First Impressions Matter: A Model of Confirmatory Bias Matthew Rabin and Joel Schrag, University of California, Berkeley and Emory University, 1997. This paper presents a model of confirmatory bias, a cognitive bias where individuals misinterpret new information to support previously held hypotheses. The authors show that confirmatory bias leads to overconfidence, as people tend to believe more strongly in their favored hypotheses than is justified. The model demonstrates that even with extensive information, an agent may come to believe in a false hypothesis with near certainty if confirmatory bias is strong enough. The paper discusses how confirmatory bias affects belief formation in situations of uncertainty. It shows that an agent who suffers from confirmatory bias may come to believe in a hypothesis that is probably wrong, meaning that a Bayesian observer who is aware of the agent's confirmatory bias would favor a different hypothesis. The authors also show that even an infinite amount of information does not necessarily overcome the effects of confirmatory bias: over time, an agent may with positive probability come to believe with near certainty in the wrong hypothesis. The paper reviews psychological evidence that humans are prone to confirmatory bias, and presents a formal model of the phenomenon. The model shows that confirmatory bias leads to overconfidence, and that an agent who suffers from confirmatory bias may come to believe in a hypothesis that is probably wrong. The authors also show that even with extensive information, an agent may come to believe in a false hypothesis with near certainty if confirmatory bias is strong enough. The paper discusses the implications of confirmatory bias in economic situations, and highlights some likely problems with such applications. The authors conclude that confirmatory bias is a significant factor in economic decision-making, and that further research is needed to understand its implications.First Impressions Matter: A Model of Confirmatory Bias Matthew Rabin and Joel Schrag, University of California, Berkeley and Emory University, 1997. This paper presents a model of confirmatory bias, a cognitive bias where individuals misinterpret new information to support previously held hypotheses. The authors show that confirmatory bias leads to overconfidence, as people tend to believe more strongly in their favored hypotheses than is justified. The model demonstrates that even with extensive information, an agent may come to believe in a false hypothesis with near certainty if confirmatory bias is strong enough. The paper discusses how confirmatory bias affects belief formation in situations of uncertainty. It shows that an agent who suffers from confirmatory bias may come to believe in a hypothesis that is probably wrong, meaning that a Bayesian observer who is aware of the agent's confirmatory bias would favor a different hypothesis. The authors also show that even an infinite amount of information does not necessarily overcome the effects of confirmatory bias: over time, an agent may with positive probability come to believe with near certainty in the wrong hypothesis. The paper reviews psychological evidence that humans are prone to confirmatory bias, and presents a formal model of the phenomenon. The model shows that confirmatory bias leads to overconfidence, and that an agent who suffers from confirmatory bias may come to believe in a hypothesis that is probably wrong. The authors also show that even with extensive information, an agent may come to believe in a false hypothesis with near certainty if confirmatory bias is strong enough. The paper discusses the implications of confirmatory bias in economic situations, and highlights some likely problems with such applications. The authors conclude that confirmatory bias is a significant factor in economic decision-making, and that further research is needed to understand its implications.