Can Relationships fix Relationships? Attachment Models of Adoptive Mothers and Attachment Patterns of Their Late-Adopted Children

The pre-adoption experiences of late-adopted children is frequently characterized by a wide range of difficulties in their families of origin, ranging from neglect, rejection, to maltreatment and sexual abuse (Howe, 1998). Due to early dysfunctional interactions late-placed children develop insecure or disorganized internal representations of attachment relationships (referred to as Internal Working Models-IWMs) that are characterized by negative expectations with regard to relationships with significant adults (Van IJzendoorn & Juffer, 2006). Many authors (Steele, Hodges, Kaniuk, Steele, 2010) have shown that although these IMWs are resistant to change, revision and further development is possible throughout one’s life as one enters new experiences. In this context, adoption has often been described as one of the most radical and powerful intervention (Rutter et al. 2007). Moreover, recent research have highlighted that children adopted by caregivers with secure IWMs demonstrated more secure attachment to them than children placed with insecure adoptive parents (Verissimo and Salvaterra, 2006; Steele et al., 2008). The purpose of this pilot-study is to verify a revision of the attachment patterns of late-placed children from insecurity towards security and to evaluate the role of the maternal attachment security in making this revision easier. This is a brief-longitudinal study consisting of two measurements with adoptive mothers and late-adopted children: the first within two months from adoption, the second six months later. The experimental sample included 28 “recently” adoptive dyads - with late-adopted children aged 4 to 7 (mean: 70 months, SD = 12.75) - and the control sample consisted of 12 genetically related dyads. The children’s behavioural patterns of attachment were assessed using the Separation-Reunion Procedure (SRP, Cassidy, 1988) and the children’s attachment narrative representations using the Manchester Child Attachment Story Task (MCAST, Green, Stanley, Smith e Goldwyn, 2000). The attachment IWMs of the adoptive mothers were classified using the Adult Attachment Interview (Main, Goldwyn, Hesse, 2003). All the coding systems required the attribution of three main categories, plus one secondary classification, as reported in table 1. At the beginning of the adoption the behavioural attachment pattern of the late-adopted children are significantly more insecure than those of non adopted peers (p = 0.002). After 6 months, the distribution of the attachment behavioural categories did not show a significant difference (p = 0.268) in the secure/ insecure distribution comparing the adopted/ non adopted samples. The results showed a significant revision of the attachment behavioral patterns of late-adopted children from insecurity towards security (p = .002, see table 2). Furthermore, late-adopted children showing such revision had predominantly adoptive mothers with secure IWMs, compared to children placed with insecure mothers (p = .047). However, the concordance between the mothers’ AAI classifications and the children’ MCAST categories did not show any statistical significance (phi = .126, p = .516). These results could suggest that attachment IWMs of late-adopted children could be revised if there is a relevant change in their relational context of life. Moreover, secure maternal IWMs could permit insecure children to reactivate or modulate attachment related emotions, needs, feelings and behaviours.