Personality-Driven Student Agent-Based Modeling in Mathematics Education: How Well Do Student Agents Align with Human Learners?

TL;DR

This study develops a Big Five personality-based student agent model to simulate human learning behaviors in mathematics education, assessing its fidelity against empirical data to ensure credible simulation for educational research.

Contribution

It introduces a novel student agent model based on Big Five traits with a comprehensive interaction pipeline, evaluating its behavioral alignment with human learners.

Findings

71.4% of student agent behaviors aligned with human learning patterns.

Collected and distilled 13 empirical studies into 14 behavioral criteria.

Validated the credibility of student agents for educational simulations.

Abstract

It is crucial to explore the impact of different teaching methods on student learning in educational research. However, real-person experiments face significant ethical constraints, and we cannot conduct repeated teaching experiments on the same student. LLM-based generative agents offer a promising avenue for simulating student behavior. Before large-scale experiments, a fundamental question must be addressed: are student agents truly credible, and can they faithfully simulate human learning? In this study, we built a Big Five Personality-based student agent model with a full pipeline of student-teacher interaction, self-study, and examination. To evaluate behavioral fidelity, we collected 13 empirical studies on Big Five traits and learning, and distilled them into 14 criteria. We found that the 71.4% of the student agents' behavior was aligned with human learners.