New preprint: Large language models and human decision-making

Dr. Kamil Fulawka, Postdoc in the Computational Social Science group of Dr. Philipp Lorenz-Spreen at SynoSys, is first author of a new preprint investigating how large language models (LLMs) can be used to study the psychology of risky choice. Together with his co-authors, he introduces a scalable approach to analyze participants’ free-text explanations of their decisions.

The study shows that LLMs can accurately identify the reasons people give for their lottery choices in more than 92% of cases. These reports reveal systematic patterns: people’s reasoning depends strongly on the structure of the decision problem rather than being fixed across individuals. Models based on these reason profiles also achieve higher predictive accuracy than prospect theory –a widely used model in decision research.

By demonstrating the scientific value of verbal data, the work opens new avenues for developing context-aware models of human decision-making.

Read the preprint here: https://osf.io/preprints/psyarxiv/yuzmw_v1