A Behavioral Demonstration of Overconfidence in Judgment

Abstract

Overprecision-an excessive confidence that one knows the truth-is both the most durable and the least understood form of overconfidence. This article outlines an approach to the study of overprecision that avoids some of the methodological problems of other approaches and better reflects the way uncertainty affects choices in everyday life. We measured the precision in judgment implied by people's tendency to adjust their point estimates of an uncertain quantity in response to the costs of overestimating or underestimating the correct answer. The results revealed robust overprecision. People adjusted their estimates less than they should have given their actual knowledge, and this effect was driven by their subjective confidence. In press, Psychological Science Abstract Overprecision-an excessive faith that one knows the truth-is both the most durable and least understood form of overconfidence. This paper offers an approach to the study of overprecision that has more in common with the way uncertainty affects our actions in everyday life and avoids some methodological problems of other approaches. We measure the precision in judgment implied by people's tendency to adjust their point estimates of an uncertain quantity in response to the costs of over-or underestimating the correct answer. The results reveal robust overprecision. People adjust their estimates less than they should given their actual knowledge, and this is driven by their subjective confidence.Author note: Data and stimuli are available at

Toward a behavioral theory of real options: Noisy signals, bias, and learning

“…In the context of learning, overprecision bias occurs when . This leads the firm to insufficiently adjust its beliefs in response to new information (Mannes & Moore, ). The association between precision bias and the rate of belief updating may be defined as α = α*(1 − ζ), where α is the rate of belief updating used in Equation , α* is the Bayesian (optimal) rate of belief updating, and ζ is a precision parameter.…”

I know I'm right! A behavioural view of overconfidence

“…Arguably, however, this procedure is unnatural for most people, and it does require an understanding of probability. We sought to address these criticisms in our recent research with a judgement task that represents more faithfully how overprecision affects people in their daily lives 7 .…”

Mapping Spatial Uncertainty in Prediction

“…Predicting the "typical" path (e.g., the most likely point of landfall in a hurricane), and predicting the uncertainty itself (e.g., the growth of the "cone of uncertainty"). This distinction between the two elements of prediction is quite analogous to the distinction or understanding of the mean, based on several samples, a task at which people are quite good (Peterson & Beach, 1967) versus assessing the variance of those samples, a task at which people are not so proficient, and at which we exhibit systematic biases (Mannes & Moore, 2013). In particular, in other contexts it has been observed that people tend to underestimate the variance of multiple samples, as if consistently underestimating the contribution of random factors to variables in the world (Kahneman, 2011;Henrion & Fischhoff, 2003;Tversky & Kahneman, 1971).…”