Representative task designs for the study of perception and action in sport

Empirical evidence indicates that an underlying antecedent of sports expertise is the ability of skilled athletes to successfully use predictive information to guide their anticipatory responses. This article discusses the nature of the relationship between expertise, perception and action using ideas from ecological psychology (Gibson, 1979) and representative task design (Brunswik, 1956). This conceptual framework suggests that a shortcoming of many research studies is a failure to accurately sample the performance environments of which skilled athletes have experience. It is proposed that the task constraints used to study perception and action should closely represent the specific performance contexts towards which investigators are attempting to generalise. Comparison of the research literature using in situ and video simulation experimental paradigms suggests that athletes' performance may vary under different task constraints. These empirical findings need to be considered in future research on the study and training of perceptual skill.