Attachment Scriptedness Among Preschoolers: A Validity Study of a Revised Scoring System

• In Event: Poster Session 13
  In Poster Session: PS 13 Topic - Parenting & Parent-Child Relationships

Abstract

The construct of attachment representations is at the center of attachment theory. Bowlby (1982) proposed that such representations are based on children’s experiences during child-parent interactions and guide both the interpretation of new attachment-related information and attachment behavior. How knowledge about attachment relationships is organized has proven challenging to assess and much more so, when addressing young children’s budding representations. Waters and colleagues (1998) suggested that young children represent attachment information as a secure base script that can be assessed through their narratives, and proposed a scale to evaluate the extent to which the organization of children’s narratives reflect such script (i.e., scriptedness). However, the development and validity of the scoring system, has scarcely been undertaken (Posada et al., 2018). In this study we tested the concurrent validity of a revised version of the scoring system that assesses the scriptedness of preschoolers’ narratives vis-à-vis key constructs in attachment theory, namely, child secure base behavior, maternal sensitivity, and maternal secure base scriptedness in two samples.

In study 1, 81 middle-class families participated. Mothers (M= 33.5 years old) were the child primary caregiver. Children were 44 months old on average (51% girls). Dyads were visited three times (1 home and 2 park visits). Children’s attachment representations were assessed with the MacArthur story completion task (spilled juice, hurt knee, and monster in the bedroom story stems; Bretherton et al., 1990) and were obtained during the home visit. The narratives were coded independently by two coders (ICC=.83) on a 5-point Likert scale. Two different observers provided independent ratings of child secure base behavior using the Attachment Q-Set (Waters, 1995) at each visit (inter-observer reliability=.79). Two researchers different from those observers of child behavior provided ratings of mothers’ behavior using the Maternal Behavior with Preschoolers Q-Set (Posada et al., 2007) at each visit (inter-observer reliability=.82). Security and sensitivity scores for each dyad were obtained by averaging all descriptions of child and mother behavior, respectively.

In study 2, 50 socioeconomically and ethnically diverse families participated. Mothers (M= 34 years old) were the child’s primary caregiver. Children were 50.1 months old on average (48% girls). Dyads completed a home visit and a lab visit during infancy, at which time mothers’ scriptedness was assessed by coding an abbreviated Adult Attachment Interview (AAI) for secure base script knowledge (Waters et al., 2013). In a lab visit at preschool-age, mother-child dyads were video-recorded during a play interaction, which was coded for maternal sensitivity, and children’s attachment representations were assessed using the same procedure as describe above for study 1. Children narratives were coded independently by two coders (ICC=.85).

Analyses (Table 1) indicate that children’s narrative scriptedness was significantly related to child secure base behavior (study 1), maternal sensitivity (studies 1 and 2), and their mothers’ scriptedness (study 2). Child verbal ability was not significantly associated with secure-base scriptedness (study 1 and 2). Results are discussed regarding the validity of the revised coding system and the consistency of findings between developing attachment representations and other key constructs in attachment theory.

Authors

• Galia Nissim, Stony Brook University
  Presenting Author

• Kristin L Bernard, Stony Brook University
  Non-Presenting Author

• Jill M Trumbell, University of New Hampshire
  Non-Presenting Author

• German E Posada, Purdue University, West Lafayette
  Non-Presenting Author