Discriminating the 'good' from the 'bad' therapy hour: A study of dyadic interaction

Haley postulated that the successful therapeutic relationship is metacomplementary - the therapist adopts the more passive role but controls the interaction by structuring the client's behavior. In light of this assertion, we investigated whether 'better' interviews could be discriminated from 'worse' interviews from a) relative client-therapist activity levels and b) the degree of structure implicit in the therapist's messages. Dyads were selected in which therapist and client had congruent perceptions of two sessions, one 'good' and one 'bad.' Group and case-by-case comparisons were made of the natural language in these interviews. Results showed metacomplementary patterns in both good and bad sessions, but client-therapist activity tended to be more asymmetrical in the worse interviews. Specifically, in the bad sessions these therapists either participated even more actively than their clients or were very passive, while in their good sessions client-therapist participation levels were more balanced. Additionally, in the good interviews therapists consistently provided a moderate degree of structure (significantly more reassurance/encouragement, information, and interpretation), as opposed to lower structure (notably reflection/restatement) or higher structure (information seeking and guidance/advise) in their bad interviews. Good/bad contrasts varied remarkably, suggesting that each dyad creates a unique culture that client and therapist evaluate phenomenologically.