We investigate cognitive mechanisms and neural systems underlying volition, cognitive control, and goal-directed action. We are interested in the dynamic regulation and adaptation of cognitive control processes to changing task demands, the representation, shielding, and switching of intentions, and the neurocognitive mechanisms mediating self-control and failures of self-control. Our research rests on an integrative theoretical framework according to which “volition” can be functionally decomposed into cognitive control mechanisms, which enable humans to anticipate future outcomes, adapt behavior to changing goals, and override impulsive or habitual responses. A key assumption is that the evolution of advanced control capacities not only expanded the flexibility of human action, but also gave rise to “control dilemmas”, which confront humans with the meta-control problem how to balance antagonistic control modes (goal shielding vs. shifting; cognitive stability vs. flexibility; goal-directed vs. habitual control). This control dilemma framework forms the basis for the research program of the CRC 940. To investigate the dynamic regulation of cognitive control, we combine behavioral tasks, functional neuroimaging (fMRI), brain stimulation with TMS and tDCS, and ecological momentary assessments of real-life behavior. Thereby we recently showed that dysfunctional interactions between brain networks involved in performance-monitoring, cognitive control, episodic prospection, and value-based decision-making reliably predict the propensity to commit real-life self-control failures.