Reproduce “PsychoGAT: A Novel Psychological Measurement Paradigm through Interactive Fiction Games with LLM Agents” on OpenResearch
Turn your favorite coding agents into research agents that review literature, develop hypotheses, and run experiments.
Universal app for Apple silicon and Intel. Requires macOS 11 or later.
About this paper
Qisen Yang, Zekun Wang, Honghui Chen, Shenzhi Wang, Yifan Pu, Xin Gao, Wenhao Huang, Shiji Song and 1 more
PsychoGAT introduces a multi-agent Large Language Model framework that gamifies psychological assessment by transforming traditional self-report scales into interactive fiction games. This system demonstrates high psychometric reliability and validity across various psychological constructs, while significantly enhancing user engagement and immersion compared to existing assessment methods.
Key insights
- Large Language Models (LLMs) can effectively serve dual roles as both 'adept psychologists' to understand psychological constructs and 'innovative game designers' to create engaging narratives.
- A multi-agent LLM architecture (Game Designer, Game Controller, Critic) facilitates a collaborative, iterative process that enhances content quality, coherence, and user experience in dynamic interactive fiction games.
- Integrating psychological assessment within an engaging game narrative, driven by LLMs, can significantly overcome participant monotony and disengagement common in traditional self-report scales.
Summary by alphaXiv.
Tip: change arxiv.org to autoarxiv.org in any arXiv abstract link to open the paper here.