Victims

4 downloads 0 Views 399KB Size Report
Stephen F. Austin State University. Daniel F. ... been small (Babcock, Green, & Robie, 2004; Feder & Wilson, 2005), leading some to sug- gest the need to ...
With the Compliments of Springer Publishing Company, LLC

Violence and Victims

www.springerpub.com/vv

Violence and Victims, Volume 28, Number 3, 2013

Interpersonal Problems and Personality Features as Mediators Between Attachment and Intimate Partner Violence David M. Lawson, PhD Stephen F. Austin State University

Daniel F. Brossart, PhD Texas A&M University We examined whether hostile dominant interpersonal problems (HDIP), antisocial features, and borderline features mediated the relationship between attachment (anxiety or avoidance) and intimate partner violence (IPV) with a sample of 132 male partner abusers. We conducted two path analyses with avoidant attachment as the predictor in one model and anxious attachment as the predictor in a second model. In both models, HDIP, antisocial features, and borderline features were the mediators with IPV as the criterion. For both models, the attachment variable had statistically significant path values to the mediating variables. However, neither antisocial nor borderline features had statistically significant path values from the mediating variable to the criterion variable (IPV). Only HDIP had a statistically significant path value from the mediating variable to the criterion variable in both models. However, only the avoidant model produced a statistically significant specific indirect effect indicating that HDIP clearly mediated the relationship between attachment and IPV. Results suggest that partner abusive men with predominantly avoidant and, to a lesser degree, anxious attachment may be at increased risk for addressing conflicts in a coercive, controlling, and vengeful manner that is manifested in physical aggression toward a partner. Further, interpersonal constructs may be better measures of psychopathology and provide more relevant clinical targets than personality constructs with male partner abusers.

Keywords: intimate partner violence; attachment; interpersonal problems; personality; mediation

A

pproximately 1 in 4 (24.3%) women have experienced severe physical violence by an intimate male partner (Black et al., 2011). Furthermore, compared to intimate partner violence (IPV) toward men, women experience more severe injuries and longer lasting symptoms such as posttraumatic stress disorder, depression, and anxiety (Black et al., 2011). However, after 30 years of examining characteristics of male partner abusers, research remains equivocal about the most salient factors and their interrelationships that contribute to and explain IPV (Hamberger, 2008; Saunders, 2004). 414

© 2013 Springer Publishing Company http://dx.doi.org/10.1891/0886-6708.VV-D-12-00031

Copyright © Springer Publishing Company, LLC

Interpersonal Problems and Personality Features

415

This situation creates a particular challenge for the field of IPV because these factors (e.g., ­psychopathology) often guide treatment. However, treatment effects for IPV have been small (Babcock, Green, & Robie, 2004; Feder & Wilson, 2005), leading some to suggest the need to examine relationship variables in addition to established intrapersonal variables associated with IPV (Saunders, 2004; Taft, Murphy, Musser, & Remington, 2004). A few researchers have examined relationship-based variables such as attachment and interpersonal problems. These variables have shown initial promise for increasing our understanding of how relationship factors contribute to IPV (Dutton, Bodnarchuk, Kropp, Hart, & Ogloff, 1997; Murphy & Blumenthal, 2000). They are especially relevant with the increase in treatment models that incorporate elements of relationship-based therapies with cognitive behavioral therapy (CBT; Lawson, 2010; Sonkin & Dutton, 2003). Relationship-based elements focus on current maladaptive interpersonal patterns that are associated with conflict and IPV. These patterns are based on the quality of early interactions with caregivers and become preset, enduring expectations about self and others that shape interactions with others. Personality features and interpersonal problems have several important links. Both represent general, deep-seated tendencies in one’s outlook on self and others with their origins rooted in early experiences. Further, some researchers view interpersonal relatedness as a component of psychopathology (Horowitz, 2004; Kiesler, 1986). This connection is supported empirically by Soldz, Budman, Demby, and Merry (1993) and Kiesler (1986), who determined that antisocial and borderline personality disorders fall within the hostile dominant regions of the circumplex model of interpersonal problems. However, some suggest that personality as an intrapersonal construct has limited clinical value because it lacks sufficient specificity as a target for treatment (Murphy, Taft, & Eckhardt, 2007). If HDIP are considered an important dimension of psychopathology but with more relevant clinical targets (e.g., “I want to get revenge against other people too much”; Horowitz, 2004), they may be a useful measure of psychopathology for clinicians in addition to or in place of an intrapersonal personality measure (e.g., Millon Clinical Multiaxial Inventory [MCMI] or Minnesota Multiphasic Personality Inventory [MMPI]).

INTERPERSONAL PROBLEMS AND INTIMATE PARTNER VIOLENCE Interpersonal problems have received little attention in IPV research and yet have the potential for increasing our understanding of factors that contribute to IPV. Interpersonal problems often emerge during childhood and by adolescence, they become established, patterns of interpersonal difficulties (Murphy & Blumenthal, 2000). They are one of the most frequently reported problems in psychotherapy and if persistent, they are viewed as a major dimension of personality disturbance (Monsen, Hagtvet, Havik, & Eilertsen, 2006). Research indicates that interpersonal problems are related to IPV (Lawson & Brossart, 2009; Murphy & Blumenthal, 2000), while attachment is related to interpersonal problems (Lawson, 2008). Interpersonal theory purports two primary postulates. First, interpersonal behaviors are organized along two dimensions: affiliation (ranging from hostile to friendly) and power (ranging from submission to dominance). These dimensions produce several polar opposite interpersonal problems. Horowitz and Vitkus (1986) proposed a circumplex model comprising eight types of interpersonal problems that correspond to eight octants: intrusive, domineering, vindictive, cold, socially avoidant, nonassertive, exploitable, and overly nurturant.

