Adaptive information search and judgment strategies in solitary and competitive tasks

A substantial body of judgment and decision-making research focuses on decisions made under risk, where all relevant option outcome and probability information is known a priori. However, most real-world decision tasks are made under uncertainty, where such population- level information is unknown. Against this background, how can, do, and should organisms obtain and use information in order to improve their judgments and decisions under uncertainty? This dissertation addresses these questions in two distinct domains: external information search in competitive tasks (papers 1 and 2) and internal search in the context of the inner-crowd (papers 3 and 4). In paper 1, we develop a new paradigm called the Competitive Sampling Game (CSG) to study how organisms adjust search in the presence of both natural uncertainty (i.e., gamble parameters) and social uncertainty (i.e., behavior of others). The paradigm produces simulation and empirical results showing that organisms should and do dramatically reduce search in the presence of competition to almost minimal levels. In paper 2, we expand on the initial results of the CSG to show how different levels of competition drive the evolution of decision strategies. In a second domain, we address how people can improve their judgments by harnessing a diverse inner-crowd using dialectical bootstrapping. In paper 3, we apply dialectical bootstrapping to a Bayesian reasoning paradigm to show how dialectical instructions induce strategy change and how people can become more Bayesian by averaging biased non-Bayesian judgments in their inner-crowd. In paper 4, we apply the inner-crowd to a cue-based estimation task and model the effects of different estimation strategies on final estimates and confidence. Our results suggest that people can use their confidence judgments to outperform the simple average of their inner-crowd. Moreover, dialectical bootstrapping increases these effects.