Seeing the forest for the trees

With growing reliance on the Internet as a primary source of input into nearly every type of decision, individuals must find ways to cope with the overwhelming quantity and variety of information accessible to them. Unfortunately, while technology increases the availability of information by facilitating both sharing and retrieval, it does not yet offer clear assistance for integrating this information into coherent preferences. Furthermore, the weighting of disparate information content is subject to the inherent decision-making biases that people have been shown to exhibit in many other contexts, and reliance on simplifying heuristics may even be exacerbated in online environments where distillation of meaning from abundant or conflicting information is especially difficult. The first paper looks at the effect of interface design on decisions, whereby individuals focus their attention according to the organization of information. Simply manipulating the way that option attributes are partitioned into categories, we induce decision-makers to ascribe different relative importance to them. The second paper examines the interpretation of opinion information, and an observed asymmetric preference for high variance experiences in positive domains and low variance experiences in negative domains. We argue that salient memories of prior experiences set reference points at extremes rather than "null" outcomes, and in turn, decision-makers perceive disproportionate likelihood and impact of realizing highly favorable outcomes in positive domains and highly unfavorable outcomes in negative domains. Finally, the third paper explores unrepresentativeness of opinion information that is made public, as a result of sample bias in the choice to provide ratings and subsequent response bias in the ratings provided. Specifically, we demonstrate that individuals are more inclined to share extreme opinions, and the opinions they do share are influenced by exposure to the opinions of others. In each of the experimental studies described, we highlight implications for improving both the provision and use of online information. While many of these findings are generalizable to offline decisions, they are especially relevant to the technology-supported decisions that have become increasingly prevalent over the Internet.