Examining specificity in psychotherapy: Open/hidden administration of alliance in experimentally induced guilt

Placebo-controlled trials face significant challenges in psychotherapy research due to difficulties in creating credible placebo conditions and the inability of blinding therapists. We tested the open/hidden administration as an alternative approach to differentiate between specific and nonspecific treatment effects.

68 healthy subjects were randomized to receive one session of person-centered therapy to implement therapeutic alliance either with their knowledge (open; n = 23), without their knowledge (hidden; n = 22), or no treatment (control; n = 23) following experimental guilt induction. The primary outcome was guilt responses after the guilt induction. Secondary outcomes were shame, pride, and affect responses. We hypothesized that both alliance conditions would reduce guilt compared to control, with no difference between alliance conditions.

Control participants showed significantly higher guilt responses than alliance conditions (p < .001, d = 1.07), whereas guilt did not differ significantly between open and hidden conditions (p = 0.810, d = 0.07). Similar patterns emerged for shame and pride, but not for affect.

These results provide evidence that the open/hidden administration offers a novel approach to test the specificity of treatment components in psychotherapy research. Furthermore, our findings indicate that the therapeutic alliance is effective regardless of the provision of a treatment rationale.