Academic journal article Canadian Journal of Behavioural Science

How Do Underlying "Self" and "Other" Dimensions Define Adult Attachment Styles?

Academic journal article Canadian Journal of Behavioural Science

How Do Underlying "Self" and "Other" Dimensions Define Adult Attachment Styles?

Article excerpt


Assumptions about dimensions underlying the four prototype model of adult attachment were explored in two studies. In the first (N = 225), associations between attachment prototype ratings and standardized measures of self and others in relationships were examined. In a second study (N = 246), measures of anxiety and avoidance were added to the analyses. Findings from correlational and hierarchical regression analyses provided support for the notion, contrary to assumptions of the model, that qualitatively different self and other dimensions underlie different attachment styles. Results are discussed in terms of the lack of equivalency in measures of "self and "other" proposed to underly each of the four attachment style ratings, the need to view attachment styles as complementary rather than mutually exclusive, and the need to continue exploration of the dimensions informing attachment.


Les hypothèses relatives aux dimensions qui sous-tendent le modèle à quatre prototypes de styles d'attachement chez l'adulte a fait l'objet de deux études. Dans une première étude (N = 225) les associations entre les évaluations des styles d'attachement et les mesures standardisées des représentations de soi et des autres ont été examinées. Dans une seconde étude, des mesures d'anxiété et d'évitement ont été ajoutées aux analyses. Les résultats des analyses de corrélation et des analyses de régression hiérarchique appuient la conclusion, contraire aux hypothèses postulées par le modèle, que des dimensions qualitativement différentes de la représentation de soi et de l'autre sous-tendent différents styles d'attachement. Les résultats sont discutés en termes de défaut d'équivalence des mesures de soi et de l'autre proposées pour évaluer les quatre styles d'attachement. La discussion insiste également sur la nécessité de poursuivre l'exploration des dimensions qui fondent l'attachement, en considérant les styles d'attachement comme complémentaires plutôt que mutuellement exclusifs.

Researchers interested in measuring adult attachment have drawn extensively on the theoretical work of John Bowlby (1969, 1973, 1980), on the infant attachment prototypes established by Mary Ainsworth and her colleagues (Ainsworth, Blehar, Waters, & Wall, 1978), and on a handful of related prototypical measures developed by several groups of attachment researchers (notably, Bartholomew & Horowitz, 1991; George, Solomon, & Main, 1985; Hazan & Shaver, 1987; Main, 1985). Different measures have been used in the past to assess general attachment styles as well as styles within specific relationship contexts, including, for example, relationships between parents with children, adults with God, and clients with therapists (e.g., Armsdon & Greenberg, 1987; Cicirelli, 1995; Collins & Read, 1990; West & Sheldon-Keller, 1994). In 1998, Fraley and Waller reported that 73% of attachment articles published in three major journals used prototypical rather than dimensional measures to assess adult attachment. A similar review conducted by the authors of this paper revealed that an overwhelming majority of articles (88%) now report either using continuous measures explicitly designed for the purpose of assessing underlying attachment dimensions or dimensional measures adapted from available categorical protocols. This paper focuses on our concerns about current trends in the measurement literature suggesting that a limited set of dimensions underlie prototypical attachment styles. Before addressing these issues, we thought it important to provide a brief historical overview of the contemporary measurement literature related to the assessment of adult attachment styles.

An earlier trend in the literature saw the frequent use of Hazan and Shaver's (1987) categorical measure designed to assess three different attachment styles in the context of romantic relationships. These paralleled the infant styles isolated earlier by Ainsworth and her colleagues (1978). …

Search by... Author
Show... All Results Primary Sources Peer-reviewed


An unknown error has occurred. Please click the button below to reload the page. If the problem persists, please try again in a little while.