Gait disturbances are disabling and common in Parkinson's disease (PD). Patients use many different, and typically self-developed, strategies to compensate for their walking abnormalities. There is a wide variety of compensatory strategies, of which external and internal cueing are the most commonly known. External cueing refers to externally produced predictive stimuli such as a metronome, whereas internal cueing refers to a movement reference generated internally, such as counting while walking. The efficacy of external and internal cueing varies dramatically across patients, and some patients can even be identified as 'non-responders' to a particular cue. Consequently, a one-size-fits-all approach simply does not suffice, which increases the need for better understanding of the key mechanisms behind these compensation strategies. Furthermore, it is still unknown how the efficacy of compensation strategies changes longitudinally. The goal of the UNITE-PD project is to address these questions. The investigators aim to investigate whether the efficacy of internal and external cueing changes over time, and whether it is differentially affected in responders and non-responders. In order to work towards a more personalized treatment for patients with PD, the investigators also aim to identify potential patient characteristics that can mediate the actual use of compensation strategies in daily life. The multicenter UNITE-PD project is divided in a joint package and individual site-specific packages. All partners will investigate the neural working mechanisms of compensation strategies in PD from different angles in the site-specific packages. The joint package focusses on the long-term effect of the compensation strategies and the potential patient characteristics that can influence the efficacy of the compensation strategies. In this project, the investigators will define responders and non-responders to external and/or internal cueing. With the use of extensive clinimetrics, the aim is to identify patient characteristics that might influence the efficacy. With the use of a custom made cueing app (which will be applied during a follow up period of six months), the long-term effect of cueing in the responders can be investigated. Together, all centers aim to include a total of 384 participants (Netherlands N = 104, Belgium N =90, Israel N = 75, Italy N = 115). Importantly, this sample size is not based on the joint workpackage described here, but on the numbers needed for the individual site-specific work packages.