Epilepsy is a frequent neurological disease which affects up to 1% of the population, in particular younger people including babies and children. With the current project, we like to study how different neuronal assemblies " talk " to each other, by analyzing the electrical activity of the human brain and by visualizing the pathways and regions, recruited by the fast neuronal activity. Three different techniques will be applied for this reason. Patients with epilepsy, who present aberrant electrical activity from time to time, and healthy subjects will be studied with multichannel electroencephalogram (EEG; up to 256 channels) and diffusion tensor imaging (DTI). EEG measures brain activity in the millisecond range, whereas DTI, a relatively new imaging technique, provides non-invasively data of nerve fiber bundles, i.e. their localization and thickness, in an individual human brain. Functional MRI, the third applied technique, shows local blood flow changes which reflect also neuronal activity non-invasively, albeit with less temporal precision but more spatial precision. In order to verify and improve our EEG techniques, we will also study epileptic mice, born with a brain malformation (similarly found in many human patients), during rest and epileptic seizures. All data will be carefully analyzed with modern mathematical algorithms to better understand which region is at which moment activated, through purely non-invasive tools. The overall goal of the project is the development and application of combined imaging techniques to human subjects and to improve our understanding of the brain "in action". While they are applied to patients with epilepsy, the results of our studies will be also of relevance for other chronic brain diseases, like autism or dementia, i.e. allowing the improvement of the diagnosis and/or monitoring of treatment effects.