Do Positive Illusions Predict Overconfidence in Judgment? A Test Using Interval Production and Probability Evaluation Measures of Miscalibration

International audienceWe address the question as to whether judgmental overconfidence, as assessed by probability miscalibration, is related to positive illusions about the self. We first demonstrate that judgmental overconfidence measured with interval production procedures can be considered a trait, due to correlations observed in miscalibration scores in two sets of general-knowledge questions of varying difficulty administered at different times. In addition, the hard-easy effect operated in different ways on over-precision and self-placement of one's performance relative to others: The more difficult the calibration task, the greater the over precision but the greater the under placement of one's performance. Finally, there was no evidence that miscalibration was related to dispositional optimism and self-efficacy. A second study extended these results by including further measures of disposition to experience positive illusions such as unrealistic optimism, a general tendency to consider oneself ``better-than-average,'' and two indexes of dispositional perception of control. The positive illusion measures showed considerable inter-correlations, but did not correlate with miscalibration on the interval production task, and correlated negatively with optimism concerning societal risks. A final study replicated this pattern of findings, but showed that disposition to positive illusions did predict miscalibration on the same questions measured with a probability evaluation technique. Our research demonstrates that ``overconfidence'' is not a unitary construct, but a series of overlapping ones. Copyright (C)2010 John Wiley & Sons, Ltd