Copyright © Springer Publishing Company, LLC

416

Lawson and Brossart

Second, two people mutually influence one another in interactions. Research supports these two postulates (Horowitz & Vitkus, 1986; Wiggins, 1982). This study focuses on three interpersonal problems located in adjacent octants of the circumplex at the dominance end of the dominance-submissive axis: vindictive, domineering, and intrusive. These scales are referred to as hostile dominant interpersonal problems (HDIP; Horowitz, Rosenberg, & Bartholomew, 1993). Clients with HDIP are more difficult to treat in psychotherapy than clients with friendly submissive problems (e.g., nonassertive) and tend to have greater difficulty understanding people (Horowitz et al., 1993). Also, HDIP are linked to interparental violence in the family of origin (Blumenthal, Neemann, & Murphy, 1998). Furthermore, increasing HDIP are related to increasing psychological aggression in relationships and severe IPV (Lawson, 2008), especially vindictive interpersonal problems (Lawson & Brossart, 2009). Finally, HDIP are associated with anger and IPV (Murphy et al., 2007).

PERSONALITY FEATURES AND INTIMATE PARTNER VIOLENCE Between 50% and 90% of court-ordered partner abusers have distinct personality disorder traits (Dutton & Starzomski, 1994; Hamberger, Lohr, Bonge, & Tolin, 1996). Furthermore, personality factors and psychopathology are among the strongest and most consistent predictors of IPV (Ehrensaft, Cohen, & Johnson, 2006), with the rates of personality disorders being up to six times higher for male partner abusers than men in the general population (Dutton, 2006). Partner abusive men have been conceptualized along two dimensions of psychopathology, antisocial, and borderline features (e.g., Ross & Babcock, 2009). Furthermore, they are more likely to evidence borderline and antisocial personality features than nonviolent men (Dutton, 2006). Men with antisocial features are characterized by instrumental attempts to control and intimidate partners as well as exhibit significant deficits in empathy and remorse and disregard for the rights of others (Dutton, 2007). Men with borderline features are characterized by an unstable self-concept and identity, impulsiveness, and anxiety-based rage related to rejection or abandonment (Dutton, 2007). In a study of partner violent men, Ross and Babcock (2009) found that men with antisocial personality disorder tended to use proactive violence (i.e., unprovoked, goal-directed, low arousal), whereas borderline personality disorder men tended to use reactive violence (i.e., unplanned, impulsive, high arousal, hostile). Behavior-based assessment models approximate Ross and Babcock’s findings (Stroops, Bennett, & Vincent, 2010).

ATTACHMENT AND INTIMATE PARTNER VIOLENCE Attachment theory offers a unique perspective in studying IPV with its emphasis on regulating closeness and distance in relationships (Fraley & Davis, 1997). According to theory, people possess an inherent need to develop emotional bonds with predictable attachment figures that provide security, protection, and intimacy. Over time, children develop strategies in response to separation and reunion with attachment figures that develop into predictable attachment patterns (Cassidy, 2008). These patterns continue into adolescent and adult relationships (Grossmann, Grossmann, Winter, & Zimmermann, 2002), predicting similar attachment patterns under stable life conditions (Waters, Merrick, Treboux, Crowell, & Albersheim, 2000).

Copyright © Springer Publishing Company, LLC

Interpersonal Problems and Personality Features

417

The continuous attachment dimensions of avoidance and anxiety provide the best description of attachment (Brennan, Clark, & Shaver, 1998) and reflect distinctly different underlying working models. Individuals with predominate anxious attachment often seek extreme intimacy, care, and attention by the attachment figure but with an underlying fear of being rejected or abandoned by them (Collins, Guichard, Ford, & Feeney, 2004). Individuals with predominate avoidance attachment tend to avoid intimacy demanding relationships with attachment figures based on a history of inconsistent or nonavailability of an attachment figure. They seek distance from others as a means to maintain autonomy and limit intimacy. Several studies support the link between attachment and IPV. Magdol, Moffitt, Caspi, and Silva (1998) conducted a longitudinal study investigating developmental antecedents of IPV and determined that an insecure attachment to parents (e.g., little warmth, trust, or positive communication) predicted men’s later IPV. Furthermore, partner abusive men report more insecure attachment styles (e.g., preoccupied or anxious) with their partners than nonviolent men (Holtzworth-Munroe, Meehan, Herron, Rehman, & Stuart, 2003; Waltz, Babcock, Jacobson, & Gottman, 2000). Finally, in a clinical sample of partner abusive men, pretreatment anxious attachment predicted mild physical abuse and psychological abuse, whereas pretreatment avoidant attachment predicted posttreatment total violence (mild and severe; Lawson & Brossart, 2009). In addition, several studies found that certain combinations of male and female attachment dimensions predict IPV. Bond and Bond (2004) found that female anxious attachment and male dismissing (avoidance) attachment predicted IPV. Furthermore, Doumas, Pearson, Elgin, and McKinley (2008) found that pairing an anxious female partner with an avoidant male partner was associated with both male and female perpetrated IPV. Together, these two studies suggest unique attachment dimensions for different genders. Most notably, avoidance attachment may be particularly relevant to men’s perpetration of IPV.

