واکنش پذیری عاطفی و خود تنظیمی در رابطه با اختلالات شخصیت

Abstract Associations of both reactive and regulative temperamental features with personality disorders (PDs) are investigated in a sample of 162 normal controls and 89 psychiatric inpatients. Reactive and regulative temperamental features were assessed by means of the BIS/BAS Scales and the Attentional Control Scale. Dimensional PD scores were obtained by using the ADP-IV. All PDs were characterized by low levels of effortful control, cluster C PDs by high BIS and cluster B PDs by high BAS. For several PDs, BIS and effortful control interacted: BIS was only related to severe PD pathology if effortful control was low. Clinical implications of these findings are discussed.

Introduction Several authors have highlighted associations with temperament as promising avenues for understanding psychopathology (Muris and Ollendick, 2005 and Nigg, 2006). Until recently, most research attention has been devoted to reactive aspects of temperament (i.e. affective-motivational reactivity). Several studies have shown that extreme (diminished as well as elevated) levels of temperamental reactivity are associated with psychopathology: high levels of behavioural inhibition system (BIS) reactivity with anxiety, depression, alcoholism and eating disorders; low levels of BIS with psychopathy; high levels of behavioural activation system (BAS) reactivity with substance abuse, manic episodes, conduct problems; low levels of BAS with depression (Bijttebier, Beck, Claes, & Vandereycken, 2009). The contribution of temperament to vulnerability for psychopathology should, however, not be viewed as merely guided by emotional reactivity. Current theories of vulnerability for psychopathology also emphasize the influence of effortful processes that enable persons to modulate their emotional reactions (Nigg, 2006). As such, the risk associated with temperamental reactivity can be decreased, making effortful control a protective factor. Given that individual differences in reactive and regulative temperament are supposed to underlie enduring personality dimensions, reactive and regulative temperament can be expected to have relevance for DSM-IV Axis I clinical disorders, and certainly for DSM-IV Axis II personality disorders (PDs). Nonetheless, in contrast to the relevance of both aspects of temperament in recent theories of vulnerability to pathology, only few studies explored their combined influence to PD symptoms, which is the aim of the present study. In Gray’s (1987) Reinforcement Sensitivity Theory (RST), reactivity to immediate incentive contexts is conceptualized in terms of motivation systems of avoidance (BIS) and approach (BAS). The BIS is sensitive to stimuli that signal conditioned aversive events (punishment), non-reward, and novelty. It inhibits behaviour that may lead to negative or painful outcomes (Fowles, 1980). In terms of individual differences in personality, higher BIS sensitivity is reflected in higher proneness to anxiety (Carver & White, 1994) and is related to the personality trait of Neuroticism (Nigg, 2006). The BAS (Fowles, 1980) is sensitive to signals of unconditioned reward, non-punishment and escape from punishment. In terms of individual differences in personality, elevated BAS sensitivity is reflected in elevated proneness to engage in goal-directed efforts and to experience positive feelings (Carver & White, 1994) and is related to the personality trait of Extraversion (Nigg, 2006). Over the years, RST developed to include a third major system: the fight–flight system (FFS: Gray, 1987). Whereas