A new 4-group model of attachment styles in adulthood is proposed. Four prototypic attachment patterns are defined using combinations of a person's self-image (positive or negative) and image of others (positive or negative). In Study 1, an interview was developed to yield continuous and categorical ratings of the 4 attachment styles. Intercorrelations of the attachment ratings were consistent with the proposed model. Attachment ratings were validated by self-report measures of self-concept and interpersonal functioning. Each style was associated with a distinct profile of interpersonal problems, according to both self-and friend-reports. In Study 2, attachment styles within the family of origin and with peers were assessed independently. Results of Study I were replicated. The proposed model was shown to be applicable to representations of family relations; Ss' attachment styles with peers were correlated with family attachment ratings.This article describes a new model of attachment styles in adulthood. Drawing on the theory of Bowlby (1973, 1980, 1982a), two types of internal working models are postulatedan internal model of the self and an internal model of others. Each internal model can be dichotomized as positive or negative to yield four theoretical attachment styles. This article summarizes the relevant childhood attachment literature, reviews recent work on adult attachment, describes the new model, and then presents two empirical studies designed to validate the proposed model.

A basic principle of attachment theory is that early attachment relationships with caregivers provide the prototype for later social relations. Working within an attachment framework, a new 4-group model of characteristic attachment styles in adulthood is proposed. In particular, two forms of adult avoidance of intimacy are differentiated: a fearful style that is characterized by a conscious desire for social contact which is inhibited by fears of its consequences, and a dismissing style that is characterized by a defensive denial of the need or desire for greater social contact. This distinction corresponds to two differing models of the self: people who fearfully avoid intimacy view themselves as undeserving of the love and support of others, and people who dismiss intimacy possess a positive model of the self that minimizes the subjective awareness of distress or social needs. The emotional and interpersonal ramifications of the two proposed styles of adult avoidance are discussed.

Three studies assessed the construct validity of the self-and other-model dimensions underlying the 4-category model of adult attachment (Bartholomew, 1990). Five methods were used to assess the hypothesized dimensions: self-reports, friend-reports, romantic partner reports, trained judges' ratings of peer attachment, and trained judges' ratings of family attachment. In each study, the convergent and discriminant validity of the dimensions were assessed by multitrait-multimethod matrices and by confirmatory factor analysis. Study 2 related the latent attachment dimensions to theoretically relevant outcome latent variables. As predicted, individuals' self models converged with direct measures of the positivity of their self-concepts, and individuals' other models converged with direct measures of the positivity of their interpersonal orientations. Study 3 related the latent attachment dimensions to 3 alternate self-report measures of adult attachment and showed that the 2 dimensions serve as an organizing framework for the different measurement approaches.

A basic premise of attachment theory is that internal models of attachment remain relatively stable across the life span. We examined stability and change in adult attachment representations over 8 months in a sample of young adults (N = 144; Mean age = 24.5 years). Attachment patterns were assessed by categorical and continuous ratings across three methods—self‐report ratings, expert ratings based on semi‐structured interviews, and reports of romantic partners—and showed moderate stability. Interview ratings tended to show higher stability than self‐report ratings. Changes in attachment ratings (across methods) were not consistently related to life events that had occurred in the intervening 8 months. The relationship between reliability and stability was discussed, and where the “true” stability could be estimated independent of unreliability, it was found to be very high (r's ranging from .72 to .96). The results highlight the importance of using multiple indicators in assessing adult attachment and using continuous rather than categorical ratings in the assessment of stability.

This study examined the characteristics of attachment hierarchies in young adulthood. Multiple components were used to assess attachment bonds: using the attachment figure as a safe haven in times of distress, using him or her as a secure base from which to venture out independently, having a strong emotional tie with the person regardless of whether the tie is positive, negative, or mixed, seeking to be in close proximity to the person, and mourning the loss of the person. The Attachment Network Questionnaire (ANQ) was developed to measure multiple adult attachment relationships and to examine the characteristics of attachment hierarchies. 223 university students completed the ANQ by listing their significant relationships and then ranking these persons in terms of the various components of attachment. A subset of the participants was followed up to examine the one month test-retest reliability of the ANQ. Young adult participants were found, on average, to have 5.38 attachment figures, including family members, romantic partners, and friends. The figures identified included both secure and insecure attachments. In addition, the ANQ demonstrated adequate test-retest reliability over one month.

The Inventory of Interpersonal Problems (IIP) has been used to identify dysfunctional patterns in interpersonal interactions. Interpersonal problems can be organized in two dimensions, and the two-dimensional space can be divided into eight equal sectors (octants). Subscales of the IIP describe each of these octants. The instrument has been used to identify (a) interpersonal problems that are discussed most often in a brief dynamic psychotherapy and (b) problems that are treated most easily. The results show that problems in the "exploitable" octant improve most frequently, whereas problems in the "dominating," "vindictive," and "cold" octants do not improve as readily. Attachment styles in adulthood were examined (following a model proposed by Bowlby), and different attachment styles were found to correspond to different types of interpersonal problems. Finally, these variables were related to the ability to describe other people clearly. The article also discusses implications for brief dynamic psychotherapy.

One hundred and twenty men referred for treatment for wife assault and forty demographic controls completed self‐report questionnaires assessing attachment patterns, anger, jealousy, Borderline Personality Organization, and trauma symptoms. This constellation, with the exception of attachment, has been found to represent a profile related significantly to the frequency of both verbal and physical abuse. In the present study, attachment was found to correlate significantly with these other measures. A measure of secure attachment correlated significantly, but negatively, with the constellation measures. A measure of fearful attachment correlated significantly and positively. A preoccupied attachment pattern correlated with the other measures less strongly than did the fearful measure. A dismissing measure was not correlated with any constellation measures. The role of attachment anger is discussed; it is suggested that fearful attachment could also be described as angry attachment. The concept of intimacy anger is introduced from early studies in attachment and applied to the explanation of assaultiveness in adult relationships.

This study explored the intergenerational transmission of violence in a community sample. A telephone survey of 1,249 adults in the City of Vancouver assessed family-of-origin violence (father to mother, mother to father, father to self, and mother to self), as well as physical and psychological abuse in intimate relationships. All forms of family-of-origin violence were predictive of all forms of relationship abuse, consistent with a general social learning model of relationship violence. There was no evidence of gender-specific or role-specific patterns of transmission. For example, father-to-mother violence was not specifically predictive of men's perpetration and women's victimization in adult relationships. Nor was parent-to-self violence more predictive of victimization than perpetration. The methodological and theoretical implications of these findings are discussed.