Search and Filter

Tacla: An Llm-Based Multi-Agent Tool For Transactional Analysis Training In Education

Date

AI Overview

Why it matters

Teachers need strong social-emotional skills to manage classroom conflicts, but they report insufficient training opportunities. Existing language model agents lack psychological depth and persona consistency needed for realistic social simulation tools.

Study at a glance
  • Question: Can a multi-agent system grounded in Transactional Analysis realistically simulate student behavior and responses to teacher conflict interventions?
  • Setting and data: Simulated classroom conflict scenario between two artificial student personas (Emma and Jacob), evaluated across 60 simulation runs.
  • Method: Multi-agent system development evaluated with 60 simulated trials across two intervention strategies, scored by large language model evaluators.
  • AI tools: TACLA, GPT-4.1 mini
Key findings
  • Adult-to-Adult teacher responses de-escalated conflict with an average score of 3.733 out of 5, whereas Controlling Parent responses escalated conflict with a score of 2.167.
  • The simulated student dialogues demonstrated high overall conversational authenticity across both intervention conditions, achieving an average realism score of 7.633 out of 10.
  • Adult-to-Adult interventions activated the Adult ego state in both student agents, while Controlling Parent interventions reinforced defensive Child responses and victim positioning.

Summary drafted with AI from the full paper and may contain errors. See the full paper for details.

Simulating nuanced human social dynamics with Large Language Models (LLMs) remains a significant challenge, particularly in achieving psychological depth and consistent persona behavior crucial for high-fidelity training tools. This paper introduces TACLA (Transactional Analysis Contextual LLM-based Agents), a novel Multi-Agent architecture designed to overcome these limitations. TACLA integrates core principles of Transactional Analysis (TA) by modeling agents as an orchestrated system of distinct Parent, Adult, and Child ego states, each with its own pattern memory. An Orchestrator Agent prioritizes ego state activation based on contextual triggers and an agent's life script, ensuring psychologically authentic responses. Validated in an educational scenario, TACLA demonstrates realistic ego state shifts in Student Agents, effectively modeling conflict de-escalation and escalation based on different teacher intervention strategies. Evaluation shows high conversational credibility and confirms TACLA's capacity to create dynamic, psychologically-grounded social simulations, advancing the development of effective AI tools for education and beyond.