition (SRP-III; Paulhus, Hemphill, & Hare, in press), and the Psychopathic Personality Inventory (PPI-R; Lilienfeld & Widows, 2005). As predicted by Fowles, 1980 and Lykken, 1995, and Newman et al., we hypothesized that primary psychopathy would be associated with a weak BIS. Like Lykken and Newman et al., we also believed that a strong BAS would be associated with secondary psychopathy.
