Representing reality: Investigating the perception-action couplings of expert soccer Goalkeepers under representative constraints

Representative Design has become a frequently used term across sport science (Pinder et al. 2011). Brunswiks original assertion that there “is little technical basis for telling whether a given experiment is an ecological normal, located in the midst of a crowd of natural instances, or whether it is more like a bearded lady at the fringes of reality” (Brunswik, 1955, p.204) led towards a key concept, representation design. Brunswik (1955, p.198) proposed that researchers “must resist the temptation {. . .} to interfere” with the environment and “instead strive to retain its natural causal texture in the stimuli presented to participants” (Dhami et al, 2004 p.962). In football goalkeeping, early work on expertise used a penalty kick task in in-situ and digital environments (See Savelsbergh et al. 2002). Whilst the findings are interesting, it is worth proceeding with caution and attention should be turned towards creating representative experiments that are reflective of the environment the individual inhabits. This study attempted to show differences in perceptual behaviours between a traditional task, and a representative task. In a professional football case study, 4 goalkeepers took part in both conditions over a season. Using the Quiet Eye(QE) (Vickers, 1996) as the perceptual mechanism for analysis, eye behaviours were recorded via an SMI-ETG mobile device, and coded with a Vision-In-Action system. QE duration was significantly longer (t2=2.66,P≤0.05) in Penalty trial (PT) than in the Representative trial (RT) (50.75%±2.84% v 45.57%±0.93%). QE onset occurred significantly (t2=4.75,P