Why do people make the choices they do? Decision research has traditionally focused on observing behavior. But choices alone often reveal little about the reasoning behind them.

A new paper co-authored by our colleague Kamil Fuławka, published in Proceedings of the National Academy of Sciences (PNAS), presents a new way to study these hidden decision processes. The team combined behavioral experiments with participants’ own explanations of their decisions and analyzed both using large language models. This made it possible to systematically identify the reasons behind individual choices and compare them with established theories of decision making.

The results show that people’s explanations contain valuable information that cannot be recovered from behavior alone. Rather than relying on fixed decision strategies, people adapt their reasoning to the structure of the decision they face.

Beyond decision science, the study demonstrates how artificial intelligence and high performance computing can help researchers combine different types of scientific data. The framework opens new opportunities to study complex human behavior by bringing together behavioral data, language, and computational models.

👉 Read the paper