ABSTRACT
Errors can be defined only in relation to correct and desired behavior. The choice of a taxonomy can influence the ease or possibility of prediction. A very specific taxonomy implies a high level of detail. Errors result from the normal operation of the human information-processing system, along with effects arising from the environment, the various pressures and biases influencing the actor, and the latter’s mental, emotional, and attentional states. Considerable effort has been expended in the past few years in estimating the probability of human error for a wide range of actions. Most of the published data on error rates have wide uncertainty bounds placed around them. These are often ignored by users who don’t follow statistics, who simply need numbers for decision making. The concept of “error proneness” has had a rather checkered history. Error proneness may simply be a statistical anomaly. Many suggestions have been made about the relation of personality to error.
