Recognition of Micro-Relations in the Context of Full Game Patterns in Soccer

Pattern recognition contributes to anticipation in team sports. Patterns of play consist of localized relations between individuals. Research in face recognition has shown that recognition performance for whole faces is reduced when local parts of a face (rather than the whole) are presented initially. From a sports perspective, this suggests that the use of small-sided games (SSG; part of a pattern), may disrupt the ability to recognize patterns of play in a full-sided game (the whole pattern). This study investigated the effects of skill level and amount of information presented in initial viewing on the recognition of sequences of play in soccer. Participants were presented with sequences from soccer matches in point light display format in either ‘whole’ or ‘part’ display mode. ‘Whole’ (W) clips showed a 11v11 soccer match and ‘part’ (P) clips presented sequences showing only two players: either two Centre Forwards (CF) or Peripheral Players (PP). Elite (n = 20), skilled (n = 34), and novice (n = 37) soccer players viewed 18 clips in a viewing phase and, following a break, completed a recognition phase in which they made judgments as to whether each clip was novel or had been presented in the viewing phase. This process was repeated in three counterbalanced conditions: ‘W-P’ (i.e., whole in viewing phase and part in recognition phase), ‘P-W’ and ‘W-W’. There was a main effect of group. Elite players’ (71.67  10.06%) recognition accuracy was higher than skilled (56.52  12.70%) and novice (53.22  13.20%) groups (p’s < 0.01). There was a main effect of condition. Participants were more accurate in the W-W (67.09  11.45%) than W-P (57.42  8.36%), and P-W conditions (56.90  13.12; p’s < 0.01). There was also a main effect of featured players. Recognition accuracy was higher for part clips featuring CFs (59.28  13.74%) than those featuring PPs (55.63  14.53%; p = 0.02). Findings suggest that CFs are important to recognizing patterns of play and 11v11 games during training may be more effective than SSG in supporting pattern recognition.