We treat the brain as multistate, metastable dynamical system, continuously transitioning between quasistable attractors that support perception, cognition, and action. Neurological injury and disease disrupt the dynamics, leading to the loss of functional states and the emergence of maladaptive patterns. Neurorehabilitation, from this perspective, can be understood not just as repairing damaged circuits but as reshaping brain dynamics—nudging neural activity toward favorable attractor states and restoring the system’s capacity for flexible metastability.
In this talk, we will outline how concepts from dynamical systems theory, control, and network neuroscience provide a framework for understanding the principles of recovery, and will show how external interventions (stimulation, rhythmic entrainment, sensory feedback, and training) can act as targeted perturbations that reconfigure trajectories of brain activity.  We will show examples from both experimental studies and computational models will illustrate how recovery can be seen as a process of re-establishing the fluid repertoire of brain states rather than returning to a single “normal” configuration. By adopting this dynamic viewpoint, neurorehabilitation becomes a problem of guiding a complex system back to health—through rhythm, resonance, and controlled instability—opening new avenues for personalized and adaptive therapeutic strategies.