Our environments are filled with statistical regularities that often go unnoticed. Research to date has extensively investigated how humans are able to extract and represent such statistical regularities. This ability, referred to as statistical learning (SL), underlies many of our fundamental cognitive abilities. However, research so far focused mainly on the learning of a single set of stable regularities, leaving an important aspect of SL largely overlooked: its flexibility. How do people flexibly update their knowledge about statistical regularities? What happens to existing knowledge in the face of changes in regularities? What neural substrates subserve this adaptation to novel structures? Can people become better at adapting to new regularities? All of these outstanding questions highlight a clear gap in the SL literature, which the current project aims to address. The current project will validate an adjusted SL paradigm that will allow us to measure a person’s adaptability. A subsequent functional magnetic resonance imaging study based on this paradigm will provide us with new insights into the neural dynamics of flexible SL. Finally, a transcranial magnetic stimulation study will allow us to investigate the malleability of adaptability in SL. Taken together, the envisioned project will further advance our understanding of human learning.