Skill Level and Graphical Detail Shape Perceptual Judgments in Tennis

Abstract

Reducing the representation of human actions from normal video to biological motion animation in perceptual tasks means removing a number of visual features from the scenery, thereby eliminating potentially useful information for successfully performing the task. To determine the impact of selected visual features on perceptual judgments in tennis, we invited skilled players and novices to predict baseline shot direction under four different display conditions (PL: point-light display; PL_TC: PL plus an animated tennis court; NV_NB: normal video without ball; NV: normal video). Skilled players clearly outperformed novices and prediction performance increased with more realistic display content. Both groups were similarly affected by display conditions and across groups significant differences between conditions were only found for PL vs NV, and PL_TC vs NV, respectively. Application of signal detection theory on response data revealed that, unlike novices, skilled players showed a bias towards preferentially expecting cross-court shots and this bias increased with enhancement in graphical detail. Results confirm previous research in that biological motion appears to provide the minimal essential information necessary for correctly predicting an opponent's intent, particularly in skilled players. In addition, findings indicate that a combination of player and scenery-related visual information is likely to facilitate visual anticipation; however, such information seems to impact skilled players' and novices' response behaviour differently.