Forty-six therapists-in-training listened to an audiotape in which one of three client-attachment-to-therapist styles was portrayed (Secure, Preoccupied-Merger, Avoident-Fearful). Participants completed an expected working alliance inventory for working with the audiotape client, as well as a measure of their own attachment dimensions. Results indicated that client attachment styles predicted differences in expected working alliance ratings, with the Secure audiotape yielding significantly higher total working alliance ratings than either the Preoccupied-Merger of Avoidant-Fearful audiotapes. The client audiotapes yielded the same ratings when therapists' own attachment dimensions were statistically controlled for, suggesting that therapists did not impose their attachment dimensions when predicting working alliance.
This article was based in part on the masters thesis by Rachel E. Crook Lyon at the University of Maryland under the direction of Charles J. Gelso.
Ealier versions of this article were presented at the 108th Annual Convention of the American Psychological Association, Washington, DC, August 2000 and the semi-annual Convention of the Association of Mormon Counselors and Psychotherapists, Salt Lake City, UT, October 2007.