ATTACHMENT AND PERSONALITY FEATURES Attachment has become more prevalent in personality research related to IPV. In a series of studies, Dutton (2007) found that partner violent men reported childhood exposure to parental violence, rejection by fathers, and insecure attachment to mothers, which were associated with fearful (anxious) attachment and borderline personality organization. Consistent with borderline personality characteristics, these men tended to experience a fear of abandonment and rejection in close relationships, which results in a need to control the relationship as a means to reduce anxiety. The IPV was one means to reduce anxiety and the immediate threat of partner abandonment. Such men report more insecure attachment styles with partners than nonviolent men (Babcock, Jacobson, Gottman, & Yerington, 2000; Holtzworth-Munroe et al., 2000). Babcock et al. (2000) found that partner abusive males with dismissing/avoidant attachment had the greatest antisocial features compared to partner abusive males with preoccupied/anxious attachment. The dismissing/avoidant men became violent primarily when their wives became defensive. The preoccupied/anxious men were similar to a borderline personality profile and tended to use IPV when their wives attempted to disengage from them. These results are similar to the research of Dutton (2007). Typology researchers have found similar results, pairing avoidance attachment with antisocial personality in one subtype and anxiety attachment with borderline personality in another subtype (HoltzworthMonroe et al., 2000; Mauricio, Tein, & Lopez, 2007).

Copyright © Springer Publishing Company, LLC

418

Lawson and Brossart

ATTACHMENT AND INTERPERSONAL PROBLEMS Horowitz et al. (1993) holds that interpersonal problems develop in part from one’s attachment history. This linkage is reflected in research with various participants. Horowitz (2004) noted that adults who consistently had negative relationships with caregivers during childhood often distrust other people, and therefore avoided intimate relationships. These individuals strongly resisted ceding control to others and often reported interpersonal problems related to hostile dominance (Horowitz et al., 1993). Furthermore, HDIP were associated with avoidant attachment. Similarly, Chen and Mallinckrodt (2002) reported that avoidant attachment was associated with HDIP with a group of graduate students. Finally, Lawson (2008) found significant relationships between HDIP and anxiety and dependent attachment with partner violent men. This research suggests a significant connection between HDIP and attachment. To our knowledge, only two studies have addressed one or more but not all of the variables proposed in this study. Mauricio et al. (2007) examined if borderline or antisocial personality disorders mediated the relationship between anxiety and avoidance attachment and physical and psychological aggression. Results indicated that both personality disorders fully mediated the relationship between avoidant attachment and physical and psychological aggression but only partially mediated the relationship between anxiety attachment and psychological aggression. Interpersonal problems were not included in their study. Furthermore, IPV data were collected only from the perpetrator, thus ­increasing the chance of a biased report. In the second study, Lawson and Malnar (2011) examined whether HDIP mediated the relationship between attachment (anxious and avoidant) and IPV (violence severity and psychological aggression). Results determined that HDIP mediated the relationship between avoidant attachment and both violence severity and psychological aggression. The HDIP did not mediate the relationship between anxious attachment and the IPV variables. However, personality features and HDIP have never been compared as mediators in the same analysis.

PURPOSE OF STUDY We examined whether antisocial and borderline personality features and HDIP mediate the relationship between avoidant and anxiety attachment and IPV. Hypotheses were (a) with avoidant attachment as the independent variable and violence severity as the dependent variable, HDIP will be a stronger mediator than antisocial and borderline personality ­features, and (b) with anxiety attachment as the independent variable and severity violence as the dependent variable, HDIP will be a stronger mediator than antisocial and borderline personality features. This study extends the IPV literature by comparing the viability of two sets of related but different constructs of psychopathology, personality features, and interpersonal problems as mediators between attachment and violence severity. Although personality measures have been used extensively in the study of IPV (e.g., MCMI), they have been criticized for having limited clinical value because they lack adequate specificity as treatment targets (Murphy et al., 2007). If interpersonal problems and personality features are equally viable measures of psychopathology, but interpersonal problems are more clinically relevant, then interpersonal problems may be the more viable psychopathology measure for IPV research and clinical work.

Copyright © Springer Publishing Company, LLC

Interpersonal Problems and Personality Features

419

METHOD Participants Participants were drawn from a pool of 155 men on probation for IPV from the southwest. The study was approved by the university Institutional Research Board. Men were informed about the study and asked to sign a consent form. Fourteen men were dropped from analysis when they rescinded their consent. Twelve cases were omitted from analysis because of excessive item omission and/or invalid MCMI-III results. The final sample consisted of 132 men with documented histories of IPV indicated by an arrest report, spouse/ partner report, or self-report. The IPV offences ranged from mild (e.g., pushing) to severe (e.g., beating up). The men had the following characteristics: ages ranged from 18 to 67 years (M 5 30.47; SD 5 10.4); 41 were African American, 42 were Hispanic, 20 were Asian, and 29 were White; and education ranged from 7 to 17 years (M 5 11.5; SD 5 2.44). All men participated in IPV over the past year ranging from 1 to 75 violent acts (Median 5 4, M 5 7.1, SD 5 12). Fifty-nine men (42.4%) had engaged in severe violence (e.g., chocked, kicked, forced partner to have sex).

