Cumulative Childhood Maltreatment and Subsequent Psychological Violence in Intimate Relationships: The Role of Emotion Dysregulation

Cumulative Childhood Maltreatment

Past research on the effects of childhood maltreatment has frequently failed to consider the effects of cumulative types of abuse or neglect on psychological IPV, even though studies increasingly recognize that individuals exposed to one category of maltreatment are likely to also have experienced other types of victimization (Elliott, Alexander, Pierce, Aspelmeier, & Richmond, 2009). For instance, individuals who have sustained childhood sexual abuse are likely to have also experienced physical or psychological abuse at the same period (Arata et al., 2005; Elliott et al., 2009; Higgins & McCabe, 2001). Thus, it would appear that children who have experienced interpersonal trauma are at an increased risk of additional victimization (Finkelhor, Ormrod, & Turner, 2007; Hodges et al., 2013). Studies have referred to this experience as cumulative childhood maltreatment, which is operationalized as the total number of different types of interpersonal victimization experienced by a given individual (Briere, Hodges, & Godbout, 2010; Cloitre et al., 2009).

In the interpersonal maltreatment literature, results are generally consistent in finding that experiencing multiple types of maltreatment may be particularly detrimental since it is related to greater psychosocial maladjustment than experiencing a single type of abuse (Briere et al., 2010; Hodges et al., 2013). Indeed, according to studies on cumulative maltreatment exposure (e.g., Briere et al., 2010; Cloitre et al., 2009), the number of different types of interpersonal trauma experienced, over and above the frequency or severity of each specific type of trauma, would act as a robust predictor of subsequent negative psychological and relational outcomes.

Cumulative Childhood Maltreatment and IPV

To our knowledge, only three studies have examined the impact of cumulative childhood maltreatment on IPV and these studies only examined its impact on IPV perpetration (Berthelot et al., 2014; Green et al., 2005; Maguire et al., 2015). However, considering that childhood maltreatment has consistently been associated with interpersonal revictimization (Lilly et al., 2014), attention must also be paid to its impact on IPV victimization.

In addition, most studies that have examined the effects of childhood maltreatment on IPV have investigated the specific impact of different types of abuse and neglect or have depicted the presence or absence of childhood maltreatment as a dichotomous variable rather than examining the correlates of cumulated types of child maltreatment (Edwards, Dixon, Gidycz, & Desai, 2014; Elmquist et al., 2016; Messing, La Flair, Cavanaugh, Kanga, & Campbell, 2012; Widom, Czaja, & Dutton, 2014). Yet, according to Elmquist et al. (2016), the combination of different types of childhood maltreatment could be related to more frequent IPV in adulthood. As such, the examination of the impact of cumulative childhood maltreatment on later psychological IPV offers an innovative approach to the study of the intergenerational transmission of violence.

Furthermore, while cumulative child maltreatment has somewhat consistently been associated with later IPV, the mechanisms underlying this relationship remain unclear. By reflecting the complex repercussions of cumulative maltreatment and by acting as a known risk factor for both IPV perpetration and victimization, disturbances in childhood maltreatment survivors’ affective and interpersonal self-regulatory capacities might help explain the link between childhood maltreatment and IPV.

Emotion Dysregulation

Emotion regulation refers to an individual’s capacity to control and tolerate strong and negative affect, without resorting to avoidance strategies that distract, soothe, or draw attention away from emotional distress (Briere, 2002). Specifically, emotion regulation can be conceptualized as a multidimensional construct (Gratz & Roemer, 2004), composed of a cognitive and a behavioral dimensions (Berzenski & Yates, 2010). The cognitive dimension of emotion regulation is based on the regulation of affects and refers to the inhibition of mood swings or of the expression of anger. The behavioral dimension of emotion regulation is reflected by the ability to refrain from externalizing negative emotions through avoidance or dysfunctional behaviors (Briere & Runtz, 2002). These behavioral strategies manifest as tension reduction behaviors, the most common of which are self-destructive and risky sexual behaviors, binge/purge eating, impulsivity, and aggressive behaviors (Briere & Runtz, 2002). Since emotion regulation is a multidimensional construct (Gratz & Roemer, 2004), researchers suggest that the cognitive capacities to accurately identify and express emotions must be investigated as potentially distinct from the behavioral capacities to control one’s response to emotions (Berzenski & Yates, 2010). Yet, very few studies have included both the cognitive and behavioral dimensions of affect dysregulation in the same integrative model (Berthelot et al., 2014; Gratz, Paulson, Jakupcak, & Tull, 2009), a shortcoming that is addressed in the current study.

