Objectives. Inflexible social interaction patterns are defining features of borderline personality disorder (BPD). Specific beliefs about the self and others may be activated across interaction situations, often leading to instable relationships. It may be pivotal to address these difficulties in early treatment phases, through appropriate therapist responsiveness, which means an adaptation of therapist's activity to their client's behaviours using emerging information in the process (Stiles, 2009, Clinical Psychology: Science and Practice, 16, 86). Design. In this process-outcome study, responsiveness is operationalized by the motive-oriented therapeutic relationship (Caspar, 2007, Handbook of psychotherapeutic case formulations, 2nd ed., 251-289, Guilford), based on the Plan analysis case formulation. The present study assesses the interplay between social interaction problems and therapist responsiveness, explaining symptoms at discharge and the therapeutic alliance. Method. In total, N = 50 clients with BPD entered the study, and standard and responsive treatments were compared. Social interaction patterns were assessed by the newly developed Borderline Interaction Patterns Scale (BIPS), applied to recorded material of three sessions per therapy. Outcome was measured by general symptoms (OQ-45), borderline symptoms (BSL-23), interpersonal problems (IIP), and the therapeutic alliance (WAI). Results. Results suggest that in standard treatment, social interaction patterns are neither related to outcome nor the therapeutic alliance. In responsive treatment, more activation of social interaction patterns predicted better outcome on IIP and lower therapist ratings of the alliance.