The ability to modify power output (PO) in response to a changing stimulus during exercise is crucial for optimizing performance involving an integration system involving a performance template and feedback from peripheral receptors. The rapidity with which PO is modified has not been established, but would be of interest relative to understanding how PO is regulated. The objective is to determine the rapidity of changes in PO in response to a hypoxic challenge, and if change in PO is linked to changes in arterial O(2) saturation (S (a)O(2)). Well-trained cyclists performed randomly ordered 5-km time trials. Subjects began the trials breathing room air and switched to hypoxic (HYPOXIC, F(I)O(2) = 0.15) or room (CONTROL, F(I)O(2) = 0.21) air at 2 km, then to room air at 4 km. The time delay to begin decreasing S (a)O(2) and PO and to recover S (a)O(2) and PO on to room air was compared, along with the half time (t (1/2)) during the HYPOXIC trial. Mean S (a)O(2) and PO between 2 and 4 km were significantly different between CONTROL and HYPOXIC (94 +/- 2 vs. 83 +/- 2% and 285 +/- 16 vs. 245 +/- 19 W, respectively). There was no difference between the time delay for S (a)O(2) (31.5 +/- 12.8 s) and in PO (25.8 +/- 14.4 s) or the recovery of S (a)O(2) (29.0 +/- 7.7 s) and PO (21.5 +/- 12.4 s). The half time for decreases in S (a)O(2) (56.6 +/- 14.4 s) and in PO (62.7 +/- 20.8 s) was not significantly different. Modifications of PO due to the abrupt administration of hypoxic air are related to the development of arterial hypoxemia, and begin within approximately 30 s.