This proposal aims to elucidate the neuronal mechanism serving the human brain’s predictive capacities. Despite the wide and growing consensus on the importance of predictions in behavior and in brain functioning, the underlying neuronal mechanisms are largely unknown. Recent empirical results suggests that predictions are embodied in the temporal structure of both stimulus-evoked and ongoing activity, and thus that neural oscillations may provide the neurophysiological basis for the brain’s representation, coordination, and enactment of sensory predictions. We examine the proposal that different brain rhythms are used for representing and enacting “when” and “what” predictions, and interrogate the mechanisms implementing predictions at different spatial scales (i.e., local and global).The research objectives are addressed using a unique combination of invasive, intracranial EEG recordings with laminar resolution in humans to investigate local mechanisms, and whole-brain magnetoencephalographic recordings and cutting-edge functional connectivity analysis to identify brain areas as sources and/or targets of predictions, thus relating local to global mechanisms. This brings an unparalleled depth of investigation into the mechanisms supporting predictions across the frequency and spatial spectra. Our proposal has the potential to unify many apparently disparate cognitive phenomena, e.g., learning, speech perception, and active sensing, but also provides insight into psychiatric diseases like schizophrenia, where alterations in predictions and oscillatory activity are implicated. The high quality and expertise of the hosts in addition to their theoretical and methodological complementariness secure success of the project and the prospect for long-standing collaborations, increasing EU scientific excellence and competitiveness.