Yury Koush1,
2, Maria Joao Rosa3, Fabien Robineau4, 5,
Klaartje Heinen6, Nikolaus Weiskopf7, Patrik
Vuilleumier4, 5, Dimitri Van de Ville, 28,
Frank Scharnowski1, 2
1Department
of Radiology and Medical Informatics, CIBM, University of Geneva , Geneva,
Switzerland; 2Institute of Bioengineering, cole Polytechnique Fdrale
de Lausanne, Lausanne, Switzerland; 3Computer Science Department,
University College London, London, United Kingdom; 4Department of
Neuroscience, CMU, University of Geneva, Geneva, Switzerland; 5Geneva
Neuroscience Center, Geneva, Switzerland; 6Institute of Cognitive
Neuroscience, University College London, London, United Kingdom; 7Wellcome
Trust Centre for Neuroimaging, Institute of Neurology,, University College
London, London, United Kingdom; 8Department of Radiology and
Medical Informatics, CIBM, University of Geneva, Geneva, Switzerland
Neurofeedback based on real-time fMRI is a novel technique that allows to train voluntary control over brain activity. So far, this technique was limited to training localized brain activity within a region of interest. Here, we overcome this limitation by presenting real-time dynamic causal modelling in order to provide neurofeedback information based on connectivity between brain areas rather than activity within a brain area. Being able to train network activity is an important extension of the neurofeedback approach that will contribute to its development as a promising research tool, and will open up a whole new range of medical applications.