To determine whether the intensity of dyspnea at a given level of respiratory motor output depends on the nature of the stimulus to ventilation, we compared the sensation of difficulty in breathing during progressive hypercapnia (HC) induced by rebreathing, during incremental exercise (E) on a cycle ergometer, and during isocapnic voluntary hyperventilation (IVH) in 16 normal subjects. The sensation of difficulty in breathing was rated at 30-s intervals by use of a visual analog scale. There were no differences in the level of ventilation or the base-line intensity of dyspnea before any of the interventions. The intensity of dyspnea grew linearly with increases in ventilation during HC [r = 0.98 +/- 0.02 (SD)], E (0.95 +/- 0.03), and IVH (0.95 +/- 0.06). The change in intensity of dyspnea produced by a given change in ventilation was significantly greater during HC [0.27 +/- 0.04 (SE)] than during E (0.12 +/- 0.02, P less than 0.01) and during HC (0.30 +/- 0.04) than during IVH (0.16 +/- 0.03, P less than 0.01). The difference in intensity of dyspnea between HC and E or HC and IVH increased as the difference in end-tidal PCO2 widened, even though the time course of the increase in ventilation was similar. No significant differences were measured in the intensity of dyspnea that occurred with changes in ventilation between E and IVH. These results indicate that under nearisocapnic conditions the sensation of dyspnea produced by a given level of ventilation seems not to depend on the method used to produce that level of ventilation.(ABSTRACT TRUNCATED AT 250 WORDS)