Anticipation of Beach Volleyball Attacks: A Comparison of Natural Versus Animated Virtual-Reality Scenes

Defending beach-volleyball attacks requires good anticipation skills due to the highly time-constrained situations (Vansteenkiste et al., 2014). For examining the underlying mechanisms of these skills, using virtual reality is a promising approach because specific variations of players movements can be manipulated (e.g., the run-up of the attacker) to evaluate important cues for anticipating attacks. As a first step, animated virtual-reality scenes were created from natural beach-volleyball attacking sequences by manually animating players’ movements in Autodesk 3dsMax. Four conditions of attacking scenes were created: animated occluded, animated not occluced, natural occluced and natural not occluded. In occluded trials, videos were occluded 5 video frames (167ms) before ball-hand contact of the attacker. Participants were instructed to decide as accurate as possible (accuracy in %) in occluded trials and to respond as fast as possible (response time in ms) in non-occluded trials. 82 sport science students were tested in 120 trials with the same 30 mother-trials in each condition (counter-balanced order). Videos were back-projected on a large screen and participants responded using a 3-button-response system (WiiMote). After the study, participants evaluated the animated scenes with a questionnaire. Response accuracy was analysed with paired t-tests. Performance was lower for animated scenes compared with natural scenes (55.3% vs. 64.4%, t(81)=7.06, p < 0.01, d = 0.71) and participants responded later in animated compared with natural scenes (-170 ms vs. -237 ms before ball-hand contact, t(81)=7.46, p < 0.01, d = 0.32). Lower accuracies in animated scenes imply that visual cues were less helpful for anticipation. Questionnaire data revealed that the dynamics of a shot negatively affected anticipation in animated scenes and that rather late cues were used for decisions, which could explain the later responses. Future virtual-reality applications should especially focus on the animation of relevant visual cues and the dynamics of the attack.