When an agent chooses between prospects, noise in information processing generates an effect akin to the winner's curse. Statistically unbiased perception systematically overvalues the chosen action because it fails to account for the possibility that noise is responsible for making the preferred action appear to be optimal. The optimal perception pattern exhibits a key feature of prospect theory, namely, overweighting of small probability events (and corresponding underweighting
of high probability events). This bias arises to correct for the winner's curse effect. (JEL D11, D81, D82, D83)There is considerable evidence that human perception of reality is noisy and biased.1 While randomness can be understood as a technological limitation of human cognition, systematic behavioral biases, such as those documented in the psychological experiments of Kahneman and Tversky (1979), are more puzzling. Since there is no obvious reason why natural or cultural evolution could not remove these biases, their prevalence suggests that they serve a purpose. This paper argues that perception biases arise as a second-best solution when some noise in information processing is unavoidable. In particular, we show that overweighting of small probability events optimally mitigates errors due to randomness. Our model also provides a framework for conceptualizing errors in decision making, allowing us to consider, for example, whether overweighting of small probabilities is a mistake or an optimal heuristic. Finally, our results demonstrate 1 McFadden (1999, p. 96) summarizes the experimental evidence as follows: "Humans fail to retrieve and process information consistently… These failures may be fundamental, the result of the way human memory is wired. I conclude that perception-rationality fails, and that the failures are systematic, persistent, pervasive, and large in magnitude."