A transdiagnostic conflict-square algorithm: a four-node computational framework for psychotherapy and functional diagnosis

Clinical services need more than categorical labels to guide safe dosing, maintain alliance, and plan functional recovery. The Conflict-Square Algorithm (CSA) offers a compact bedside grammar for moment-to-moment decisions during psychotherapy. Clinicians track four observable signals—defense, anxiety/affect tolerance, progression, and superego/shame—and gate intervention intensity by three safety thresholds (A–C). Each clinically meaningful moment is summarized as one auditable episode line in plain language: trigger, observable response, threshold, action, and expected functional impact (Mini-ICF-APP). To strengthen reproducibility, we provide i) operational threshold definitions with observable markers and common misclassification errors, grounded in established anxiety-channel descriptions [striated muscle, smooth muscle, and cognitive-perceptual disruption (CPD)]; ii) a short scope and contraindication checklist; iii) several consecutive worked micro-episodes demonstrating node shifts, threshold transitions, and dose modulation over time; and iv) a minimal machine-readable schema plus a threshold-gating state diagram. As a proof-of-concept feasibility demonstration, we report aggregate coding statistics from three published psychotherapy training videos distributed by the ISTDP Institute (transcribed for analysis with written permission; N = 2,809 speaker turns) using a three-label therapist intervention mapping (invite progression/defense work/anxiety regulation) aligned with CSA nodes. CSA is presented as a teachable, testable decision framework—not as a validated diagnostic instrument—and we outline a pragmatic validation program (rater agreement, safety-rule adherence, usability, and functional outcomes) and future multimodal extensions (e.g., optional physiological monitoring for biofeedback and threshold detection).