Emotion Dysregulation and Cumulative Childhood Maltreatment

Researchers have suggested that the experience of maltreatment during childhood may interfere with the development of adaptive emotion regulation by exposing children to overwhelming or extreme emotional demands while simultaneously failing to teach them how to regulate emotional arousal, to control their behaviors in the context of emotional arousal, or to tolerate emotional distress (Gratz et al., 2009). Others have also suggested that the experience of violence, hostility, or indifference in childhood might foster affective predispositions to aggression by hindering the capacity to get rid of negative feelings with adequate coping mechanisms (Allen, 2011; Berthelot et al., 2014). Essentially, when faced with insufficient affect regulation skills or overwhelming negative emotions, adult survivors of childhood maltreatment are often unable to cope and tend to resort to avoidance strategies to reduce the impact and duration of their negative experiences (Briere & Lanktree, 2008). Empirical data have supported those theoretical propositions by showing that adult survivors of child maltreatment tend to exhibit emotion dysregulation as reflected by emotional instability, problems in inhibiting the expression of strong affects, and the use of tension-reduction behaviors (i.e., self-injury, binge–purge eating, impulsivity, excessive risk-taking, and aggression; Briere, 2002; Briere & Rickards, 2007). Exposure to cumulative childhood maltreatment has been specifically proposed to result in an intricate pattern of affective and interpersonal disturbances that might reflect self-regulatory disturbances (Briere & Scott, 2015; Cloitre et al., 2009), which was confirmed in several empirical studies (Briere et al., 2010, 2008; Dugal, Godbout, Bigras, & Bélanger, 2015; Elliott et al., 2009).

Emotion Dysregulation and IPV

Emotion regulation is central to adaptive coping in relational settings, as the capacity to deal with emotionally straining situations is critical for successful intimate relationships (Berzenski & Yates, 2010). Thus, emotion dysregulation could be at the root of negative relational interactions and lead to psychological IPV as a way to cope when faced with negative or intense emotions (Allen, 2011; Bliton et al., 2015). For instance, women in treatment for IPV often cite nonregulated negative emotions, arising during couple conflicts, as precipitants to their use of violence toward their partner (Stuart et al., 2006). Yet, studies exploring these issues rather emphasize the importance of emotion dysregulation as a predictor for IPV in men, given that IPV perpetrated by women is more likely used in a context of self-defense (Hamberger, 2005). Still, empirical data have confirmed that both male and female survivors of childhood maltreatment are more vulnerable to resort to aggression when relational problems challenge their ability to regulate negative emotions (Berthelot et al., 2014).

Although several studies have concluded to the existence of a relationship between emotion dysregulation and IPV, very few have examined how emotion dysregulation may be related to psychological IPV victimization. According to those studies, difficulties in emotion regulation could contribute to interpersonal victimization through defective risk detection abilities (i.e., inability to identify and respond effectively to dangerous situations or people; Iverson, McLaughlin, Adair, & Monson, 2014) and deficits in effective coping strategies (i.e., the emotional control necessary to resolve conflicts in an appropriate manner; Lilly et al., 2014). Difficulties in emotion regulation could thus hamper one’s ability to deal with domestic disputes, a known risk factor for conflicts between partners turning violent (Johnson, 2006), thereby increasing the possibility of being victimized by a romantic partner. Indeed, a growing body of research has ascertained that bidirectional violence, in which partners are both perpetrators and victims of IPV, is a prevalent type of IPV in the general population (Archer, 2000; Straus, 2011). Consequently, researchers suggest that both IPV perpetration and victimization should be included in models examining IPV (Bélanger, Mathieu, Dugal, & Courchesne, 2015).

