A new method to assess attachment representations in children by applying a story completion procedure in doll play (SCPDP) is presented. Transmission and continuity of attachment were tested in 28 German families by using the Strange Situation procedure (SS) with the mother and her infant, the Adult Attachment Interview (AAI) with the mother when the child is 5 years old, and the attachment representations (SCPDP) of the 6-year-olds. Relying on the twofold distinction (secure/insecure), results revealed a significant continuity of attachment from 1 to 6 years of age, and a correspondence between maternal AAI and child's attachment quality in SS, as well as a correspondence between maternal AAI and the 6-year-olds' attachment representations. Using configural frequency analysis, we found continuity in patterns of security and insecurity when looking across the measures of infant and pre-school attachment and maternal adult attachment. Processes underlying the high match of mothers' and children's attachment representations are discussed.

Attachment in middle childhood increasingly attracts the interest of developmental psychologists and clinicians. Recent studies using attachment narratives elicited by story stems reported gender-specific aspects of attachment development, potentially evoked by developmental tasks during this period of the life span. There is evidence that children with risk factors present more insecure and disorganized attachment narratives compared to children without risk. Yet, there is little research concerning the joint effects of gender, risk, and age for attachment classifications. The paper presents a pooled analysis of 22 samples (eight risk samples) including 887 children (411 girls), aged between 4.5 and 8.5 years who were assessed with the same "German Attachment Story Completion Procedure" (GASCP). Girls were 1.8 times more likely to present secure and 0.4 times less likely to present disorganized narratives compared to boys when controlling for risk status and age. Children from risk samples were more likely (odd ratio 5.4) to display disorganized and less likely to show a secure attachment (odd ratio 0.3) compared to those from no-risk samples in multilevel logistic regressions. Remarkably, the effect of risk was not moderated by age and gender, and gender effects were not moderated by age.

Viewed from the perspective of attachment theory, coping with the separation and divorce of parents requires that children reorganize their mental model of attachment. Secure attachment models may be disrupted, while insecure attachment models may be strengthened. According to findings from research on divorce, this process of family reorganization takes about two years. Attachment representations in middle childhood can be assessed by new methods that rely on the symbolic medium of story completion in doll play. Based on five coded attachment story narratives, the Attachment Story Completion Task (ASCT) by Bretherton and Ridgeway, and the German Coding and Classification System by Gloger-Tippelt and König, permits the assessment of secure, avoidant, ambivalent and disorganized attachment representations reliably, and to rate attachment security on a four-point-scale. In this study we compare the attachment representations in ASCT of 51 children from two parent families with 60 six-year-old children from one parent, all mother families who experienced separation/divorce, and were still in the phase of family reconstruction. The children of single mothers demonstrated fewer secure and more insecure attachment representations, in particular more avoidant ones. Moreover, in the children from one parent, but not from two parent families a significant gender effect emerged. Results have been discussed considering further risk factors in the one parent families and possible influences of children's attachment representations in the school setting.

A phase model describing the course of the woman’s first pregnancy is proposed. In complementing the dominant treatments of pregnancy within medical and clinical domains, a more extended view of pregnancy as both a biological and psycho-social process is outlined. The pregnancy course is considered an integral segment within the woman’s full life course – both dependent upon and influencing other segments. In the present model, four ideal phase types are distinguished: a disruption phase of radical change, an adaptation phase of readjustment, a centering phase in which the task of production is central, and a final phase of anticipation of birth and preparation for the expected child. Within each phase three levels of description are separated: a biological, psychological and social level. Special attention is given to the mutual interaction among levels, along with the effects of preceding conditions on the phases of pregnancy. Finally, implications of the present model for research and application are considered.

No abstract