Rapid reconfigurations of brain activity support efficient neuronal communication and flexible behaviour. Suboptimal brain dynamics impair this adaptability, possibly leading to functional deficiencies. We hypothesize that impaired flexibility in brain activity can lead to motor and cognitive symptoms of Parkinson's disease (PD). To test this hypothesis, we studied the 'functional repertoire' the number of distinct configurations of neural activityusing source-reconstructed magnetoencephalography in PD patients and controls. We found stereotyped brain dynamics and reduced flexibility in PD. The intensity of this reduction was proportional to symptoms severity, and explained by beta-band hyper-synchronization. Moreover, the basal ganglia were prominently involved in the abnormal patterns of brain activity. Our findings support the hypotheses that: symptoms in PD reflect impaired brain flexibility, this impairment preferentially involves the basal ganglia, and beta-band hypersynchronization is associated with reduced brain flexibility. These findings highlight the importance of extensive functional repertoires for behaviour and motor.Parkinson's disease (PD) is a severe and disabling neurodegenerative disorder. It is characterized by reduced amplitude of movements, and slowing of cognitive processes, imposing major individual and social burden (24). The main histopathological finding in PD is a severe nigrostriatal dopamine depletion (25,26). PD has traditionally been regarded predominantly as a motor disease. However, recent clinical and neuroimaging findings have questioned these views (27). For example, the first symptoms to appear are often not motor, and the clinical phenotype clearly goes well beyond the motor impairment, with other domains, such as executive functioning, involved (28,29). Structural Magnetic Resonance Imaging (MRI) has also confirmed that PD is more widespread than previously thought (27). Although functional magnetic resonance (fMRI) studies have identified impairment in the corticostriatal network in PD patients, the observed changes in functional connectivity extend into many other brain systems (30). Similarly, magnetoencephalography (MEG) studies have also reported widespread functional connectivity alterations in PD (31-34) .The brain can be represented as a network, with, at the macro-scale, brain areas as 'nodes' and structural or functional relationships between the brain areas as 'edges ' (35, 36), following which the topology of the network can be characterised using graph theory. A rapidly growing body of research now uses graph theory to characterize large-scale patterns of brain activation in health (37) and disease (38). Graph theory has highlighted the multifaceted nature of large-scale interactions in PD, with studies reporting more-connected networks (39-42), less-connected networks (43-48), and a combination of both (49,50). However, most research in this area has assumed brain activity to be stationary, and did not explore the structure of dynamic fluctuations in activity...