MEASURES Attachment was assessed with the 18-item Adult Attachment Scale (AAS; Collins & Read, 1990), which comprises three scales: (a) close assesses comfort with closeness and intimacy; (b) depend assesses comfort depending on others and belief that people can be relied on when needed; and (c) anxiety assesses worry about being rejected and abandoned. Higher scores (from 1 not at all characteristic of me to 5 very characteristic of me) indicate increasing comfort with closeness, comfort depending on others, and increasing anxiety about being rejected or abandoned. Men responded to the items based on their perspective of romantic relationships. Test–retest reliability over a 2-month interval with a sample of college students was .68 for close, .71 for depend, and .52 for anxiety (Collins & Read, 1990). Sperling, Foelsch, and Grace (1996) found convergent validity of the AAS with several attachment ­measures (e.g., Attachment Style Measure; Hazan-Shaver Attachment Self-Report). Collins and Read (1990) found the following average alpha coefficients based on three separate samples: close, .80; depend, .79; and anxiety, .84. Alpha coefficients for this study were close, .76; depend, .77; and anxiety, .80. Collins (1996) holds that the close and depend scales can be combined to reflect an avoidant scale. The close and depend scales are correlated (r 5 .53 and r 5 .57 in this study), whereas both scales were negatively correlated with anxiety (close: r 5 2.34 and r 5 2.46 in this study; and depend: r 5 2.25 and r 5 2.33 in this study). Avoidance reflects comfort or discomfort in approaching or avoiding in relationships. Lower scores indicate increasing avoidance. Higher scores indicate increasing approaching. Combining the close and depend scales to form the avoidance scale along with the anxiety scale has been employed in attachment-related studies (e.g., Collins, 1996; Collins & Feeney, 2000; Lawson, 2010). Shaver and Fraley (2004) hold that Collin’s anxiety and avoidance scales are viable measures of these two attachment dimensions. Following Collins, we converted the AAS scales to two dimensions: anxiety and avoidance. Alpha coefficients for this study were anxiety, .80 and avoidance, .74.

Copyright © Springer Publishing Company, LLC

420

Lawson and Brossart

The frequency and severity of partner abuse was assessed with the 23-item Modified Conflict Tactics Scale (MCTS; Pan, Neidig, & O’Leary, 1994). The MCTS comprises three scales: psychological aggression (e.g., “sulked and/or refused to talk,” “insulted, yelled, or swore”), mild physical aggression (e.g., “controlled spouse physically,” “pushed, grabbed, and shoved”), and severe physical aggression (e.g., “choked or strangled,” “beat up”). We used only the 12 physical aggression items to assess for IPV. A violence seve­ rity weighted score was determined that combined mild and severe physical aggression items. The violence severity scale was computed by taking into account the severity and frequency of the violence (Stets & Straus, 1990). The weighted scale considers the injuryproducing potential of each violent act by applying a larger weighted multiplier with increasingly more severe types of violence multiplied times the frequency. For example, mild physical violence (e.g., threw something) was unweighted. More severe violence such as “kicked, bit” 3 2, “beat up” 3 5, and so on up to “used knife or gun” 3 8. In this study, the coefficient alpha was .86 for violence severity. Men in treatment as well as their partners completed the MCTS. However, the analysis included only the partner’s MCTS scores to minimize underreporting bias by the men. Interpersonal functioning was assessed with the 32-item short form of Inventory of Interpersonal Problems-Short Circumplex (IIP-SC; Soldz et al., 1995). The IIP-SC includes eight subscales: domineering, vindictive, overly cold, socially avoidance, nonassertive, exploitable, overly nurturant, and intrusive. The IIP-SC closely corresponds to the circumplex model of interpersonal functioning. In this study, coefficient alphas for the subscales ranged from .68 to .84. Similar to Taft et al. (2004), we used a 12-item composite score to measure interpersonal problems related to hostile dominance: vindictive (“I want to get revenge against people too much”), domineering (“I am too aggressive toward other people”), and intrusive (“It is hard for me to keep things private from other people”). The 175-item Millon Clinical Multiaxial Inventory III (MCMI-III; Millon, 1994) assessed for antisocial and borderline personality features. The antisocial scale assesses the degree of disregard for the values/standards of society and for the rights of people. The borderline scale assesses for patterns of unstable interpersonal relationships, emotional dysregulation, and fear of abandonment. Coefficient alphas in this study were .75 and .77, respectively.

PROCEDURES The men were administered the AAS, MCTS, IIP-SC, and the MCMI-III in groups of 15–30. They completed demographic information and signed consent forms, including consent for their partners to complete the MCTS. Participants were interviewed individually to determine the nature of the offence, criminal history, family history, and a psychosocial history. Partners were contacted by phone and responded to the MCTS based on their participating partner’s behavior.

RESULTS The correlations among the variables are in Table 1. All intercorrelations were significant at p , .01 or p , .001 except the relationship between anxiety and violence severity (p 5 .11). All significant intercorrelations, whether positive or negative, were associated with increasing levels of antisocial and borderline features, avoidance (negative correlations indicate more avoidance) and anxiety, HDIP, and IPV.

Copyright © Springer Publishing Company, LLC

Interpersonal Problems and Personality Features

421

TABLE 1.  Means, Standard Deviations, and Correlations Among the Variables (n 5 132) Variable

1

2

3

4

5

1. Antisocial

__

2. Borderline

.70***

__

3. Anxiety attachment

.25**

.31***

4. Avoidance attachment

2.32*** 2.41*** 2.33*** __

5. Violence severity

.25**

.33***

.11

.30**

6. Hostile dominant   interpersonal problems

.41***

.51***

.31***

2.47*** .32***

Means (SD) 52.2 (23.8) 53.6 (26.6)

__

2.8 (0.63) 3.1 (0.60) __

31 (51.6) 2.9 (1.4)

**p , .01. ***p , .001.

Mediating Effect of Personality Features and Hostile Dominant Interpersonal Problems Several path models were examined to test for mediation (Byrne, 2006), with antisocial, borderline, and HDIP variables as mediators; violence severity as the dependent variable; and avoidant or anxious attachment style as the independent variable. The strategy consisted of testing a fully saturated model (a partially mediated model) against a fully mediated model for each attachment variable. Figure 1 is an example of the fully saturated model tested. The fully mediated model is similar except that there is no path from the attachment style variable to violence severity. Although the current sample tends to be on the small side (n 5 132), it should be sufficient for the path models that follow, in that there were at least 10 subjects per parameter that required statistical estimates (Jackson, 2003).

