Conversational Learning Diagnosis via Reasoning Multi-Turn Interactive Learning
2026-03-03 • Computers and Society
Computers and Society
AI summaryⓘ
The authors developed ParLD, a new method to better understand how students think during conversations with tutors in learning settings. Unlike previous methods that directly analyzed dialogue without a strong psychological basis, ParLD breaks down the process into three steps: predicting student behavior, analyzing dialogue and behaviors to update the cognitive state, and reasoning about future responses to check its own work. These parts work together to provide more reliable and insightful information about students' learning progress. The authors tested ParLD and found it effective in predicting student performance and supporting tutoring.
learning diagnosiscognitive statelanguage modelsconversational learningmulti-agent collaborationbehavior schemadialogue analysisself-reflectionperformance predictiontutoring support
Authors
Fangzhou Yao, Sheng Chang, Weibo Gao, Qi Liu
Abstract
Learning diagnosis is a critical task that monitors students' cognitive state during educational activities, with the goal of enhancing learning outcomes. With advancements in language models (LMs), many AI-driven educational studies have shifted towards conversational learning scenarios, where students engage in multi-turn interactive dialogues with tutors. However, conversational learning diagnosis remains underdeveloped, and most existing techniques acquire students' cognitive state through intuitive instructional prompts on LMs to analyze the dialogue text. This direct prompting approach lacks a solid psychological foundation and fails to ensure the reliability of the generated analytical text. In this study, we introduce ParLD, a preview-analyze-reason framework for conversational learning diagnosis, which leverages multi-agent collaboration to diagnose students' cognitive state over multiple dialogue turns. Specifically, ParLD comprises three main components: (1) Behavior Previewer, which generates a student behavior schema based on previous states and learning content; (2) State Analyzer, which diagnoses the tutor-student dialogue and behavior schema to update the cognitive state; and (3) Performance Reasoner, which predicts the student's future responses and provides verifiable feedback to support ParLD's self-reflection with the Chain Reflector. They operate sequentially and iteratively during each interaction turn to diagnose the student's cognitive state. We conduct experiments to evaluate both performance prediction and tutoring support, emphasizing the effectiveness of ParLD in providing reliable and insightful learning diagnosis.