Two experiments were carried out to examine the effect of moderate and maximal exercise on the cognitive performance of experienced soccer players. Experiment 1 examined the speed and visual search in familiar (game) and unfamiliar (non-game) contexts. Participants had to detect, as quickly as possible, the presence or absence of a ball in tachistoscopically presented slides. Participants were tested at rest and while exercising at 70 and 100% maximum power output. A main effect of exercise intensity was demonstrated and Tukey post-hoc tests showed that performance during maximal exercise was significantly better than in the other two conditions. We concluded that exercise significantly improves speed of visual search. Experiment 2 examined the effects of exercise on speed of search, speed of decision following ball detection, overall speed of decision and accuracy of decision at rest and while exercising at 70 and 100% maximum power output. A repeated-measures multivariate analysis of variance and Tukey post-hoc tests showed that performance during exercise was significantly better than at rest. Observation of the separate univariate analyses of variance demonstrated that most of the variance could be accounted for by overall speed of decision and speed of decision after ball detection. We concluded that exercise induces not only an improvement in a simple task, like speed of visual search, but also an overall increase in speed of information processing. Theories concerning the effect of emotionally induced arousal on cognitive performance do not accurately predict the effect of physically induced arousal on cognitive tasks.