Antisocial

E2

Borderline

Violence Severity

E3 Anxiety Attachment Hostile Dominant Interpersonal Problems

E10 E12

Figure 1.  Fully saturated model with anxious attachment style.

Copyright © Springer Publishing Company, LLC

422

Lawson and Brossart

Antisocial

E1

 .30

*

.07

*

Borderline

40

.

.10

Violence Severity

E2 Avoidance Attachment

.21*

.4

6*

Hostile Dominant Interpersonal Problems

E4 E3

Figure 2.  Fully mediated model of avoidant attachment style with standardized solution.

In the model with avoidant attachment as the predictor variable, the fully saturated model includes a path from the avoidant attachment to violence severity. This model is saturated, x2 (0) 5 0. The fully mediated model, which does not contain a path from the attachment variable to the violence variable, resulted in x2 (1) 5 1.75, p 5 .19. Model fit could be considered moderately good given that CFI 5 .995, standardized root mean square residual (SRMR) 5 .02, and root mean square error of approximation (RMSEA) 5 .078 (90% CI 5 0–.265). Because the x2 was not statistically significant, the complete mediation model could not be rejected in favor of the partially mediated model (includes both direct and indirect effects; see Figure 1; MacKinnon, 2008). Figure 2 contains the completely mediated model for the avoidant model with standardized values. Statistically significant (p , .05) values are indicated by an asterisk. For the model with anxious attachment as the predictor, the results were similar in that the fully mediated model could not be rejected, x2 (1) 5 .88, p 5 .35. Fit was very good with CFI 5 1.00, SRMR 5 .02, and RMSEA , .01 (90% CI 5 0–.231). Figure 3 shows the complete mediation model with standardized values.

Antisocial

E1 .07

.25

*

Borderline

.10

*

.33

Violence Severity

E2 Anxiety Attachment

.21*

.33

*

Hostile Dominant Interpersonal Problems

E4 E3

Figure 3.  Fully mediated model of anxious attachment style with standardized solution.

Copyright © Springer Publishing Company, LLC

Interpersonal Problems and Personality Features

423

The total standardized indirect effects for the avoidant attachment model was 2.16, p 5 .02. The specific indirect effect of the HDIP variable was 2.09, p 5 .04 (95% CI 5 2.21–.03). Confidence intervals were based on the bias-corrected bootstrap method recommended by Preacher and Hayes (2008). This specific indirect effect represents the ability of the HDIP variable to mediate the effect of avoidant attachment on violence severity while controlling for all other mediators. The other specific indirect effects for the two personality variables were not statistically significant, antisocial was 2.02, p 5 .61, and borderline was 2.04, p 5 .51. The total standardized indirect effects for the anxious model was .12, p 5 .02. For the anxious attachment model, the specific indirect effect of the HDIP variable was 2.07, p 5 .055, (95% CI 5 2.02–.16). In reference to Figure 3, one sees that although the individual paths leading into and out of the HDIP variable were statistically different from zero, the ability of the HDIP variable to mediate the effect of anxious attachment on violence severity while controlling for the other mediating variables was not quite large enough to reach statistical significance. The other specific indirect effects for the two personality variables were not statistically significant, antisocial was .02, p 5 .62 and borderline was .03, p 5 .53.

DISCUSSION This study examined whether men’s HDIP and personality features mediated the relationship between anxious and avoidance attachment and IPV. Both hypotheses were supported with the avoidant model providing stronger evidence than the anxious model. Both models provided a good fit to the data; however, the avoidant model provided a statistically significant specific indirect effect that increases confidence that the HDIP variable was the stronger mediator in the avoidant model. Thus, avoidant attachment, in particular, clearly predicts violence severity through its relationship with HDIP. Results suggest that partner abusive men with predominantly avoidant and, to a lesser degree, anxious attachment may be at increased risk for addressing conflicts in a coercive, controlling, and vengeful manner that is manifested in physical aggression toward a partner. Thus, HDIP may be a more efficient and readily applicable psychopathology construct for clinicians. These themes can be identified and targeted in treatment (see Limitations and Application section). Results suggest that HDIP warrant consideration when mediation models are used to examine predictors of IPV. In addition, given that the avoidant attachment model performed better overall than the anxious attachment model, the results support earlier research suggesting that avoidant attachment is particularly relevant for partner violent men (Doumas et al., 2008). The failure of antisocial and borderline personality features to be significant mediators in either path model contradicts previous research suggesting a clear meditation function of these features between attachment dimensions and IPV (Mauricio et al., 2007). The most apparent explanation for these results is that HDIP was a stronger mediator than antisocial or borderline features when all three mediators were included the path analyses. Research provides a possible explanation for the relationship between attachment, HDIP, and IPV. Avoidant attachment is characterized by distress in intimacy-demanding relationships, an orientation toward self-reliance and withdrawal, and deactivation of attachment, presuming others inaccessible and as a result, a hostile orientation to others (Feeney, 2008). This is especially the case with partners who they perceive as demanding. Conversely, partners with anxiety attachment tend toward engagement and hyperactivation of the attachment system to avoid abandonment (Collins et al., 2004). During conflict, men with avoidant

