Attachment theory suggests, first, that patterns of dyadic behavior cohere across salient relationships and, second, that such linkages are mediated by working models, defined as cognitive/emotional representations of relationships abstracted from dyadic experience. In this longitudinal study, adolescents' (age 19) Adult Attachment Interview (AAI) coherence ratings and classifications (e.g. working model proxies) were related prospectively to their observed dyadic behaviors with romantic partners in young adulthood (age 20-21). Results demonstrated significant associations between adolescents' representations of their relationships with parents and the later quality of their interactions with romantic partners. Next, a model was tested whereby participants' working models, as inferred from the AAI, mediate the across-time correlation between a subset of observationallv assessed parent-child dyadic behaviors (age 13) and the romantic relationship behaviors of these participants eight years later in young adulthood (age 20-21). Results of mediational analyses were consistent with the fundamental tenet of the organizational-developmental model that salient parent-child experiences are internalized and carried forward into adult relationships.

Disoriented, punitive, and caregiving/role-reversed attachment behaviors are associated with psychopathology in childhood but have not been assessed in adolescence. One hundred twenty low-income late adolescents (aged 18 – 23) and parents were assessed in a conflict-resolution paradigm. Their interactions were coded with the Goal-Corrected Partnership in Adolescence Coding Scales. Confirmatory factor analysis demonstrated that the three disorganized constructs (punitive, care-giving, and disoriented interaction) were best represented as distinct factors and were separable from a fourth factor for collaboration. The four factors were then assessed in relation to measures of attachment disorganization, partner abuse, and psychopathology. Results indicate that forms of disorganized behavior first described in early childhood can also be reliably assessed in adolescence and are associated with maladaptive outcomes across multiple domains.

Attachment theory suggests, first, that patterns of dyadic behavior cohere across salient relationships and, second, that such linkages are mediated by working models, defined as cognitive/emotional representations of relationships abstracted from dyadic experience. In this longitudinal study, adolescents' (age 19) Adult Attachment Interview (AAI) coherence ratings and classifications (e.g. working model proxies) were related prospectively to their observed dyadic behaviors with romantic partners in young adulthood (age 20-21). Results demonstrated significant associations between adolescents' representations of their relationships with parents and the later quality of their interactions with romantic partners. Next, a model was tested whereby participants' working models, as inferred from the AAI, mediate the across-time correlation between a subset of observationallv assessed parent-child dyadic behaviors (age 13) and the romantic relationship behaviors of these participants eight years later in young adulthood (age 20-21). Results of mediational analyses were consistent with the fundamental tenet of the organizational-developmental model that salient parent-child experiences are internalized and carried forward into adult relationships.

Borderline symptoms are thought to emerge from the interaction of temperamental factors and environmental stressors. Both parental invalidation and attachment disorganization have been hypothesized to play an etiological role. However, to date the quality of parent-child interaction has not been observed directly. In this study, 120 young adults were assessed for features of borderline personality disorder on the SCID II, for severity of childhood maltreatment on interview and self-report measures, and for disturbance in parent-child interaction during a videotaped conflict discussion task. Borderline traits, as well as suicidality/self-injury specifically, were associated with more role confusion and more disoriented behavior in interaction with the parent. Among young adults with recurrent suicidality/self-injury, 40% displayed high levels of role confusion compared to 16% of those who were not suicidal. Neither form of disturbed interaction mediated the independent effect of childhood abuse on borderline symptoms. A parent-child transactional model is proposed to account for the findings.