A Dyadic Analysis of Partner Maltreatment Using an Attachment-Based Diathesis–Stress Model: A Lab-Based Study

Objectives: Interpersonal and contextual factors are increasingly acknowledged to play an important role in understanding intimate partner maltreatment, abuse, and aggression. Nevertheless, several gaps in the literature exist. First, there is little dyadic and observational research into partner maltreatment. Second, there is little application of theoretical models with an interpersonal focus to guide research, policy, and practice. The present study adopted a diathesis–stress framework to understand the dyadic associations between attachment orientations and partner maltreatment, and the moderating role of perceived stress. Method: A total of 172 heterosexual couples completed measures of attachment and perceived stress and participated in a video-recorded discussion task. The interactions were coded for behaviors indicative of partner maltreatment. Actor–partner interdependence models were used to examine actor effects of attachment anxiety or avoidance on one’s own and one’s partner’s perpetration of partner maltreatment. The moderating effect of stress on actor–partner interactions was also examined. Results: Actor attachment avoidance, but not actor attachment anxiety, predicted actor perpetration of maltreatment, whereas both actor attachment anxiety and actor attachment avoidance predicted partner perpetration of maltreatment. Actor attachment avoidance was more positively associated with partner perpetration of maltreatment among actors with higher, relative to lower, levels of stress. Tests of gender invariance revealed few differences between men and women. Conclusions: The findings contribute to an emerging body of literature that uses dyadic modeling and observational approaches to examine the etiology of partner maltreatment and suggest that when contextual and dyadic factors are modeled, the interdependent nature of partner maltreatment cannot be overlooked.