While Large Reasoning Models (LRMs) generate extensive chain-of-thought reasoning, we lack a principled framework for understanding how these thoughts are structured. In this paper, we introduce a novel approach by applying Schoenfeld's Episode Theory, a classic cognitive framework for human mathematical problem-solving, to analyze the reasoning traces of LRMs. We annotated thousands of sentences and paragraphs from model-generated solutions to math problems using seven cognitive labels (e.g., Plan, Implement, Verify). The result is the first publicly available benchmark for the fine-grained analysis of machine reasoning, including a large annotated corpus and detailed annotation guidebooks. Our preliminary analysis reveals distinct patterns in LRM reasoning, such as the transition dynamics between cognitive states. This framework provides a theoretically grounded methodology for interpreting LRM cognition and enables future work on more controllable and transparent reasoning systems.
@article{arxiv.2509.14662,
title = {Understanding the Thinking Process of Reasoning Models: A Perspective from Schoenfeld's Episode Theory},
author = {Ming Li and Nan Zhang and Chenrui Fan and Hong Jiao and Yanbin Fu and Sydney Peters and Qingshu Xu and Robert Lissitz and Tianyi Zhou},
journal= {arXiv preprint arXiv:2509.14662},
year = {2025}
}