Adult Attachment Representations and Secure Base Use and Support in Couple Problem Solving Interactions

Abstract

Validity is more than a rich suite of empirical correlates. They have to be the right correlates. In the case of the Adult Attachment Interview, validation includes showing links to the secure base concept, the central concept in Bowlby's and Ainsworth's attachment theory. Important evidence for the AAI's validity comes from longitudinal research linking infant Strange Situation classifications to the same person's AAI classifications 20 years later (e.g., Waters et al. 2000). In addition, Crowell et al. (2002) provided strong support for the attachment relevance of the AAI by showing that narrative coherence in the AAI is significantly related to secure base use and support in couple problem solving interactions. Interestingly, the secure base concept is not explicitly mentioned in the AAI scoring manual. This raises a question as to whether early representations of secure base experience play an important role in the AAI, or are they replaced by a different kind of representation as cognitive abilities mature and a person begins to organize their personality around an elaborate personal narrative (see McAdams & Pals, 2006).The AAI Secure Base Scale (AAI_SB scale) is a newly developed scale (Waters, Waters, & Crowell, in preparation) designed to measure a person's knowledge and access to script-like representations of secure base experience from a subset of the AAI questions. This dissertation examines the AAI_SB scale's links to AAI Coherence and Mother/Father Accepting and Rejecting scales. In addition to evaluating the scale's discriminant validity, it tests the hypothesis that secure base script knowledge explains most of the relation between AAI Coherence and adults' secure base use and support in videotaped problem solving interactions. Data for the study was selected from the original Crowell et al. (2002) study. A sample of 60 AAI transcripts were scored using the AAI_SB scale. Correlations were used to evaluate convergent and discriminant validity in relation to conventional AAI Coherence, Acceptance, and Rejection scales. In addition, hierarchical multiple regression analysis was used to test the hypothesis that script knowledge accounts for most of the secure base use and support variance explained in the original Crowell et al. (2002) study. The correlation between AAI Coherence and Secure Base Script knowledge was .64, p  .01. In addition, Secure Base Script knowledge accounted for most of the correlation of AAI Coherence with Secure Base Use and Secure Base Support in the couple problem solving interactions.