Functional mapping of the human brain with positron emission tomography (PET) can best be performed by obtaining multiple short measurements of cerebral blood flow in a single sitting. In this manner regional changes in blood flow accompanying the increased neuronal activity from a movement, sensation, or even cognition task, have been identified. However, localizing a functional region with PET has been severely limited by the poor resolving properties of PET devices. Using a new method of data analysis we recently reported the mapping of visual field stimuli on human visual cortex with surprisingly high reliability as measured by the low standard deviation in positions across different subjects (as low as 1 mm). In this work the analysis technique enabling such high-resolution functional brain mapping is fully described. Additionally, simulations are presented to illustrate its advantages and limitations.