We tested peripheral, spinal and cortical excitability during 3 minutes of unresisted finger tapping at the maximal possible rate, which induced fatigue. Subsequently, we studied the temporal dynamics of muscle fatigue, expressed in the tapping movement profile, and its relationship to neural systems using mixed model analyses. The tapping rate decreased by 40% over the duration of the task. The change in the amplitude of the range of motion was not significant. The excitability of the flexor and extensor muscles of the index finger was tested via evoked potentials obtained with various types of stimulation at various levels of the motor system. the change in spinal excitability with time was evaluated considering the simultaneous changes in muscle excitability; we also considered how spinal excitability changed over time to evaluate cortical excitability. Excitability in the flexor and extensor muscles at the different levels tested changed significantly, but similar excitability levels were observed at notably different tapping rates. Our results showed that only 33% of the decrease in the tapping rate was explained by changes in the excitability of the structures tested in the present work.Determining the central mechanisms involved in muscle fatigue is important from a physiological perspective and can also have relevant implications from an applied perspective when we refer to sports, ergonomics or certain pathological conditions. These mechanisms have been studied thoroughly in the case of isometric muscle contractions; they include changes in excitability both at the spinal cord and M1 networks 1-8 .Another type of muscle activity corresponding to the contractions performed during rhythmic repetitive movements (RRMs) is essential in daily living and may result in fatigue. Traditionally, their central expressions of fatigue have been studied at the point at which an activity has been completed 9,10 , which is a limitation because the CNS recovers very quickly when the activity ends 11,12 . Several works recently tested fatigue at the central level immediately following the end of unresisted RRMs without allowing time for CNS recovery 5,13,14 . The reduction in the maximal movement rate was greater after 30 s of finger tapping (ft) than that after 10 s, and the reduction was accompanied by an increase in the excitability of M1 GABA b interneurons, which was more pronounced after 30 s of performing the task 5,13 . Interestingly, spinal excitability, which was tested by measuring cervicomedullary evoked potentials (CMEPs), increased during the waning of the tapping rate 14 , which is essentially different from the outcome when fatigue is caused by isometric exercise performed for the same duration and executed with the same body segment 14 .However, the description of how corticomuscular excitability changes with RRM fatigue development has not been performed, as previously done for isometric activities [15][16][17] . Another un-resolved point is whether the