The attachment script representation procedure in an Italian sample: Associations with Adult Attachment Interview scales and with maternal sensitivity GABRIELLE COPPOLA 1 , BRIAN E. VAUGHN 2 , ROSALINDA CASSIBBA 3 , & ALESSANDRO COSTANTINI 3 1 University ‘‘G. D’Annunzio’’ of Chieti, Italy, 2 Auburn University, AL, USA, and 3 University of Bari, Italy Abstract This study provides data supporting the reliability and validity of an Italian version of the adult attachment script representation task, designed by Waters & Rodrigues-Doolabh (2004). Specifically, we tested hypotheses concerning positive relations between attachment scriptedness scores and two other representational measures, derived from the Adult Attachment Interview (AAI). In addition, we tested the hypothesis that secure base script scores should predict maternal sensitivity in the context of mother – infant interaction. Thirty-one mothers completed narrative protocols and received scriptedness scores using the Waters & Rodrigues-Doolabh (2004) criteria. Prior to the attachment script assessment, mothers had been assessed using the AAI and had been observed in the context of infant– mother interactions to assess maternal sensitivity. Assessment of instrument reliability was satisfactory (Cronbach’s a 4.70) and both hypotheses were supported; the attachment scriptedness score (based on 4 attachment narratives) was positively and significantly associated with the AAI coherence score, the continuous security score derived from the AAI State of Mind scales, and with maternal sensitivity. These data extend to another socio-cultural milieu, previous findings supporting reliability, convergent, and predictive validity of the attachment script representation task as a measure of adult attachment. Keywords: Secure base script, Adult Attachment Interview, maternal sensitivity Introduction Although Bowlby (1973) first introduced and defined the notion of internal working models, it has been only during the last 20 years, in line with more recent contributions from cognitive psychology (Neisser, 1986; Shank & Abelson, 1977), that research in the field of attachment has expanded the focus from attachment behaviors and interactions to the cognitive organization of attachment representations. Internal working models, built upon repeated experiences of interaction with the attachment figure, are no longer considered as a set of attachment relevant information stored in a dual format, either episodic or semantic (Bowlby, 1973), but rather as a more complex and differentiated system of mental representations concerning attachment (Bretherton, 1990, 1992). This system is thought to be hierarchically organized: at a very basic level, these models include episodic information Correspondence: Gabrielle Coppola, Dipartimento di Scienze Biomediche, Facolta ` di Psicologia, Universita ` ‘‘G. D’Annunzio’’ di Chieti-Pescara, Blocco A di Psicologia, Campus Universitario, Via dei Vestini n. 31, 66100, Chieti Scalo, Italy. E-mail: g.coppola@unich.it Attachment & Human Development, September 2006; 8(3): 209 – 219 ISSN 1461-6734 print/ISSN 1469-2988 online Ó 2006 Taylor & Francis DOI: 10.1080/14616730600856065