Cumulative Childhood Maltreatment, Emotion Dysregulation, and Psychological IPV

To our knowledge, only a few studies have examined the relationship between childhood maltreatment, emotion dysregulation, and IPV. A study conducted by Berthelot et al. (2014) reported that, in a clinical population of men and women consulting for sexual or relational problems, the experience of childhood maltreatment increased the risk of IPV perpetration through elevated Post-traumatic stress disorder (PTSD) symptoms and anger personality traits. These authors, along with others such as Iverson et al. (2014), suggested that a lower threshold for anger combined with PTSD-related fear and hypervigilance could underlie emotion regulation difficulties. Hence, emotion dysregulation might be at the heart of partner violence since it increases the risk of experiencing IPV perpetration and victimization. However, this hypothesis needs further empirical validation.

In a sample of undergraduate students, Gratz et al. (2009) found evidence for the mediational role of emotion dysregulation in the relationship between childhood maltreatment and physical IPV perpetration as committed by men toward women. However, this mediational model did not apply to women from their sample. In a sample of undergraduate women, Berzenski and Yates (2010) found that emotion dysregulation acted as a partial mediator of the relationship between child maltreatment and physical, psychological, and sexual IPV victimization and perpetration. Finally, in a sample of women survivors of childhood maltreatment and IPV, Lilly et al. (2014) reported similar results, showing that disturbances in emotion regulation resulting from childhood maltreatment increased the risk of physical, psychological, and sexual IPV victimization. However, previous studies primarily relied on nonclinical samples of students (Berzenski & Yates, 2010; Gratz et al., 2009) did not examine both IPV perpetration and victimization in the same model (Berthelot et al., 2014; Gratz et al., 2009; Lilly et al., 2014), did not differentiate cognitive from behavioral indicators of emotion dysregulation (Gratz et al., 2009), did not specifically examine psychological IPV (Berzenski & Yates, 2010), and did not examine those variables through a multivariate integrative model (Gratz et al., 2009).

Participants

A total of 162 adults (65 men, 97 women) consulting for sexual or relational problems participated in the study. Their mean age was 38.2 years (SD = 12.5, range = 17–70). Participants were full-time workers (64.2%), students (12.3%), unemployed (7.5%), stayed at home (6.6%), or retired (8.5%). They were either married (21.6%, n = 35), in a common-law partnership or cohabiting (42.6%, n = 69), or in a relationship with a regular partner (22.2%, n = 36). Some participants were in a relationship during the past year but were now single/with occasional partner(s) (3.1%, n = 5) or single/not committed to a relationship (10.5%, n = 17). The majority of participants reported being heterosexual (87.6%), whereas 5.0% reported being homosexual and 6.2% bisexual. As for education, 17.8% of participants held a high school diploma, 39.5% attained a college or professional studies degree, 25.3% completed undergraduate studies, and 15.4% completed graduate studies. A total of 40.6% reported an annual income of CAD $19,999 or less, 28.8% reported an income between CAD $20,000 and CAD $39,999, and 30.7% reported an income of CAD $40,000 or more, which indicate a high proportion of low-income participants in this sample. With regards to the experience of trauma before and after the age of 18 years, such as death of a family member or natural disaster, it was reported by 40.1% (before the age of 18 years) and 43.2% (after the age of 18 years) of participants. The current sample presents characteristics (i.e., low sexual desire in women and erectile dysfunction in men as main reasons for consultation, high level of psychological distress in both men and women) that are consistent with other studies that have included participants consulting for sexual or relational problems (Godbout, Lafrenaye-Dugas, Hébert, & Goulet, 2015).

