ABSTRACT
Functional neuroimaging is a powerful technology for mapping human brain function, and has undergone enormous development during the past decade (Fox, 1997; Seitz et al., 2000). Most widely used are measurements of stimulation-related haemodynamic changes. These changes can be assessed with measurements of the regional cerebral blood flow (rCBF) using positron emission tomography (PET), and of the blood oxygenation level-dependent
approximately 5 to 9 mm (Frackowiak et al., 1994; Calamante et al., 1999). The temporal resolution of PET and fMRI, however, is relatively poor, being in the range of approximately 6 s to 1 min, due, respectively, to the tracer kinetics and the haemodynamic characteristics of the measurements. Nevertheless, the reconstructed tomographic imaging data allow one to detect activity changes occurring simultaneously in different parts of the brain, including the different parts of the cerebral cortex, subcortical structures as the basal ganglia and thalamus, and the cerebellum. It should be born in mind, however, that the observed haemodynamic changes represent only indirect measures of brain activity. Although, under physiological conditions, there is a tight coupling of activation-related metabolic and haemodynamic changes to increases in neural activity (Fox et al., 1984; Blomqvist et al., 1994; Bandettini et al., 1997; Hoge et al., 1999), bioelectric neural activity has a time-course in the range of several milliseconds, faster than the haemodynamic measures by three orders of magnitude. Therefore, one of the assumptions underlying functional imaging with PET and fMRI is that a state of activation has to be kept constant over a sufficiently long period of time in order to capture functional changes in the different parts of the brain, during a condition approaching a steady state.
