Federal Project Grant Award Summary

Yale University received a $339,513 CAREER award from the National Science Foundation's Division of Information and Intelligent Systems under the Computer and Information Science and Engineering program (CFDA 47.070), effective October 1, 2025 through October 31, 2027. The project will develop theoretical foundations and machine learning algorithms designed to model and predict human decision-making as it occurs in practice, rather than as traditional behavioral theories prescribe. The research will employ graph-theoretic approaches to understand how pairwise interactions between alternatives influence choice behavior, building on recent literature modeling decision irrationality through interference effects within choice sets.

The deliverables include theoretical analysis of interference-based modeling approaches for human choices, expanded applications to complex choice problems including ranking scenarios, and development of defensive tools to protect consumers from manipulative platform designs. Additionally, the project will produce measurement frameworks for quantifying irrationality and testing protocols to identify platform designs that promote more rational decision-making. By operationalizing behavioral economic principles within a machine learning framework, the research aims to advance both the theoretical understanding of human behavior and the practical design of large-scale web systems that better serve user interests.