A model of first-session therapeutic alliance development is proposed, wherein the interpersonal styles of the client and the therapist interact to influence the quality of the alliance that develops. It is proposed that when these individuals’ interpersonal styles do not fit together well, some flexibility of interpersonal strategies or styles is helpful to ensure alliance formation. Quality of fit of interpersonal styles is defined within the interpersonal circumplex model as complementarity, which is characterized by similar levels of communion (warmth, affiliation, engagement) and reciprocal levels of agency (dominance, control, influence). Flexibility of interpersonal strategies is defined within the interpersonal circumplex model as vector length or profile definition. This variable represents the frequency and intensity with which interpersonal behaviors of one characteristic kind are used at the expense of other behavioral options.
In the present studies, complementarity was tested as a moderator of the relationship between vector length and the quality of the therapeutic alliance in five archival data sets. These five data sets included three psychotherapy samples, one sample of rheumatoid arthritis patients, and one analogue study featuring videotaped reenactments of psychotherapy transactions. Measures of complementarity and vector length varied. Therapeutic alliance was measured with the Working Alliance Inventory in each study.
The proposed model was not supported in any of the five studies, nor were the supporting hypotheses upheld. Methodological issues may account for the null findings, including small samples, mismatched measures and consequent standardization of data, and the use of trait ratings instead of interaction-level ratings. Limitations of archival samples are discussed, and directions for future research are enumerated.