Behavioral Biases and Group Decision
Human behavioral biases are a topic of great importance within the field of economics, but they also are of significance for a broad range of human endeavors. Within economics we most often encounter these biases in puzzling patterns of individual behavior, as in insurance purchases, affinity for gambling, or stock portfolio choices. But, behavioral biases are also thought to play an important role in many critical aggregate phenomena, from market swings to partisan political deadlock. This thesis presents a model for how several of these biases may have arisen together during evolution and suggests that individual biases are intricately linked to the dynamics of group decision making: 1) confirmation bias acts as a stabilizing influence when individuals provide input to group decisions, holding their input steady in the face of rival motives or distracting information; 2) group decisions tend to accentuate belief polarization; and 3) probability weighting, when seen through this evolutionary lens, simply undoes the polarization that occurred when the shared belief was acquired. The final chapter presents an experiemental approach to overcoming behavioral biases in the context of deliberative democracy. Small group dialog often triggers biases that lead to belief polarization, but when properly structured, participants coalesce around causal knowledge of the problem domain, rather than entrenched conceptions of the solution.