Talk by Prof. Simon Eickhoff (CSN Virtual Seminar)

We hereby announce the next talk in the 'CSN Virtual Seminar':

Looking back to move forward: Meta-Analytic priors for machine-learning on fMRI

by Prof. Simon Eickhoff, INM-7, Jülich Research Centre and Heinrich Heine University Düsseldorf

Abstract

Whereas the potential inference from any single neuroimaging study is limited to method-inherent drawbacks, the high degree of standardization in neuroimaging research allows pooling and integration of activation results from several thousands of experiments. Moreover, large-scale databases of neuroimaging results have emerged over the last years that compile this wealth of information. In this talk, I would like to outline, how emerging meta-analytic tools may be used to capitalize on these resources and provide new insights into several aspects pertaining to the organization of the human brain: i) The localization of brain functions and its relationship to task-specific confounds ii) The functional roles underlying, e.g., morphometric, findings, including formal inference for functional decoding iii) The identification of functional connectivity in a task-based state through the mapping of co-activations. iv) How these may complement information on connectivity in the unconstrained, endogenously driven task-free "resting" state v) The delineation of cortical modules by data-driven clustering of co-activation patterns, which, in combination with the other described methods described in this talk, entails the possibility for functional brain mapping and atlasing. Finally, I will illustrate how this knowledge on human brain organization can be leveraged for inference on socio-affective or cognitive traits in previously unseen individual subjects or psychopathology in mental disorders. Providing a bidirectional translation, such application will in turn provide information on the respective brain regions and networks.