Curvilinear relations between insight-, cognition-, and skills-oriented technique use and outcome across treatments, or Goldilocks and the three psychotherapies

Abstract

Objective: A "Goldilocks" effect is when "just right" conditions are present for a phenomenon to occur. In psychotherapy research, the Goldilocks effect is when moderate intervention levels (not very low or high) correlate with more improvement than very high or very low levels. Finding curvilinear relations requires the wider range of what a specific intervention can take, which can be seen when examining technique and outcome across different treatments rather than within a single therapy orientation.

Method: For 151 patients with panic disorder (66% female, 33% racial/ethnic minority) in a randomized comparative trial of panic-focused psychodynamic psychotherapy (B. L. Milrod et al., 1997), panic control therapy (Craske et al., 2000), and applied relaxation training (Schwalberg & Chambless, 2006), observers assessed insight-, cognition-, and skills-focused techniques using the multitheoretical list of therapeutic interventions (McCarthy & Barber, 2009) from Weeks 1, 5, and 9 session recordings. Outcome was assessed at Weeks 1, 5, 9, and termination by the Panic Disorder Severity Scale (Shear et al., 1997).

Results: When looking across treatments, very high or low (not moderate) insight-oriented interventions were associated with the most symptom improvement by the subsequent assessment point and at termination. Moderate (not very high or low) skills- and cognition-oriented interventions correlated with more improved outcome at the subsequent assessment and termination. These findings describe when interventions are used in general but not within a specific protocol treatment.

Conclusions: Curvilinear relations between technique and symptom change might more closely depict how interventions relate to outcome than might more conventional linear approaches. (PsycInfo Database Record (c) 2025 APA, all rights reserved).