Copyright © Springer Publishing Company, LLC

424

Lawson and Brossart

attachment may experience demand from their partners to remain engaged in response to their attempts to disengage (Lawson & Malnar, 2011), often referred to as a pattern of demand-withdrawal (Christensen & Heavey, 1990). Pressure from a demanding partner who attempts to engage a withdrawing, avoidant partner may create setting conditions for IPV by the avoidant partner. Research has identified this pattern with some partner violent men (Allison, Bartholomew, Mayseless, & Dutton, 2008). In contrast, anxious attachment leads many partner abusive men to experience fear of abandonment resulting in an exaggerated need to control their partner to prevent abandonment and, therefore, reduce the anxiety (Allison et al., 2008). The IPV is one means to achieve this end. From an attachment perspective, there may be two different dynamics leading to IPV for partner violent men, although our results suggest ultimately that it is a hostile dominant interpersonal style characterized by control, vengeance, and aggression that is the means through which an avoidant or anxious attached male partner expresses violence toward a female partner.

LIMITATIONS AND APPLICATION This study had several limitations. The data were cross sectional rather than longitudinal, thus limiting conclusions. Future research may include meditational analyses across treatment to better delineate causal links among variables. The modest sample size prevented the inclusion of latent variables such as the individual scales comprising HDIP and different forms of IPV. Furthermore, this study examined only self-report interpersonal problems as mediators. Future studies could include partner data that represents actual couple interaction focusing on demand/withdrawal interaction patterns from observational data. Also, partners could provide responses on the IIP-SC to examine the complementary elements of interpersonal theory. This would make it possible to examine more fully interpersonal theory with couples. Notwithstanding these limitations, this study included several improvements over previous research. First, although Mauricio et al. (2007) conducted the first study examining personality factors as mediators between attachment and IPV, their measure of IPV was obtained from perpetrators rather than victims. Victims, as opposed to perpetrator reports, typically are considered a more reliable measure of IPV. This study employed partner’s report of IPV rather than perpetrators report, thus enhancing measurement validity. Second interpersonal problems previously had not been included with personality measures in meditational models with IPV. Thus, it provides another measure of psychopathology beyond the intrapersonal level. Finally, HDIP provide clinicians with a tangible expression of interpersonal aspects of psychopathology that can be readily targeted in IPV treatment along with attachment themes (e.g., abandonment). Based on the current results, clinicians can be aware that even though men with anxious or avoidant attachment may have different motivations to engage in IPV, both dimensions may be related to a general interpersonal orientation toward control, intrusion, and revenge that may lead to violence toward their partner. The three variables comprising the HDIP composite scale provide clear clinical phenomena: vindictive (e.g., “I want to get revenge against people too much”), domineering (e.g., “I am too aggressive toward other people”), and intrusive (e.g., “It is hard for me to keep things private from other people”). Self-statements related to these themes can be targeted for exploration and cognitive restructuring. Furthermore, IPV treatments are beginning to expand beyond CBT and feminist treatment models to include relationship-based therapies

Copyright © Springer Publishing Company, LLC

Interpersonal Problems and Personality Features

425

(e.g., brief psychodynamic therapy; Lawson, 2010; Sonkin & Dutton, 2003) that borrow heavily from attachment and brief psychodynamic theories. Relationship-based interventions target current maladaptive interpersonal patterns linked to conflict and IPV. This study provides a basis for conceptualizing the interplay between attachment and interpersonal problems because they are related to IPV.

REFERENCES Allison, C. J., Bartholomew, K., Mayseless, O., & Dutton, D. G. (2008). Love as a battlefield attachment and relationship dynamics in couples identified for male partner violence. Journal of Family Issues, 29, 125–150. Babcock, J. C., Green, C. E., & Robie, C. (2004). Does batterers’ treatment work? A meta-analytic review of domestic violence treatment. Clinical Psychology Review, 23, 1023–1053. Babcock, J. C., Jacobson, N. S., Gottman, J. M., & Yerington, T. P. (2000). Attachment, emotional regulation, and the function of marital violence: Differences between secure, preoccupied, and dismissing violent and nonviolent husbands. Journal of Family Violence, 15, 391–409. Black, M. C., Basile, K. C., Breiding, M. J., Smith, S. G., Walters, M. L., Merrick, M. T., . . . Stevens, M.R. (2011). The National Intimate Partner and Sexual Violence Survey (NISVS): 2010 Summary Report. Atlanta, GA: National Center for Injury Prevention and Control, Centers for Disease Control and Prevention. Blumenthal, D. R., Neemann, J., & Murphy, C. M. (1998). Lifetime exposure to interparental physical and verbal aggression and symptom expression in college students. Violence and Victims, 13, 175–196. Bond, S. B., & Bond, M. (2004). Attachment styles and violence within couples. Journal of Nervous & Mental Disease, 192, 857–863. Brennan, K. A., Clark, C. L., & Shaver, P. R. (1998). Self-report measurement of adult attachment: An integrative overview. In J. A. Simpson & W. S. Rholes (Eds.), Attachment theory and close relationships. (pp. 46–76). New York, NY: Guilford Press. Byrne, B. (2006). Structural equation modeling with EQS. Mahwah, NJ: Lawrence Erlbaum Associates. Cassidy, J. (2008). The nature of the child’s ties. In J. Cassidy & P. R. Shaver (Eds.), Handbook of attachment: Theory, research, and clinical applications (pp. 3–22). New York, NY: Guilford Press. Chen, E. C., & Mallinckrodt, B. (2002). Attachment, group attraction, and self-other agreement in interpersonal circumplex problems and perceptions of group members. Group Dynamics, Theory, Research and Practice, 6, 311–324. Christensen, A., & Heavey, C. L. (1990). Gender and social structure in the demand/withdrawal pattern of marital conflict. Journal of Personality and Social Psychology, 59, 73–82. Collins, N. L. (1996). Working models of attachment: Implications for explanation, emotion, and behavior. Journal of Personality and Social Psychology, 71, 810–832. Collins, N. L., & Feeney, B. C. (2000). A safe haven: An attachment theory perspective on support seeking and caregiving in intimate relationships. Journal of Personality and Social Psychology, 78, 1053–1073. Collins, N. L., Guichard, A. C., Ford, M. B., & Feeney, B. C. (2004). Working models of attachment: New developments and emerging themes. In J. A. Rholes & J. A. Simpson (Eds.), Adult attachment: Theory, research, and clinical implications (pp. 196–239). New York, NY: Guilford Press. Collins, N. L., & Read, S. J. (1990). Adult attachment, working models and relationship quality in dating couples. Journal of Personality and Social Psychology, 58, 644–663. Doumas, D. M., Pearson, C. L., Elgin, J. E., & McKinley, L. L. (2008). Adult attachment as a risk factor for intimate partner violence. Journal of Interpersonal Violence, 23, 616–634.

