Bartholomew’s two-dimensional model of attachment. 

Context in source publication

Context 1

... do we understand relationships in which abuse appears to be inextricably linked with intimacy? How is it that someone can love, defend, and remain attached to a partner who is psychologically or physically abusive? And how is it that an individual can be assaultive toward someone whom he or she can’t bear to be without? These apparent paradoxes can be understood by looking at two critical tenets of attachment. First, attachment fulfills a basic need for survival (Bowlby, 1988). Thus, the tenacity of the attachment bond is dependent more on maintaining a link to the perceived safety of the attachment figure than to the quality of the attachment relationship. Second, individuals whose attachment needs have been frustrated throughout their relationship history and who feel par- ticularly vulnerable to the potential loss of an attachment figure may strike out violently in order to regain proximity to an intimate partner (Bartholomew et al. , 2001). In the following subsections of this introduction, we expand on this attachment model, explore how it relates to intimate abuse, and review previous research on attachment and relationship abuse. Then we proceed to overview the current study. According to attachment theory, internal representations or working models of close attachment relationships begin in childhood and are incorporated into the developing personality structure, eventually guiding the formation of later social relations outside the family (Bowlby, 1973). Bowlby proposed that these internal working models reflect the extent to which individuals believe themselves worthy of love and attention from others (the self-model) and the extent to which they believe that others will respond to them in a supportive way (the other model). Research in adult peer attachment originated with Hazan and Shaver (1987) who demonstrated conceptual parallels between affectional bonds that are formed between adults and affectional bonds formed between children and their caregivers. Throughout development, changes occur in the content and structure of an individual’s attachment relationships, shifting from asymmetric complimentary attachments (such as the infant–caregiver relationship) to more symmetric or reciprocal attachments (such as adult romantic attachment relationships). Hazan and Zeifman (1994) found that the transfer from complimentary to reciprocal attachments is gradual and that during early adulthood sexual partners tend to ascend to the top of the attachment hierarchy and assume the position as primary attachment figures. To assess individual differences in adult attachment orientations, Bartholomew incorporated Bowlby’s conception of self and other representations in a two- dimensional model of adult attachment (Bartholomew, 1990; Bartholomew & Horowitz, 1991; Griffin & Bartholomew, 1994). Bartholomew identified four pro- totypic attachment patterns in terms of the intersec- tion of Bowlby’s two dimensions of self and other (see Fig. 1). The positivity of the self-dimension, or one’s sense of internalized self-worth, is reflected in an individual’s tendency to be self-confident rather than anxious in close relationships. The positivity of the other dimension, or the perceived supportiveness of close others, is reflected in a tendency to seek out others for support, rather than avoid intimacy. The secure pattern (positive view of self and others) is characterized by high self-esteem, and an ability to establish and maintain close intimate bonds with others without losing a sense of self. The fearful pattern (negative view of self and others) is characterized by low self-esteem and active avoidance of intimacy due to fear of rejection. This fear, however, is coupled with a desire for social contact and approval, resulting in conflicting attachment needs of closeness and distance. The preoccupied pattern (negative view of self and positive view of others) is characterized by low self-worth, excessive dependency on others’ love and approval in close relationships, and an over- involved, demanding interpersonal style. The dismissing pattern (positive view of self and negative view of others) is characterized by a compulsive self-reliance and a defensive downplaying of the importance of intimate relationships. Bowlby (1973, 1982) maintained that the strength of attachment bonds is unrelated to quality of attachment relationships. Threatened individuals will seek proximity to their attachment figure, thereby activating the attachment system and facilitating the formation of attachment bonds. Moreover, individuals will seek proximity to an attachment figure even when the attachment figure is the original source of the threat. Further, because a punitive attachment figure will set up the circumstances for the attachment system to be activated, the attachment bond will not only persist but may even be actively enhanced (Bowlby, 1982). For instance, a parent’s rejection of a child’s efforts to be close often evokes precisely the opposite effect to what was intended. Fearful that proximity to the parent is being jeopardized, a child may become even more clingy in an effort to maintain proximity (see Crittenden, 1988, 1992). Although Bowlby’s theory may be applicable to any victimized individual, this concept has been most exten- sively applied to battered women. Dutton and Painter (1981) proposed a theory of traumatic bonding, which suggests that the power imbalance and intermittency of abuse typical of abusive relationships enhances the strength of emotional bonds to abusive partners. This theory was validated in a study by Dutton and Painter (1993), which showed that women were more strongly attached to their assaultive partners when there was more abuse and the abuse was inconsistent. Dutton and Painter’s theory incorporates the concept of attachment processes, but does not address individual differences in attachment which may be associated with the receipt of relationship abuse. In a sample of women who had recently left an abusive relationship, Henderson et al. (1997) found that 88% of the women had a predominant attachment pattern associated with a negative self-model (fearful or preoccupied), at least double that of a typical nonclinical sample (e.g., Scharfe & Bartholomew, 1994). Further, findings suggested that preoccupied women may be at increased risk for returning to abusive partners (based on their ratings of intentions and feelings), whereas fearful women may have more difficulty disengaging initially (based on abusive relationships of longer duration). Associations between attachment anxiety and abuse receipt also have been found for both men and women in college samples (e.g., Bartholomew et al. , 2001; Roberts & Noller, 1998). There also is evidence of an association between attachment anxiety and perpetration of violence, with research focusing predominantly on male perpetration. Dutton and associates found that assaultive men were more likely to be fearful and preoccupied and less likely to be secure than a matched comparison group (Dutton et al. , 1994). Further, fearfulness and preoccupation were positively correlated with the perpetration of psychological abuse and a constellation of dysfunctional personality traits (anger, jealousy, Borderline Personality Organization, and trauma). Dutton and colleagues explain these findings in terms of “intimacy anger,” suggesting that a violent man’s assaultive episodes represent an adult parallel to the angry protest behavior exhibited by an infant when separated from an attachment figure. They suggest that a man’s violence is often precipitated by the perceived loss of an attachment partner and demonstrates an active effort to bring the attachment figure back. Thus, both fearful and preoccupied individuals, characterized by attachment anxiety, are at risk for high levels of intimacy-anger. Associations between attachment anxiety and male intimate violence have been replicated ...