Childhood Maltreatment

A self-report questionnaire assessing eight types of childhood maltreatment (i.e., sexual, physical and psychological abuse, physical and psychological neglect, witnessing physical and psychological violence, and bullying) was administered to participants. For childhood sexual abuse, participants indicated, on a yes/no scale, if they experienced any unwanted sexual contact (e.g., touching and penetration) or any sexual contact with a person 5 years older or in a position of authority before the age of 18 years (2 items, r = .69). For the other types of maltreatment, participants were asked to report how many times they experienced each event in a typical year (before the age of 18 years), on a 7-point Likert scale ranging from 0 (never) to 7 (almost every day). Physical and psychological abuse were assessed using indicators derived from the Early Trauma Inventory-Self-Report (ETISR-SF; Bremner, Bolus, & Mayer, 2007). Physical abuse referred to having been slapped, burned, punched or kicked, hit, or shoved (5 items, Cronbach’s α = .72). Psychological abuse referred to having been put down, degraded, ridiculed, insulted, made to feel like you did not count, or told you were no good by a caregiver (3 items Cronbach’s α = .89). Neglect was assessed using items derived from the Comprehensive Child Maltreatment Scale for Adults (Higgins & McCabe, 2001). Psychological neglect referred to having been ignored or not felt loved nor understood by a caregiver (3 items Cronbach’s α = .84). Physical neglect referred to not having received regular meals, baths, clean clothes, needed medical attention, or being confined in a room alone for extended periods of time. Witnessing interparental violence was assessed by 2 items derived from the Childhood Maltreatment Questionnaire (Godbout, Lussier, & Sabourin, 2006; Godbout et al., 2009): (a) witnessing psychological violence (“My parents insulted each other or shouted at each other”) and (b) witnessing physical violence (“My parents shoved, hit or fought each other with hands, feet or objects”). One item, based on the recommendations of the Centers for Disease Control and Prevention (Gladden, Vivolo-Kantor, Hamburger, & Lumpkin, 2014), assessed bullying by asking participants the typical annual frequency at which they experienced intimidation or harassment by peers. For the purpose of this study, each scale was dichotomously coded (0 = absence, 1 = presence) and the results were added to obtain a continuous score of cumulative maltreatment, ranging from 0 to 8, indicating the number of different types of child maltreatment experienced.

Emotion Dysregulation

Affect regulation and tension-reduction activities were assessed using the Inventory of Altered Self-Capacities (IASC; Briere, 2000). The affect regulation scale included 9 items and evaluated affect instability and affect skills deficits by examining problems in affect regulation and control expressed by mood swings, problems in inhibiting the expression of anger, and inability to regulate dysphoric states without externalization. The tension-reduction activities scale included 9 items and referred to the tendency to react to painful internal states with potentially dysfunctional externalizing or avoidant behaviors that distract, soothe, or reduce internal distress (e.g., self-mutilation and sexual acting-out). Participants indicated the frequency with which they experienced each of the items in the last 6 months on a 5-point Likert scale ranging from 1 (never) to 5 (very often). Total scores of each scale ranged from 9 to 45, with higher scores reflecting more important emotion regulation difficulties. To determine whether participants were above or below the clinical cut-off, scores were transformed into T-scores (Briere, 2000). The psychometric qualities of the original standardized and validated measure (Briere & Runtz, 2002) were satisfactory. In the current sample, Cronbach’s α was .88 for the affect dysregulation scale and .62 for the tension-reduction activities scale.

Intimate Partner Violence

A self-report questionnaire was administered to assess inflicted and sustained psychological IPV. Three questions assessing psychological IPV were derived from the French adaptation (Hébert & Parent, 2000) of the Revised Conflict Tactics Scale (CTS-2; Straus et al., 1996). Participants were asked whether they experienced insulting, yelling, threatening, or calling names with their partner. An additional question, specifically evaluating control of the partner, was derived from an instrument assessing violence against women in intimate relationships (Lavoie & Vézina, 2001) and was incorporated as part of psychological IPV. The severity of violence was also examined as descriptive statistics using the subscales of the psychological intimate partner violence scale of the CTS-2: one of the items referred to “minor” IPV, whereas three items indicated “severe” IPV. For each item, participants were asked to report the frequency, during the last year, at which they perpetrated and sustained these psychologically violent behaviors on a 6-point Likert scale ranging from 0 (this never happened) to 6 (more than 20 times during the past 12 months). For the purpose of this study, scores were averaged into continuous scores, ranging from 0 to 6 and indicating the annual frequency of psychological IPV perpetration (Cronbach’s α = .80) and victimization (Cronbach’s α = .78), as per the authors’ recommendations (Straus et al., 1996). Although the severity of psychologically violent acts was examined by this measure, it was not accounted for in the final model since we were interested in examining the chronicity or frequency of exposure to psychological violence in cumulative maltreatment survivors’ romantic relationships, as cumulative maltreatment has been shown to lead to more frequent IPV in adulthood (Elmquist et al., 2016).

