Bounded Rationality in Decision-Making Under Uncertainty

Abstract

To better introduce the behavioral economics approach and reinforce the theoretical basis for supporting bias-aware user modeling and evaluation, we need to have a deeper understanding of the concepts, theories, recent progress, and empirical findings on users and their biased decisions in varying scenarios. To achieve this, this chapter takes a step back from specific computational IR models and focuses on explaining the fundamental frameworks (e.g., theories of two systems), research progress, and practical implications of behavioral economics research on boundedly rational decision-making activities. Our review focuses on the major human biases and heuristics that are both widely examined in behavioral economics studies and also clearly contradict one or more assumptions that are explicitly or implicitly made in formal IR models. Although the theories on bounded rationality may not be able to match the precision and quantifiability of formal computational models, as argued by Kahneman, this statement of limitation from the classic economics side is “just another way of saying that rational models are psychologically unrealistic” [Kahneman (American Economic Review 93(5):1449, 2003)]. This argument also serves as part of the motivations for this book and the author’s broad research agenda on IR research.