Repeated-sprint exercise provides an interesting model to investigate the mechanisms governing the decline in power output during whole-body, high-intensity dynamic activities which require high contraction rates similar to those encountered in many athletic activities 
. The majority of the energy required for all-out sprinting is derived from phosphocreatine (PCr) hydrolysis and anaerobic glycolysis, and repeated sprints therefore result in large changes in both PCr and hydrogen ion (H+
) concentration 
. There is also an increased aerobic contribution when sprints are repeated but, nonetheless, even the latter sprints in a repeated-sprint bout remain predominantly anaerobic 
. Accordingly, most previous explanations of fatigue during repeated-sprint exercise have focused on factors associated with cellular muscle fatigue 
. These include limitations in anaerobic energy supply from adenosine triphosphate (ATP) and PCr 
and intramuscular accumulation of selected metabolic by-products including inorganic phosphate (Pi
) and H+
Fatigue induced by repeated-sprint efforts has also been related to neural adjustments, as demonstrated by a reduction in the central nervous system’s drive to the active musculature 
, an impaired muscle activation 
, a reduced EMG amplitude 
, and alterations in the neural strategies for voluntary activation of the contracting muscles 
. The results from such experiments indicate that, given the high level of activation to sustain maximal power output, a suboptimal motor unit activity could impair the ability to repeatedly generate maximum power outputs 
. Therefore, while a large component of muscle fatigue during repeated-sprint exercise is likely to be due to local intramuscular factors, neural factors might also contribute to the decline of maximal power output.
While previous studies have measured changes in metabolites 
and neural factors 
during RSE, only a limited understanding of the causes of fatigue during repeated-sprint exercise can be obtained by monitoring the many changes which occur concurrently. A limitation of this approach is that it is difficult to discern which of the measured factors, if any, is really responsible for the observed fatigue. One way to study the importance of, and the interplay between, the changes in both muscle metabolism and neuromuscular function on performance during repeated-sprint exercise is to measure the recovery of performance and to relate this to the recovery of both muscle metabolites and neuromuscular function. For example, despite large changes in both PCr and H+
concentration during a 30-s all-out test, it has been reported that the recovery of single 10-s sprint performance is not related to changes in muscle pH, but is strongly related to the resynthesis of PCr 
. While a similar result might be hypothesised for repeated-sprint exercise, we have previously reported that the recovery of single- and repeated-sprint performance follow different time courses 
, suggesting that their recovery might be mediated by different mechanisms.
Consequently, the goal of this study was to investigate for the first time the recovery of repeated-sprint performance, following an exhausting repeated-sprint bout, and to compare this with the recovery of muscle ATP, PCr and H+
, and neuromuscular activity (via surface electromyography (EMG) signals - a reasonable proxy for net motor unit activity e.g., 
). We hypothesized that the degree to which these quantities were correlated with mechanical power during repeated-sprint exercise would allow judgements to be made regarding the involvement of putative energy-store, fatigue-metabolite and neuromuscular activity changes in the recovery and therefore composition of mechanical performance during repeated-sprint exercise.