Simulating Couple Conflict: Designing A Multi-Agent System for Therapy Training and Practice

TL;DR

This paper introduces a multi-agent simulation system for couples therapy training that models emotional dynamics and conflict patterns, providing a realistic and controllable environment for therapists to practice and improve their skills.

Contribution

It presents a novel multi-modal, stateful simulation framework that models therapy as a multi-agent dynamical system with structured interaction stages, enhancing realism and control.

Findings

Therapists more accurately identified state transitions in the simulation.

Participants rated the system as more realistic and responsive than baseline.

The simulation effectively models demand-withdraw conflict patterns.

Abstract

Couples therapy requires managing complex, evolving emotional dynamics between partners, but traditional training methods for therapists, like role-play, lack realism, consistency, and control. We present a multi-modal simulation that models therapy as a controlled, multi-agent dynamical system with structured interaction stages. Therapists practice with a pair of client-agents who go through six evolving stages that respond to therapist actions. This simulation enables practice with demand-withdraw conflict patterns in a closed-loop environment. The simulation uses a sense-plan-act architecture: it detects the therapist's input, updates agents' interaction states based on psychotherapy theory and transcript analysis, and generates realistic verbal and emotional responses. In an experiment with 21 licensed U.S. therapists, participants more accurately identified state transitions and rated the system as more realistic and responsive than a prompt-based baseline, demonstrating the value of stateful, interpretable simulation for therapist training.