The freezing burden will be quantified in subjects with Parkinson's Disease (PD)before and after 6 weeks of training. Two types of interventions (20 subjects in each group) will be tested: 1) Open-loop group (OLG); 2) Closed-loop group (CLG). Each session of the OLG training includes walking courses aimed at provoking freezing episodes. The experimenter will trigger an auditory rhythmic stimulation (RAS) in walking conditions likely to invoke freezing (e.g., turning) and the subject will learn to synchronize his/her gait with the auditory cues, i.e., to keep the walking pace and coordination and, as a result, to avoid freezing. Similar principles will apply for the CLG training; however, the RAS will be elicited automatically by a device that recognizes an approaching freezing episode.

The freezing burden will be quantified in subjects with PD before and after 6 weeks of training. Two types of interventions (20 subjects in each group) will be tested: 1) Open-loop group (OLG); 2) Closed-loop group (CLG). Each session of the OLG training includes walking courses aimed at provoking freezing episodes. The experimenter will trigger an auditory rhythmic stimulation (RAS) in walking conditions likely to invoke freezing (e.g., turning) and the subject will learn to synchronize his/her gait with the auditory cues, i.e., to keep the walking pace and coordination and, as a result, to avoid freezing. Similar principles will apply for the CLG training; however, the RAS will be elicited automatically by a device that recognizes an approaching freezing episode.

We anticipate that after intensive training, the central nervous system (CNS) of subjects with PD will be able to anticipate impending freezing episodes based on awareness of the environmental conditions (e.g., an approaching turn) and/or based on sub-conscious response to a deteriorating gait pattern. As a result, an automated motor response that paces and coordinates gait will be internally triggered by the CNS and the approaching freezing episode will be averted. The overall freezing burden will therefore decrease in trained subjects.

Inclusion Criteria:

1. Have a diagnosis of PD.
2. Suffer from the freezing of gait (FOG) symptoms. Subjects must score 2 or more on item #3 of the subjective FOG questionnaire (FOG-Q) and exhibit two or more FOG episodes during a short, functional FOG evaluation procedure that includes FOG-provoking conditions (e.g., turns, doorways) and 5 laps of walking in a figure 8 shaped trajectory.
3. Able to walk unassisted for at least 5 minutes with ample rest.

Exclusion Criteria:

1. Having serious co-morbidities or acute illness that would make training inappropriate.
2. Have had brain surgery.