To explore if and how the common rebreathing (paper bag) approach to hyperventilation works, two experiments were carried out. In the first experiment, 12 normal volunteers, aware of the common physiological rebreathing rationale, were twice asked to overbreath intensely and then either to rebreath or to restart normal ventilation. Alveolar CO₂ increased more quickly and physical symptoms disappeared earlier in the rebreathing condition. The second experiment had a similar design; however, this time the subjects were led to believe that, after both hyperventilation provocation tests, they were rebreathing in a semiclosed tube system. On one of the occasions, the tube system was, in fact, open. The CO₂ restoration again differed in the two conditions. In this second experiment, the moment of symptom disappearance was not significantly earlier in the rebreathing condition. The combined results of the studies suggest that expectation and suggestion play a role in the effects of rebreathing approaches on hyperventilation.