Copyright © Springer Publishing Company, LLC

426

Lawson and Brossart

Dutton, D. G. (2006). Rethinking domestic violence. Vancouver, Canada: UBC Press. Dutton, D. G. (2007). The abusive personality: Violence and control in intimate relationships (2nd ed.). New York, NY: Guilford Press. Dutton, D. G., Bodnarchuk, M., Kropp, R., Hart, S. K., & Ogloff, J. R. P. (1997). Wife assault treatment and criminal recidivism: An 11-year follow-up. International Journal of Offender Therapy and Comparative Criminology, 41, 9–23. Dutton, D. G., & Starzomski, A. J. (1994). Psychological differences between court-ordered and self-referred wife assaulters. Criminal Justice and Behavior: An Intentional Journal, 21, 203–222. Ehrensaft, M. K., Cohen, P., & Johnson, J. G. (2006). Development of personality disorder symptoms and the risk for partner violence. Journal of Abnormal Psychology, 115, 474–483. Feder, L., & Wilson, D. B. (2005). A meta-analytic review of court-mandated batterer intervention programs: Can courts affect abusers’ behavior? Journal of Experimental Criminology, 1, 239–262. Feeney, J. A. (2008). Adult romantic attachment. In J. Cassidy & P. R. Shaver (Eds.), Handbook of attachment: Theory, research, and clinical applications (pp. 456–481). New York, NY: Guilford Press. Fraley, R. C., & Davis, K. (1997). Attachment formation and transfer in young adults’ close friendships and romantic relationships. Personal Relationships, 4, 131–144. Grossmann, K. E., Grossmann, K., Winter, M., & Zimmermann, P. (2002). Attachment relationships and appraisal of partnership: From early experience of sensitive support to later relationship representation. In L. Pulkkinen & A. Caspi (Eds.), Paths to successful development (pp. 73–105). Cambridge, UK: Cambridge University Press. Hamberger, L. K. (2008). Twenty-five years of change in working with partner abusers—Part II: Observations from the trenches about changes in understanding of abusers and abuser treatment. Journal of Aggression, Maltreatment & Trauma, 17, 1–22. Hamberger, L. K., Lohr, J. M., Bonge, D., & Tolin, D. F. (1996). A large sample empirical typology of male spouse abusers and its relationship to dimensions of abuse. Violence and Victims, 11, 277–292. Holtzworth-Munroe, A., Meehan, J. C., Herron, K., Rehman, U., & Stuart, G. L. (2000). Testing the Holtzworth-Munroe and Stuart (1994) batterer typology. Journal of Consulting and Clinical Psychology, 68, 1000–1019. Holtzworth-Munroe, A., Meehan, J. C., Herron, K., Rehman, U., & Stuart, G. L. (2003). Do subtypes of martially violent men continue to differ over time? Journal of Consulting and Clinical Psychology, 71, 728–740. Horowitz, L. M. (2004). Interpersonal foundations of psychopathology. Washington, DC: American Psychological Association. Horowitz, L. M., Rosenberg, S. E., & Bartholomew, K. (1993). Interpersonal problems, attachment styles, and outcome in brief dynamic psychotherapy. Journal of Consulting and Clinical Psychology, 61, 549–560. Horowitz, L. M., & Vitkus, J. (1986). The interpersonal basis of psychiatric symptoms. Clinical Psychology Review, 6, 443–469. Jackson, D. L. (2003). Revisiting sample size and number of parameter estimates: Some support for the N:q hypothesis. Structural Equation Modeling, 10, 128–141. Kiesler, D. J. (1986). The 1982 interpersonal circle: An analysis of DSM-111 personality disorders. In T. Millon & G. Klerman (Eds.), Contemporary directions in psychopathology toward DSM-IV (pp. 571–597). New York, NY: Guilford Press. Lawson, D. M. (2008). Attachment, interpersonal problems, and family functioning: Differences between intimate violent versus non-intimate violent men. Psychology of Men and Masculinity, 9, 90–105. Lawson, D. M. (2010). Comparing cognitive behavioral therapy and integrated therapy in group treatment for partner violent men. Psychotherapy: Theory, Research, Training, Practice, 47, 122–133.

