In the cognitive sciences, we currently witness a „pragmatic turn” away from the traditional representation-centered framework towards a paradigm that focusses on understanding the intimate relation between cognition and action. Such an „action-oriented“ paradigm has earliest and most explicitly been developed in robotics, and has only recently begun to gain impact on cognitive psychology and neurobiology. The basic notion is that cognition should not be understood as a capacity of deriving world-models, which then might provide a “database” for thinking, planning and problem-solving. Rather, it is emphasized that cognitive processes are not only closely intertwined with action but that cognition can actually best be understood as „enactive“, as a form of practise itself. Cognition, on this account, is grounded in a pre-rational understanding of the world that is based on sensorimotor acquisition of real-life situations. control strategy. One key concept I will discuss is that of sensorimotor contingencies, i.e., law-like relations between actions and associated changes in sensory input. I will propose to advance this concept further and suggest that actions not only play a key role for perception, but also in developing more complex cognitive capabilities.