The overprediction of fear: is it a form of regression toward the mean?

Fearful people tend to overpredict the amount of fear they will experience in subjectively threatening situations. Little is known about the determinants of overprediction bias, although it has been suggested that it is a form of regression toward the mean. In this article I argue that the regression effect is a description (phenomenon) rather than explanation, and so the regression toward the mean cannot "explain" the overprediction bias. Regression may be due to psychologically meaningful factors, and should not be dismissed as a statistical artifact. However, there are present several reasons why the overprediction of bias is unlikely to be a form of regression. This conclusion is supported by reanalyses of two recent studies.