Is 'error proneness' specific to errors? Predictive validity of the cognitive failures questionnaire

摘要

Over two decades ago, the cognitive failures questionnaire (CFQ) was introduced as a tool for obtaining information concerning the frequency of everyday errors. Since then errors have been related to the construct of error proneness, which is supposed to be a steady trait in any individual. The CFQ has been adapted for populations other than English speakers and used in a great variety of experimental and applied settings. The aim of this paper is twofold: 1) to further address the factorial structure of the CFQ; 2) to verify the predictive validity of the error proneness construct in a simulated driving task. Results of the first study showed that seven factors might underlie the CFQ structure: action monitoring and control, memory for names, decision, mind-wandering, over-reliance on memory, blunders, absent-mindedness. However, we show that the CFQ total score may be preferred. Furthermore, the experimental study here reported showed that the CFQ total score is a weak predictor of driving performance and only predicts the time spent out-of-path, suggesting that 'error proneness', as measured by this questionnaire, is mainly linked to constructs such as absent mindedness or distractibility rather than a dispositional tendency to make errors.