the BIS responds to conditioned aversive stimuli, the FFS responds to unconditioned aversive stimuli. In 2000, Gray and McNaughton presented a major revision of RST (Corr, 2008). The BAS is now assumed to be responsive to (un)conditioned positively valenced stimuli. The Fight/Flight/Freeze System (FFFS) adopts the punishment system role that was originally ascribed to the BIS. Finally, the BIS is now believed to be responsible for the resolution of goal conflict in general (reactive control), i.e. to regulate situations in which both the BAS and the FFFS have been activated (Bijttebier et al., 2009). Only a handful of studies have investigated patterns of reactivity associated with Axis II PDs. High levels of BIS have been related to all cluster C PDs (Caseras et al., 2001, Farmer and Nelson-Gray, 1995, Fullana et al., 2004 and Pastor et al., 2007), cluster A paranoid and schizotypal PDs (Gilbert et al., 2005 and Pastor et al., 2007) as well as cluster B borderline PD; whereas low levels of BIS have been related to cluster B antisocial and narcissistic PDs (Fowles, 1980 and Pastor et al., 2007). High levels of BAS have been related to all cluster B PDs (Carver and White, 1994, Farmer and Nelson-Gray, 1995, Harmon-Jones, 2003, Pastor et al., 2007 and Quay, 1993) as well as the cluster A paranoid and schizotypal PDs (Pastor et al., 2007), which reflects – according to Fowles (1992) the tendency toward positive schizophrenia symptoms in these PDs. Finally, low levels of BAS have been described in cluster A schizoid and cluster C-avoidant PDs, which are both characterized by low levels of extraversion (Pastor et al., 2007). Current theories of psychopathology (Nigg, 2006) not only emphasize the role of temperamental reactivity, but also the influence of effortful processes that enable persons to regulate their emotional reactivity and as such decrease the risks associated with reactivity (Bijttebier et al., 2009). The notion of effortful control (EC; Rothbart, 1989) includes both behavioural forms of self-control as well as attentional processes (e.g., the ability to voluntarily focus or shift attention) and is related to the personality trait Conscientiousness (Nigg, 2006). As far as we know, few studies have focussed on the association between (lack of) EC and PDs. Hoermann, Clarkin, Hull, and Levy (2005) compared three subgroups of borderline patients with different levels of EC with respect to symptoms, interpersonal relations and personality organization. Subgroup 1 (high EC), exhibited the fewest problems in symptoms, interpersonal functioning and personality organization, whereas subgroup 3 (low EC) was characterized by the most problems in these areas. Subgroup 2, high in some aspects of EC and low in others, was situated between groups 1 and 3. Both reactive and effortful dimensions of temperament have been related separately to PDs. However, integration of both dimensions in one study is needed to investigate their joint influence on PDs and test current theorizing about temperament and psychopathology. The work of Depue and Lenzenweger (2005) offers a clear theoretical framework to understand the influence of both reactive and regulative features on personality disturbance. The reactive dimensions (BIS and BAS) of temperament provide the qualitative emotional content of contemporaneous behaviour, and the regulative dimension (EC) of temperament modulates the probability of elicitation of all the reactive systems. For example, the antisocial PD is characterized by a high ratio of BAS/BIS reactivity and a low level of EC. Our aim is to investigate the joint – interactive or additive – influence of reactive and effortful dimensions of temperament on PDs. In the first case (interactive), EC acts as a moderator on the association between temperamental reactivity and PDs, so they may not be studied separately. In the second case (additive), reactivity and EC each play a unique role and have additive effects on personality psychopathology. Based on the literature, we expect that high levels of BIS are related to cluster A, B (e.g., borderline, histrionic) and C PDs; and low levels of BIS to cluster B antisocial and narcissistic PDs. High levels of BAS are considered to be related to all cluster B PDs. Fewer hypotheses can be formulated concerning the joined (additive or interactive) influence of both reactive and regulative features on PDs. Based on the literature on anxiety problems (Muris & Ollendick, 2005), we may expect that cluster C PDs are not only determined by the main effects of BIS and EC, but also by their interaction (BIS × EC), meaning that high BIS only determines problematic cluster C PDs if EC is low. In the same line of reasoning and referring to research on conduct disorders (Uzieblo, Verschuere, & Crombez, 2007), the cluster B antisocial PD is not only determined by the main effects of BAS and EC, but also by their interaction (BAS × EC), meaning that high BAS only determines the antisocial PD if EC is low. These studies, suggest in both cases an interactive effect instead of solely an additive effect of BIS/BAS and EC on PDs.

Results 3.1. Gender and group differences on reactive/regulative temperamental aspects and dimensional PD scores To determine gender (male/female) and group (control/patients) differences on temperamental and PD scores, we performed MANOVA’s with gender and group as independent variables and reactive/regulative temperament and PD scores as dependent variables (Table 1). Table 1. Means and standard deviations of male and female controls and patients on the measures of (BIS/BAS), self-regulation (ACS) and personality disorders (ADP-IV). Controls Patients Total group Controls vs. Patients Males vs. Females Males Females Total Males Females Total Males Females M (SD) M (SD) M (SD) M (SD) M (SD) M (SD) M (SD) M (SD) F(1, 247) F(1, 247) BIS 18.7 (3.3) 21.5 (3.6) 20.3 (3.7) 21.3 (4.2) 23.8 (3.6) 22.6 (4.1) 19.7 (3.9) 22.3 (3.8) 24.3⁎⁎⁎ 28.4⁎⁎⁎ BAS 35.8 (5.3) 35.0 (5.3) 35.4 (5.3) 35.7 (6.0) 35.4 (4.8) 35.5 (5.4) 35.8 (5.6) 35.1 (5.2) 0.0 0.6 Self-Regulation ACS 56.5 (13.7) 54.4 (12.8) 55.3 (13.2) 42.9 (13.2) 41.8 (13.7) 42.3 (13.4) 51.5 (15.0) 50.0 (14.4) 54.5⁎⁎⁎ 0.8 Personality Disorders PAR 13.9 (5.5) 12.9 (4.8) 13.3 (5.2) 20.7 (9.4) 20.3 (9.1) 20.5 (9.1) 16.4 (7.8) 15.4 (7.5) 60.4⁎⁎⁎ 0.5 SZ 15.5 (6.0) 12.1 (4.6) 13.6 (5.5) 20.5 (7.8) 21.0 (7.7) 20.7 (7.7) 17.3 (7.1) 15.1 (7.2) 69.6⁎⁎⁎ 3.1 ST 17.1 (8.2) 16.5 (6.5) 16.8 (7.3) 28.0 (10.4) 27.3 (10.1) 27.6 (10.2) 21.1 (10.4) 20.3 (9.4) 92.5⁎⁎⁎ 0.3 AS 12.2 (5.6) 10.3 (3.6) 11.1 (4.7) 16.8 (8.4) 15.0 (7.9) 15.9 (8.1) 13.9 (7.1) 12.0 (5.9) 33.0⁎⁎⁎ 5.0⁎ BDL 19.7 (8.0) 19.7 (7.4) 19.7 (7.6) 34.3 (13.1) 38.5 (13.7) 36.6 (13.5) 25.1 (12.3) 26.2 (13.4) 154.5⁎⁎⁎ 2.4 HYS 14.9 (5.3) 14.6 (5.6) 14.8 (5.5) 23.5 (7.7) 24.8 (8.8) 24.2 (8.3) 18.0 (7.5) 18.2 (8.4) 112.2⁎⁎⁎ 0.4 NARC 16.9 (6.4) 15.2 (5.5) 15.9 (6.0) 22.9 (8.2) 19.7 (6.1) 21.1 (7.3) 19.1 (7.7) 16.7 (6.1) 37.6⁎⁎⁎ 8.3⁎⁎ AV 14.8 (6.0) 14.6 (6.0) 14.7 (6.0) 23.5 (8.9) 27.2 (9.9) 25.5 (9.6) 18.0 (8.3) 18.9 (9.7) 117.3⁎⁎⁎ 3.0 DEP 15.2 (5.6) 15.7 (6.13) 15.5 (5.9) 24.1 (8.6) 28.2 (9.7) 26.3 (9.4) 18.5 (8.0) 20.0 (9.6) 122.9⁎⁎⁎ 5.5⁎ OC 20.5 (6.7) 19.3 (7.19) 19.8 (7.0) 28.7 (8.9) 29.7 (8.2) 