In the last decades there has been a progressive advance in the development of techniques able to explore in humans neurophysiologic and neurochemical processes. Positron emission tomography (PET) is a very powerful technique allowing to study a quite variable range of physiological and biochemical processes in the healthy subjects and in diseases. Apart from its capacity to provide pathophysiological information, PET is also important for the objective assessment of therapeutic efficacy. Initial studies were performed measuring cerebral metabolic rate for glucose (CMRglc) and cerebral blood flow (CBF), representing an indirect index of synaptic activity. The advent of receptor tracers allowed measuring other important physiological parameters, such as receptor occupancy, and endogenous release. In neuropsychiatric disorders, as Alzheimer disease, schizophrenia, epilepsy and Huntington disease, PET has been useful to elaborate hypothesis of the pathogenesis, to relate symptoms to biological variables and to study individuals at increased risk. The new concepts of neurovascular unit and default network, preferentially active at rest, can significantly change the approach of PET, with images reflecting a complex scenario, not merely limited to neural activity, but involving the activity of the entire neurovascular unit and the multifunctional role of astrocytes. To detect dysfunction of the dialog between glutamatergic neurons and astrocytes could lead to a better understanding of altered functional brain images. In this direction a professional network between PET researchers and basic scientists, could give a determinant improvement in the capability to understand the complex physiological and pathophysiological cerebral world. J. Cell. Physiol. 226: 948–961, 2011. © 2010 Wiley-Liss, Inc.