Informational constraints, option generation, and anticipation

Objectives: To determine the option generation strategies used by expert and novice tennis players when anticipating an opponent's intentions using information that would normally be picked up sequentially prior to the opponent striking the ball. Design and method: Altogether, 12 expert and 14 novice tennis players completed an option generation task when presented with rallies from real matches under two display conditions, namely video film and animation. Rallies were presented in such a way that participants either had access to contextual information and postural cues (video) or solely contextual information (animation; e.g., player positioning, shot sequencing). Results: The experts were more accurate than novices in both display conditions. Participants generated less options in the video compared with the animated condition. More often than not, participants chose the first option they generated. The expert participants generated more task-relevant and fewer task-irrelevant options than novices, with this effect being stronger in the animated than the video condition. The number of options generated was negatively related to performance in the video condition only. Conclusions: In dynamic, time-constrained tasks, performers adapt their option generation strategy depending on the information available. In keeping with Long Term Working Memory theory (Ericsson & Kintsch, 1995), when constrained to anticipate based on contextual information alone, effective anticipation is underpinned by being able to access the likely outcome and potential relevant alternatives. Moreover, when pertinent postural cues become available, option generation strategies consistent with the Take The First heuristic model may be optimal (Johnson & Raab, 2003). Implications for performance and training are discussed.