29.2 (8.5) 23.5 (8.5) 22.9 (9.0) 85.0⁎⁎⁎ 0.0 PAR, paranoid PD; SZ, schizoid PD; ST, schizotypal PD; AS, antisocial PD; BDL, borderline PD; HIS, histrionic PD; NARC, narcissistic PD; AV, avoidance PD; DEP, dependent PD; OC, obsessive–compulsive PD; BIS, Behavioural Inhibition Scale; BAS, Behavioural Activation Scale; ACS, Attentional Control Scale. ⁎ p < 0.05. ⁎⁎ p < 0.01. ⁎⁎⁎ p < 0.001. Table options With respect to the reactive temperamental aspect (BIS/BAS), female participants scored significantly higher on the BIS scale than males; on the BAS scale, no significant gender differences emerged. With respect to the regulative temperamental aspect (ACS), no significant differences between male/female participants were found. In contrast, patients scored significantly lower than normal controls on the total ACS. On all PD scales, patients scored significantly higher than normal controls. Males scored significantly higher on the antisocial and narcissistic PD scales compared to females, and females scored significantly higher than males on the dependent PD scale. 3.2. Correlations between reactive/regulative temperamental aspects and dimensional PD scores controlled for gender The correlations between reactive and regulative temperamental aspects and PD scores controlled for gender are given in Table 2. Table 2. Correlations between reactivity and self-regulation and the 10 personality disorder scores controlled for gender. Cluster A PDs Cluster B PDs Cluster C PDs Paranoid Schizoid Schizotypal Antisocial Borderline Histrionic Narcissistic Avoidant Dependent Obsessive–compulsive BIS_Total 0.34⁎⁎⁎ 0.17⁎⁎ 0.20⁎⁎⁎ −0.05 0.37⁎⁎⁎ 0.28⁎⁎⁎ 0.18⁎⁎ 0.49⁎⁎⁎ 0.48⁎⁎⁎ 0.45⁎⁎⁎ BAS_Total 0.05 −0.15⁎⁎ 0.02 0.23⁎⁎⁎ 0.18⁎⁎ 0.25⁎⁎⁎ 0.22⁎⁎⁎ −0.05 0.11 0.06 Self-regulation ACS_Total −0.43⁎⁎⁎ −0.32⁎⁎⁎ −0.39⁎⁎⁎ −0.30⁎⁎⁎ −0.50⁎⁎⁎ −0.41⁎⁎⁎ −0.34⁎⁎⁎ −0.48⁎⁎⁎ −0.53⁎⁎⁎ −0.42⁎⁎⁎ BIS, Behavioural Inhibition Scale; BAS, Behavioural Activation Scale; ACS, Attentional Control Scale. ⁎⁎ p < 0.01. ⁎⁎⁎ p < 0.001. Table options 3.3. Prediction of dimensional PD scores based on gender, group, reactive and regulative temperamental aspects, and their interaction Finally, we tried to find out if reactive and regulative temperamental characteristics as well as their interactions were able to predict – additive or interactive – the dimensional PD scores. Significant interactions between reactive and regulative temperamental characteristics mean that their effects are partly interactive, and not only additive. The results of the hierarchical regression analyses with GENDER, GROUP, BIS, BAS, ACS and their interactions as predictor variables and the dimensional PD scores as dependent variables are displayed in Table 3. GENDER and GROUP were entered in the first step, BIS, BAS and ACS in the second step, and the interactions in the third step. The BIS × BAS interaction was not included because we were particularly interested in the interaction of both BIS/BAS and ACS. Table 3. Prediction of dimensional personality disorder scores based on gender, group, reactivity and self-regulation and their interaction. Cluster A PDs Cluster B PDs Cluster C PDs PAR SZ ST AS BDL HIS NARC AV DEP OC β β β Β β β β β β β R2 = 0.20 ⁎⁎⁎ R2 = 0.24 ⁎⁎⁎ R2 = 0.28 ⁎⁎⁎ R2 = 0.14 ⁎⁎⁎ R2 = 0.39 ⁎⁎⁎ R2 = 0.32 ⁎⁎⁎ R2 = 0.15 ⁎⁎⁎ R2 = 0.33 ⁎⁎⁎ R2 = 0.34 ⁎⁎⁎ R2 = 0.26 ⁎⁎⁎ GENDER 0.05 0.14⁎ 0.03 0.14⁎ −0.06 −0.02 0.16⁎⁎ −0.06 −0.09 0.02 GROUP 0.45⁎⁎⁎ 0.47⁎⁎⁎ 0.52⁎⁎⁎ 0.34⁎⁎⁎ 0.62⁎⁎⁎ 0.56⁎⁎⁎ 0.36⁎⁎⁎ 0.57⁎⁎⁎ 0.58⁎⁎⁎ 0.52⁎⁎⁎ R2 = 0.30 ⁎⁎⁎ a R2 = 0.29 ⁎⁎⁎ a R2 = 0.31 ⁎⁎⁎ a R2 = 0.27 ⁎⁎ a R2 = 0.50 ⁎⁎⁎ a R2 = 0.42 ⁎⁎⁎ a R2 = 0.24 ⁎⁎⁎ a R2 = 0.48 ⁎⁎⁎ a R2 = 0.52 ⁎⁎⁎ a R2 = 0.39 ⁎⁎⁎ a GENDER 0.12⁎ 0.16⁎⁎ 0.04 0.05 −0.01 −0.00 0.16⁎⁎ 0.06 0.01 0.13 GROUP 0.28⁎⁎⁎ 0.40⁎⁎⁎ 0.43⁎⁎⁎ 0.31⁎⁎⁎ 0.47⁎⁎⁎ 0.46⁎⁎⁎ 0.26⁎⁎⁎ 0.38⁎⁎⁎ 0.38⁎⁎⁎ 0.35⁎⁎⁎ BIS 0.19⁎⁎ 0.02 0.00 −0.27⁎⁎⁎ 0.14⁎⁎ 0.06 0.01 0.33⁎⁎⁎ 0.28⁎⁎⁎ 0.31⁎⁎⁎ BAS 0.03 −0.16⁎⁎ 0.01 0.25⁎⁎⁎ 0.16⁎⁎⁎ 0.24⁎⁎⁎ 0.21⁎⁎⁎ −0.09⁎ 0.07 0.03⁎ ACS −0.26⁎⁎⁎ −0.15⁎⁎ −0.21⁎⁎⁎ −0.25⁎⁎⁎ −0.26⁎⁎⁎ −0.19⁎⁎⁎ −0.21⁎⁎⁎ −0.21⁎⁎⁎ −0.28⁎⁎⁎ −0.17⁎⁎ R2 = 0.34 ⁎⁎⁎ a R2 = 0.29 ⁎⁎⁎ R2 = 0.32 ⁎⁎⁎ R2 = 0.30 ⁎⁎⁎ a R2 = 0.54 ⁎⁎⁎ a R2 = 0.44 ⁎⁎⁎ a R2 = 0.25 ⁎⁎⁎ R2 = 0.51 ⁎⁎⁎ a R2 = 0.55 ⁎⁎⁎ a R2 = 0.40 ⁎⁎⁎ GENDER 0.13⁎⁎ 0.16⁎⁎ 0.06 0.06 0.00 0.01 0.18⁎⁎ 0.08 0.02 0.14⁎⁎ GROUP 0.25⁎⁎⁎ 0.41⁎⁎⁎ 0.42⁎⁎⁎ 0.30⁎⁎⁎ 0.44⁎⁎⁎ 0.44⁎⁎⁎ 0.24⁎⁎⁎ 0.36⁎⁎⁎ 0.35⁎⁎⁎ 0.33⁎⁎⁎ BIS 0.21⁎⁎⁎ 0.01 0.01 −0.26⁎⁎⁎ 0.17⁎⁎⁎ 0.08 0.02 0.35⁎⁎⁎ 0.31⁎⁎⁎ 0.32⁎⁎⁎ BAS 0.02 −0.16⁎⁎ −0.00 0.23⁎⁎⁎ 0.16⁎⁎⁎ 0.22⁎⁎⁎ 0.21⁎⁎⁎ −0.10 0.07 0.03 ACS −0.26⁎⁎⁎ −0.15⁎⁎ −0.21⁎⁎⁎ −0.24⁎⁎⁎ −0.26⁎⁎⁎ −0.18⁎⁎⁎ −0.21⁎⁎⁎ −0.22⁎⁎⁎ −0.29⁎⁎⁎ −0.17⁎⁎ BIS × ACS −0.16⁎⁎ 0.03 −0.04 −0.02 −0.17⁎⁎⁎ −0.06 −0.08 −0.14⁎⁎ −0.16⁎⁎ −0.11⁎ BAS × ACS −0.04 0.05 −0.03 −0.12⁎ −0.05 −0.05 0.01 0.04 −0.03 0.03 BIS × BAS × ACS −0.09 0.05 −0.10 −0.11 −0.10 −0.15⁎ −0.07 −0.06 −0.08 −0.07 PAR, paranoid PD; SZ, schizoid PD; ST, schizotypal PD; AS, antisocial PD; BDL, borderline PD; HIS, histrionic PD; NARC, narcissistic PD; AV, avoidance PD; DEP, dependent PD; OC, obsessive–compulsive PD; BIS, Behavioural Inhibition Scale; BAS, Behavioural Activation Scale; ACS, Attentional Control Scale. a A significant increase in R2 compared to prior step. ⁎ p < 0.05. ⁎⁎ p < 0.01. ⁎⁎⁎ p < 0.001. Table options All cluster C PDs are associated with a high level of BIS, a low level of EC and their interaction. This interaction refers to the moderating function of EC on the expression of BIS. Patients with high levels of BIS and low levels of EC score higher on cluster C PDs than patients with high levels of BIS and high levels of EC (Fig. 1). The trend indicated in Fig. 1 is the same across all cluster C PDs. Mean dependent PD z-scores for participants with low/high levels of BIS×low/high ... Fig. 1. Mean dependent PD z-scores for participants with low/high levels of BIS × low/high levels of Attention Control. Figure options Within the cluster B PDs, all PDs are characterized by a high level of BAS and a low level of EC. The antisocial PD is further predicted by a low level of BIS, whereas the borderline PD is predicted by a high level of BIS. For the borderline PD, it also holds that a high level of BIS is modulated by the level of EC (BIS × ACS), meaning that the higher BIS is and the lower EC is, the higher the score on the borderline PD. For the antisocial PD, it holds that a high level of BAS is modulated by the level of EC (BAS × ACS), indicating that the higher BAS is and the lower EC is, the higher the score on the antisocial PD (Fig. 2). Mean antisocial PD z-scores for participants with low/high levels of ... Fig. 2. Mean antisocial PD z-scores for participants with low/high levels of BAS × low/high levels of Attention Control. Figure options All cluster A PDs are characterized by low levels of EC. The paranoid PD is predicted by high BIS, low EC, and their interaction. The schizoid PD is predicted by low BAS and low EC. “GENDER” (0 = female, 1 = male) has a positive influence on the paranoid, schizoid, narcissistic and obsessive–compulsive PDs, meaning that males score higher on these PDs than females. Also the variable “GROUP” (0 = normal controls; 1 = patients) shows a significantly positive influence on all PDs, meaning that patients score higher on the PDs scales than normal controls.