The target of this clinical trial is Freezing of Gait symptom (FoG), and associated falls. The project is aimed to evaluate the effects of an innovative experimental protocol to treat Parkinson (PD) Patients with FoG. This new physiotherapy protocol is based on the recovery of the correct mental representation of the movement, through Action Observation. A method that can facilitate the process of re-learning correct motor strategies, and at the same time avoid the phenomena of external cues dependency. Our version of the Action observation method uses video-clips of gait exercises. In these video-clips the audio part was obtained with the sonification of the kinematics of the body movements. We postulate that action observation, through the activation of the mirror system, is able to reactivate stored motor programs concerning walking ability, which can be used to facilitate recovery of defective motor control and overcome freezing of gait. Moreover, given that patients with PD and FoG may have major shortages of attention resources, a multisensory approach (audio-visual) would help to further reduce the attention load, facilitating learning processes. Therefore half of participants received an experimental protocol with Action Observation plus signification while the other half received a traditional protocol combining visual and auditory cues.

FoG is a disabling and distressing symptom strictly associated to falls. The little or no pharmacological responsiveness of FoG has led to an increasing interest in rehabilitation interventions aimed at functional recovery and autonomy. Currently, standard protocols employed for rehabilitation are based on the use of external (visual and auditory) sensory cues. However, cued strategies generate an important dependence on the environment. Teaching motor strategies without cues (i.e. action observation - AO) can be an alternative/innovative approach to rehabilitation, that matters most on appropriate allocation of attention and lightening cognitive load. One way to increase the effectiveness of AO, is the use of a multisensory learning mode (visual and auditory) to facilitate the recovery of motor gestures thanks to enhanced perceptual processes, which is known to be reduced in PD with FoG. Sonification could be an important method to enhance therapeutical effects in action observation rehabilitation process. Sonification of movements amplifies the activity of the human action observation system including subcortical structures of the motor loop.

Methods. We compared the effects of two different therapeutic protocols. The experimental protocol was based on action observation plus sonification; patients of the experimental group re-learned 8 motor gestures watching video-clips showing an actor performing the same gestures, and then tried to repeat the gesture. Each video-clip was composed by images and sounds of the gestures. The sounds of gestures were obtained with the sonification technique, by transforming kinematic data (velocity) recorded during the execution of gesture, into pitch variations (for an example see: bit.ly/sonif_example). The same 8 motor gestures were re-learned in the standard protocol, with a common sensory stimulation method (active comparator group). We evaluated all patients of the two groups with functional and clinical scales before, immediately after, at 1 month, and 3 months after each treatment.The duration of each protocol will be about 2 months, 15 sessions, 2 times a week.

Data Safety Monitoring Plan. The patients' assignment to the two groups was as follow: a list of 20 patients (10 patients for each group) was created and the order fully randomized. The list was filled with the patients following the order of arrival from the Neurology Clinic (Cattinara Hospital, Trieste). All the evaluations (neuropsychological, neurological, and physiotherapy) are blind respect to the patient's group assignment. The person in charge to make assignment patient-treatment will not be the PI.

Sample size calculation. The dependent variable (the primary outcome measure) is the score of the N-FOG Questionnaire. Sample size has been calculated considering a repeated measures mixed ANOVA of the primary outcome measure, with the software G*Power 3. The result gives a sample size of 20 patients: 10 for each group (experimental and active comparator group). The sizing of the sample is justified by similar protocols evaluations that with an identical sample size (n = 10 per group) found statistically significant results.

Inclusion Criteria:

• Parkinson's Disease Diagnosis (UK Brain Bank)
• H&R stage < or = to 3;
• score of 1 on the 1st question of the NFOG-Q;
• no comorbidities that would preclude physiotherapy treatment;
• Back Depression Scale score < or = to 16;
• stabilized pharmacological therapy

Exclusion Criteria:

• dementia: MMSE > or = to 24
• past neuropathies, ictus or myelopathies
• orthopaedic comorbidity that may impede walking
• presence of DBS
• severe psychiatric pathology