Follow You, or Follow Me? Examining Therapist Responsiveness to Client and Responsiveness to Self Using Differential Equations Model and Multilevel Data Disaggregation
Publication or External Link
This study examined the effects of therapist responsiveness on client’s perception of working alliance in their first counseling session, using the Ordinary Differential Equations (ODE) model to quantify the mutual influence and responsiveness dynamics between the therapist-client dyad, and multilevel data disaggregation to disentangle the between-therapist and within-therapist effects. Based on the framework in Interpersonal Theory (Kiesler, 1988), archival video recordings of 111 sessions, coming from the first counseling session from 38 therapists and their 111 clients in a university clinic, were rated by 11 trained undergraduate raters on therapists’ and clients’ respective levels of Control and Affiliation on 2-minute segments. The rating data were entered into the ODE model to derive the dynamic coefficients capturing therapists’ responsiveness to clients and to themselves in that session, which were then disaggregated into between-therapist and within-therapist components. Responsiveness was operationally defined as the extent to which the therapist changes (increases or decreases) their controlling or affiliative behaviors given the level of the therapist’s and client’s previous control or affiliation. Using clients’ evaluation of the first-session working alliance as the dependent variable, Hierarchical Linear Modeling results indicated that only for the Control dimension at the between-therapist level did therapist responsiveness significantly predict client report of working alliance. Specifically, client working alliance ratings for the first session were highest for therapists who were generally responsive in an anti-complementary way along the Control dimension across their clients on caseload (e.g., working alliance was high when therapist increased their level of control in response to a higher level of client or when therapist decreased their level of control in response to a lower level of client control), were moderate for therapists who were generally non-responsive, and lowest for therapists who were generally responsive with their clients in a complementary way (e.g., working alliance was low when therapist increased their level of control in response to a lower level of client control or when therapist decreased their level of control in response to a higher level of client control). None of the other associations were significant. Theoretical and practical implications, limitations, and future directions were discussed.