Copyright © Springer Publishing Company, LLC

Interpersonal Problems and Personality Features

427

Lawson, D. M., & Brossart, D. (2009). Attachment, interpersonal problems, and treatment outcomes in group therapy for partner violent men. Psychology of Men and Masculinity, 10(4), 288–301. Lawson, D. M., & Malnar, S. G. (2011). Interpersonal problems as a mediator between attachment and intimacy partner violence. Journal of Family Violence, 26, 421–430. MacKinnon, D. P. (2008). An introduction to statistical mediation analysis. New York, NY: Routledge. Magdol, L., Moffitt, T. E., Caspi, A., & Silva, P. A. (1998). Developmental antecedents of partner abuse: A prospective-longitudinal study. Journal of Abnormal Psychology, 107, 375–389. Mauricio, A. M., Tein, J. Y., & Lopez, F. G. (2007). Borderline and antisocial personality scores as mediators between attachment and intimate partner violence. Violence and Victims, 22, 139–157. Millon, T. (1994). Millon Clinical Multiaxial Inventory-III manual. Minneapolis, MN: National Computer System. Monsen, J. T., Hagtvet, K. A., Havik, O. E., & Eilertsen, D. E. (2006).Circumplex structure and personality disorder correlates of the Interpersonal Problems Model (IIP-C): Construct validity and clinical implications. Psychological Assessment, 18, 165–173. Murphy, C. M., & Blumenthal, D. R. (2000). The mediating influence of interpersonal problems on the intergenerational transmission of relationship aggression. Personal Relationships, 7, 203–218. Murphy, C. M., Taft, C. T., & Eckhardt, C. I. (2007). Anger problem profiles among partner violent men: Differences in clinical presentation and treatment outcome. Journal of Counseling Psychology, 54, 189–200. Pan, H. S., Neidig, P. H., & O’Leary, K. D. (1994). Male-female and aggressor-victim differences in the factor structure of the Modified Conflict Tactics Scale. Journal of Interpersonal Violence, 9, 366–382. Preacher, K. J., & Hayes, A. F. (2008). Asymptotic and resampling strategies for assessing and comparing indirect effects in multiple mediator models. Behavior Research Methods, 40, 879–891. Ross, J. M., & Babcock, J. C. (2009). Proactive and reactive violence among intimate partner violent men diagnosed with antisocial and borderline personality disorder. Journal of Family Violence, 24, 607–617. Saunders, D. G. (2004). The place of typology of men who are “martially” violent within a nested ecological model: A response to Holtworth-Munroe and Meehan. Journal of Interpersonal Violence, 19, 1390–1395. Shaver, P. R., & Fraley, R. C. (2004). Self report measures of Adult Attachment. Adult Attachment Lab. Retrieved from http://psychology.ucdavis.edu/labs/Shaver/measures.htm Soldz, S., Budman, S., Demby, A., & Merry, J. (1993). Representation of personality disorders in circumplex and five-factor space: Explorations with a clinical sample. Psychological Assessment, 5, 41–52. Soldz, S., Budman, S., Demby, A., & Merry, J. (1995). A short form of the Inventory of Interpersonal Problems Circumplex Scales. Assessment, 2, 53–63. Sonkin, D. J., & Dutton, D. G. (2003). Treating assaultive men from an attachment perspective. In D. G. Dutton & D. J. Sonkin (Eds.), Intimate violence: Contemporary treatment approaches (pp. 105–134). New York, NY: Haworth Press. Sperling, M. B., Foelsch, P., & Grace, C. (1996). Measuring adult attachment: Are self-report instruments congruent? Journal of Personality Assessment, 67, 37–51. Stets, J. E., & Straus, M. A. (1990). Gender differences in reporting marital violence and its medical and psychological consequences. In M. A. Straus & R. J. Gelles (Eds.), Physical violence in American families: Risk factors and adaptations to violence in 8,145 families (pp. 151–166). New Brunswick, NJ: Transaction. Stroops, C., Bennett, L., & Vincent, N. (2010). Development and predictive ability of a behaviorbased typology of men who batter. Journal of Family Violence, 25, 325–335.

Copyright © Springer Publishing Company, LLC

428

Lawson and Brossart

Taft, C. T., Murphy, C. M., Musser, P. H., & Remington, N. A. (2004). Personality, interpersonal, and motivational predictors of the working alliance in group cognitive-behavior therapy for partner violent men. Journal of Consulting and Clinical Psychology, 72, 349–354. Waltz, J., Babcock, J. C., Jacobson, N. S., & Gottman, J. M. (2000). Testing a typology of batterers. Journal of Consulting and Clinical Psychology, 68, 658–669 Waters, E., Merrick, S., Treboux, D., Crowell, J., & Albersheim, L. (2000). Attachment security in infancy and early adulthood: A twenty-year longitudinal study. Child Development, 71, 684–689. Wiggins, J. S. (1982). Circumplex models of interpersonal behavior in clinical psychology. In P. C. Kendall & J. N. Butcher (Eds.), Handbook of research methods in clinical psychology (pp. 183–221). New York, NY: Wiley. Correspondence regarding this article should be directed to David M. Lawson, PhD, Department of Human Services, Stephen F. Austin State University, Nacogdoches, TX 75962-3019. E-mail: [email protected]

Copyright © Springer Publishing Company, LLC