The coherent firing of large neural populations produces electrical and magnetic fields that peak at specific times and frequencies and can be recorded at the scalp of humans as the EEG/MEG. Specific rhythms have been shown to be of great research and clinical importance. However, the exact anatomical localization of these epochs of coherent neuronal activity is necessary for further progress. While there is no doubt that invasive recording of this activity is very informative, it is not generally applicable in humans. On the other hand, the limitations in spatial resolution of the EEG/MEG has long hampered mapping its well known temporal and frequency organization to the equally well known, fine-grain spatial organization of brain functions.

During the past decade our group has developed a functional EEG/MEG tomography of the neural current densities, based on the use of constraints from individual anatomy obtained from the MRI. The resulting Neuroimaging modality (VARETA) has a spatial resolution in the order of several mm. and an unlimited temporal resolution. This method has been used to quantify the development of resting state EEG rhythms sources from age 5 to 97, mapped into the MNI probabilistic Atlas. This study shows the emergence and attenuation with age of rhythmic activity in the alpha band in sensorimotor areas.

Deviations from this maturational pattern are assessed with a modification of Statistical Parametric Mapping known as quantitative EEG Tomography (qEEGT. Examples of such deviations are narrow band increases of activity in the delta band in cortical areas near brain lesions. Excess theta has been found to co-localize with edematous brain tissue. Also, these slow oscillations appear to be among the earliest signs of ischemia. A recent validation of qEEGT was the successful discrimination between asymptomatic carriers and non carriers of the E280A Presenilin-1 mutation that produces an early-onset form of Alzheimer's disease.

Further improvements in EEG/MEG tomography are possible by use of information provided by the fMRI, in particular when concurrent EEG/fMRI recordings are available. We have recently used the EEG/BOLD neurovascular relation to identify the extended neural circuits involved in the generation of the alpha rhythm during the resting state. While the actual generators of scalp alpha oscillations seem to be concentrated in the visual cortex, the modulation of this activity seems to be mediated by reciprocal inhibition of this structure with the thalamus and insula through a complex pattern of effective connectivity that is specific to a certain mode of oscillation. These studies are now being extended to other brain states.

EEG/MEG tomography, through the fusion of electrophysiological information with other imaging modalities is steadily decreasing the temporal and spatial limits for visualizing dynamic changes in brain activation and connectivity.