The purpose of this investigation was to determine the relationship between the repetition rate of a simple sensory stimulus and regional cerebral blood flow (rCBF) in the human brain. Positron emission tomography (PET), using intravenously administered H2(15)O as the diffusible blood-flow tracer, was employed for all CBF measurements. The use of H2(15)O with PET allowed eight CBF measurements to be made in rapid sequence under multiple stimulation conditions without removing the subject from the tomograph, thus minimizing changes in base-line CBF and in head position due to longer intervals between scans. Nine normal volunteers each underwent a series of eight H2(15)O PET measurements of CBF. Initial and final scans were made during visual deprivation. The six intervening scans were made during visual activation with patterned-flash stimuli given in random order at 1.0-, 3.9-, 7.8-, 15.5-, 33.1-, and 61-Hz repetition rates. In each subject the region of greatest rCBF increase was determined. Within this region the rCBF was determined for every test condition and then expressed as the percentage change from the value of the initial unstimulated scan (rCBF% delta). Anatomical localization of the region of greatest rCBF response was performed employing bony landmarks from a lateral skull radiograph, a template of the cranium created from a transmission attenuation scan and a stereotaxic atlas. In every subject, striate cortex rCBF% delta varied systematically with stimulus rate. Between 0 and 7.8 Hz, rCBF% delta was a linear function of stimulus repetition rate. The rCBF response peaked at 7.8 Hz and then declined. The rCBF% delta during visual stimulation was significantly greater than that during visual deprivation for every stimulus rate except 1.0 Hz. The anatomical localization of the region of peak rCBF response was determined for every subject to be the mesial occipital lobes along the calcarine fissure, primary visual cortex. We conclude that stimulus rate is a significant determinant of rCBF response in the visual cortex. Investigators of brain responses to selective activation procedures should be aware of the potential effects of stimulus rate on rCBF and other measurements of cerebral metabolism. For cerebral responses to selective activation to be meaningfully interpreted, the stimulus repetition rate must be taken into consideration. Response amplitude may be maximized by proper rate selection or be undetectable due to selection of too high or too low a repetition rate. Stimulus rate must be controlled for when responses to unlike stimuli or performance tasks are compared or ambiguities will be present as to whether response differences are