Analytic Strategy

Correlations and multiple regressions were first conducted to assess the associations among the study variables. Then, path analysis was used to test the hypothesized model that cumulative childhood maltreatment increases the risk of presenting affect dysregulation and tension-reduction activities, which in turn lead to the experience of psychological IPV. Path analyses were conducted using Mplus, version 7 (Muthén & Muthén, 1998-2012) and the Maximum Likelihood Estimation with Robust Standard Errors (MLR), which is robust to non-normality and protects against heteroscedasticity (White, 1980). To examine the mediational role of emotion dysregulation indicators, we computed direct effects (i.e., path coefficient from child maltreatment to perpetrated IPV and from child maltreatment to sustained IPV) and indirect effects (i.e., the product of the path coefficients from childhood maltreatment to affect dysregulation and from affect dysregulation to perpetrated or sustained IPV). To examine the magnitude and significance of the indirect effects, 95% bootstrap confidence intervals were computed (Mackinnon & Fairchild, 2009). This bias-corrected method is based on the distribution of the product of coefficients and generates confidence limits for the true value of coefficients for indirect effects. When zero is not in the confidence interval, the indirect effect is considered significant. To examine the gender moderation hypothesis, a multiple-group, gender-invariance analysis was conducted. Overall model fit was tested by considering together the comparative fit index (CFI; Bentler, 1990), the root mean square error of approximation (RMSEA; Steiger, 1990), the standardized root mean square residual (SRMR), the chi-square statistic, and the ratio of chi-square to degrees of freedom (X ²/df). A combination of a nonstatistically significant chi-square value, a CFI value of .90 or higher, RMSEA and SRMR values below .06, and a ratio of chi-square to df less than 3 are thought to represent a good fit (Hu & Bentler, 1999; Kline, 2011; Ullman, 2001).

Objective 1: Descriptive Statistics and Bivariate Correlations

Objective 2: Integrative Model of the Role of Affect Dysregulation and Tension-Reduction Activities in the Relationship Between Cumulative Maltreatment and Psychological IPV Perpetration and Victimization

The proposed model adequately fitted the data, CFI = .99, RMSEA = .08, CI [.00, .20], χ ²[2] =4.57, p = .10 , ratio χ ²/df = 2.29, and SRMR = .03. However, in line with the principle of parsimony, a nonsignificant association between affect dysregulation and psychological IPV victimization (β = .13, ns) was removed. The adjusted model provided slightly improved fit to the data: CFI = .99, RMSEA = .07, CI [.00, .16], χ ²[3] =5.64, p = .13, ratio χ ²/df = 1.88, and SRMR = .03. As presented in Figure 1, this model indicated that cumulative childhood maltreatment was positively associated with both the affect dysregulation and the use tension-reduction activities, which in turn increased the risk of experiencing psychological IPV. More specifically, the use of tension-reduction activities in trauma survivors increased the risk of experiencing both psychological IPV perpetration and victimization, whereas affect dysregulation only increased the risk of perpetrating psychological IPV. The covariance between the residual terms of affect dysregulation and tension-reduction activities was .61 (p < .001). The covariance between the residual terms of perpetrated and sustained psychological IPV was .74 (p < .001). To confirm the generalizability of the mediational model across age, especially since the range of participants’ ages was wide (17–70), an additional analysis was conducted to control for age of participants. Results yielded a well-fitting model, and adding age to the mediational model did not change the significance and strength of the associations between variables.

Figure 1.

Path analysis of the role of affect dysregulation and tension -reduction activities in the relationship between cumulative childhood maltreatment and psychological intimate partner violence (IPV) perpetration and victimization.

*p < .05. **p < .01. ***p < .001.

Results indicated that the indirect effect of cumulative childhood maltreatment through affect dysregulation was significant for perpetrated psychological IPV (β= .04, 95% CI .00 to .04), which suggests a mediational effect between these variables. The indirect effects of cumulative childhood maltreatment via the use of tension-reduction activities was also significant for both perpetrated psychological IPV (β = .08, 95% CI .01, .08) and sustained psychological IPV (β = .11, 95% CI .04, .17). Results suggested mediational effects of tension-reduction activities in the relationship between cumulative childhood maltreatment and perpetrated psychological IPV and in the relationship between cumulative childhood maltreatment and sustained psychological IPV. Overall, the final model explained 12.0% of the variance for psychological IPV perpetration and 8.3% for psychological IPV victimization. Standardized coefficients are presented in Figure 1.

Objective 3: Gender Moderation of the Integrative Model