Simulating Students with Large Language Models: A Review of Architecture, Mechanisms, and Role Modelling in Education with Generative AI
By: Luis Marquez-Carpintero, Alberto Lopez-Sellers, Miguel Cazorla
Potential Business Impact:
Lets computers act like students to test teaching.
Simulated Students offer a valuable methodological framework for evaluating pedagogical approaches and modelling diverse learner profiles, tasks which are otherwise challenging to undertake systematically in real-world settings. Recent research has increasingly focused on developing such simulated agents to capture a range of learning styles, cognitive development pathways, and social behaviours. Among contemporary simulation techniques, the integration of large language models (LLMs) into educational research has emerged as a particularly versatile and scalable paradigm. LLMs afford a high degree of linguistic realism and behavioural adaptability, enabling agents to approximate cognitive processes and engage in contextually appropriate pedagogical dialogues. This paper presents a thematic review of empirical and methodological studies utilising LLMs to simulate student behaviour across educational environments. We synthesise current evidence on the capacity of LLM-based agents to emulate learner archetypes, respond to instructional inputs, and interact within multi-agent classroom scenarios. Furthermore, we examine the implications of such systems for curriculum development, instructional evaluation, and teacher training. While LLMs surpass rule-based systems in natural language generation and situational flexibility, ongoing concerns persist regarding algorithmic bias, evaluation reliability, and alignment with educational objectives. The review identifies existing technological and methodological gaps and proposes future research directions for integrating generative AI into adaptive learning systems and instructional design.
Similar Papers
Integrating LLM in Agent-Based Social Simulation: Opportunities and Challenges
Artificial Intelligence
Lets computer characters act more like real people.
Large Language Models in Architecture Studio: A Framework for Learning Outcomes
Computers and Society
Helps architecture students learn and create better.
Social Simulations with Large Language Model Risk Utopian Illusion
Computation and Language
Computers show fake, too-nice people in chats.