A multiple-cue learning approach as the basis for understanding and improving soccer referees' decision making.

A significant proportion of all referee decisions during a soccer match are about fouls and misconduct. We argue that most of these decisions can be considered as a perceptual-categorization task in which the referee has to categorize a set of features into two discrete classes (foul/no-foul). Due to the dynamic nature of tackling situations in football, these features share a probabilistic rather that a deterministic relationship with the decision criteria. Accordingly, these processes can be studied on the basis of a multiple-cue learning framework as proposed by Brunswick (1955), which focuses among others on how people learn from repeated exposure to probabilistic information. Such learning processes have been studied on a wide range of tasks, but until now not (to our knowledge) in the area of judging sport performance. We suggest that decision accuracy of referees can be improved by creating a learning environment that fits the requirements of this theoretical perspective.