Seven subjects made a wrist flexion movement as rapidly as possible in response to a cutaneous shock on the opposite hand. In some trials, an auditory warning signal was given 0.5 s beforehand. In random trials, transcranial magnetic stimulation (TMS) was used to elicit EMG responses (MEPs) in forearm flexor and extensor muscles 0-500 ms before the cutaneous shock. H-reflexes were elicited in flexor muscles at the same intervals. The warning stimulus reduced reaction time from about 400 ms to 200 ms. MEPs in the flexor muscles were significantly suppressed from 125 ms after the warning stimulus until the time of the cutaneous shock whilst MEPs in the extensors, and H-reflexes in the flexor were either unaffected, or reduced by a smaller amount at a later time. Responses in relaxed contralateral muscles were unchanged. If the task was changed to a choice reaction, in which the imperative stimulus (but not the warning signal) indicated whether to flex or extend the wrist, then there was no change in the MEPs or H-reflex in the warning period. A similar effect was seen if the duration of the warning period was extended from 0.5 to 2 s in a simple reaction (flexion) task. We conclude that increased excitability of the corticospinal output is not required to speed up reaction times. The time taken to discharge cortical output elements is relatively unimportant compared with the time needed to process the sensory input and link it to the motor output.