Attachment research has provided useful concepts towards an interpersonal approach to psychopathology. Although it suggests that psychological processes are defined by three relational dichotomies (self vs. other, relatedness vs. autonomy, and dependent vs. dependent-on positions), it has not yet assessed separately and simultaneously the eight components generated by the intersection of these dichotomies (e.g. self requesting relatedness, self providing relatedness etc). The aim of the present research was to investigate the validity and potential clinical usefulness of such a neglected approach.

In Study 1, thirty Cognitive Analytic Therapy outpatients at Guy's Hospital, London, were interviewed before treatment on their couple relationships and filled out the Brief Symptom Inventory (BSI) and Inventory of Interpersonal Problems (IIP). A content-analytic method was applied to the transcripts. In Study 2, a new attachment questionnaire was completed by 400 undergraduates, along with the BSI and IIP.

In Study 1, at a six-month follow-up, BSI scores were predicted by the Self Requests Autonomy [R Sq = .29, F(1,23) = 10.84, p = .003] and the Other Provides Closeness [R Sq = .23, F(1,23) = 9.72, p=.001] attachment dimensions, while IIP scores were predicted by Self Provides Support [R Sq = .11, F(1,23) = 7.87, p=.003]. In Study 2, factor analysis yielded eight components matching the theoretical expectations and correlating with BSI and IIP scores (rs ranging from .18 to.41).

The present findings provide some preliminary support for the validity and potential clinical relevance of the method.