Under normal conditions, motor evoked potentials (MEPs) elicited by transcranial magnetic stimulation increase in amplitude if the subject exercises the examined muscle immediately before recording. The authors examined the effect of nonfatiguing exercise on the amplitude of MEPs on 42 psychiatric, medicated inpatients (14 with depression, 14 with schizophrenia, and 14 with mania) compared with 14 healthy control subjects. For each subject, a total of 50 baseline and 50 postexercise MEPs were recorded. The mean (+/- standard deviation) postexercise MEP facilitation, expressed as a percentage of mean baseline values, was significantly lower (p </= 0.02) in all three patient groups (148 +/- 91% in the depressed, 107 +/- 43% in the schizophrenic, and 143 +/- 63% in the manic patients) compared with the control subjects (268 +/- 223%). There was no clear evidence that psychotropic medication could be fully responsible for the modified MEP response to exercise, but the degree to which they were responsible is impossible to assess. These findings indicate that the reduced postexercise MEP facilitation is not related to any particular psychiatric illness, but rather that it represents a suppressed reaction of cortical excitability shared by patients with either depression, mania, or schizophrenia. The authors propose that, despite the lack of specificity, the study of postexercise MEP alterations could be useful in the neurophysiologic evaluation of motor cortex excitability in some psychiatric illnesses, except in those patients who may have diminished somatosensory input associated with voluntary muscle contraction resulting in decreased cortical excitability.