Essays on Judgment and Decision Making

Abstract

Most judgments and choices in life arrive not as disconnected and isolated problems but as the next in a series enmeshed in a social context. The works contained herein, in one way or another, examine the role of context in shaping human judgment and decision making. The first essay reveals the surprising effects of the context of one’s own prior judgments of confidence. In seven experiments (N = 5,484), we explore the counterintuitive finding that confidence in one’s own judgments decreases over a series of difficult quantitative estimates. Our findings suggest that rather than evaluating confidence in isolation, participants evaluate confidence in reference to their stated confidence on earlier judgments. We theorize that confidence in earlier judgments increases in hindsight due to biased forgetting of disconfirming evidence, resulting in the downward trend we observed. The second essay examines sunk-cost bias in a social context. Viewed through the lens of economic rationality, sunk-cost bias is clearly irrational. In four experiments and a replication (N=2,754 U.S. adults), however, we found that decision makers received rewards precisely when they factored in sunk costs. Across three domains, decision makers who chose to escalate commitment after a prior investment were rated as more competent, warm, and confident than those who did not, a pattern that reversed in the absence of a prior investment. The pattern persisted with real financial stakes, an effect mediated by the perceived competence of the decision maker. The third essay builds upon the second by examining escalation of commitment to a failing course of action as a signal of trustworthiness. In two experiments (N = 2,198), we found that decision makers who escalate commitment to a failing course of action are trusted 29% more by third-party observers than decision makers who de-escalate. Decision makers who escalate commitment actually are 15% more trustworthy. This signal was surprisingly robust to incentives for strategic signaling. Taken together, these essays contradict a foundational assumption of the rational actor model that history, whether your own recent judgments or the decision process, is irrelevant.