UC San Diego

Choice architecture, or the design of the context in which people make judgments and decisions, can influence people’s behaviors in systematic ways. In this dissertation, I focus on two aspects of choice architecture – framing (which description to use) and defaults (which option to preselect). Because the options under consideration remain the same regardless of which frame is used or which option is set as the default, inconsistent responses are commonly viewed as irrational. This perspective of choice architecture, however, overlooks the subtle information that choice architects may convey to decision makers through their choice of frame and default. Building on this information leakage framework, I aim to provide a more comprehensive understanding of how this implicit social interaction affects judgment and decision making. Chapter 1 provides a causal link between frames, inferences, and judgments, and demonstrates that inferences are sufficient to generate attribute framing effects. Chapter 2 examines whether attribute framing is malleable to the informativeness of the choice of frame. While framing effects remained unaltered when the uninformative frame selection process was described using hypothetical vignettes, participants ceased to respond systematically to frames when the incentive structure was manipulated in a two-player repeated game. Finally, Chapter 3 shows that when decision makers can self-select into different default settings, they strategically exploited the asymmetric signal that their choices convey. Overall, the current dissertation provides further support for choice architecture as an implicit social interaction, and offers new insights on factors that moderate the effectiveness of behavioral interventions.