Over recent years, activation studies that have been undertaken using brain imaging techniques, such as functional magnetic resonance imaging, positron emission tomography or near infrared spectroscopy, have greatly improved our knowledge of the functional anatomy of the brain. Nevertheless, activation studies do not directly quantify the variations of synaptic transmission (neuronal activity) but detect it indirectly either through the visualisation of changes in cerebral blood flow, oxidative or glycolytic metabolism (for positron emission tomography), or through the measurement of a global index that is dependent on both cerebral blood flow and oxidative metabolism (for functional magnetic resonance imaging and near infrared spectroscopy). Such approaches are based on the concept of a tight parallelism--termed coupling--between variations in neuronal activity, metabolism and cerebral blood flow. However, several "uncoupled" situations between these parameters have been reported over the last decade through experimental, pharmacological and pathophysiological studies. The aim of this review is to focus on these data that have to be taken into account for the interpretation of the results obtained in activation paradigms.