Electronic Thesis and Dissertation Repository


Doctor of Philosophy




Dr. Greg Moran


Traditional attachment theory suggests that because maternal state of mind regarding attachment is generally stable by adulthood, mothers should interact similarly with their children and, consequently, should share a similar quality of attachment with each. Early empirical work, however, suggests that the quality of siblings’ relationships is frequently different. Using varied theoretical and methodological approaches, this dissertation expanded upon the existing literature to further explore the nature and underpinnings of variability in the quality of mother-infant attachment across siblings.

Study 1 comprehensively described patterns of attachment within the family, investigating the extent to which the quality of siblings’ relationships with their common mother are a) similar to each other; and b) consistent with the quality of maternal state of mind. While concordance in family classifications was common, so was non-concordance – unexpectedly so, from the perspective of traditional theory, indicating that patterns of attachment within the family are complex and warrant further exploration.

Studies 2 and 3 explored the roots of variability in family attachment relationships. Study 2 examined links between family attachment patterns and maternal mentalization, which reflects mothers’ capacity to represent their child in terms of mental states. Patterns of mentalization across siblings did not vary with maternal state of mind, nor were they linked with similarity in siblings’ attachment classifications. However, when attachment security was represented as a continuous dimension, similarity in mentalization across siblings was associated with similarity in their relationship quality.

Study 3 returned to the field’s traditional focus on maternal sensitivity, but examined its role from a family systems perspective including shared and non-shared components. Shared sensitivity contributed to similarity in the quality of siblings’ relationships, but non-shared sensitivity did not account for differences. This work also revealed that siblings’ relationship quality diverged even when assessed continuously and contemporaneously, complementing previous work based on categories assigned at different ages.

Together, these studies highlighted that variability in family attachment is a normative phenomenon requiring more comprehensive integration into theory and research. Future directions for the field are discussed, including the utility of applying approaches from beyond the realm of attachment to advance research in this area.