Proceedings of the Annual Meeting of the Cognitive Science Society

Overprecision (overconfidence in interval estimation) is a biaswith clear implications for economic outcomes in industriesreliant on forecasting possible ranges for future prices andunknown states of nature - such as mineral and petroleumexploration. Prior research has shown the ranges peopleprovide are too narrow given the knowledge they have – thatis, they underestimate uncertainty and are overconfident intheir knowledge. The underlying causes of this bias are,however, still unclear and individual differences research hasshed little light on traits predictive of susceptibility. Takingthis as a starting point, this paper directly contrasts the NaïveSampling Model and Informativeness-Accuracy Tradeoffaccounts of overprecision – seeing which better predictsperformance in an interval estimation task. This was achievedby identifying traits associated with these theories – ShortTerm Memory and Need for Cognitive Closure, respectively.Analyses indicate that NFCC but not STM predicts intervalwidth and thus, potentially, impacts overprecision.