Recent evidence suggested that Action Observation Therapy (AOT), based on observation of actions followed by immediate reproduction, could be a useful rehabilitative strategy for promoting functional recovery of children affected by Unilateral Cerebral Palsy (UCP).

The AOT is based on the discovery of mirror neurons, a class of visuomotor neurons that are activated when the individual performs a targeted action, both when observing the same action performed by another individual.

An important assumption of AOT is that the mirror system, thanks to its visuomotor properties, is able to coordinate visual information with the observer's motor experience. Indeed, the actions outside the individual's behavioral repertoire are elaborated and categorized only on the basis of visual characteristics, without inducing any phenomenon of motor resonance in the observer's brain.

Therefore, given the deficiency of their motor repertoire, children affected by UCP could have a reduced activation of the mirror system during the observation of actions performed by healthy subjects. Nevertheless, this activation could increase during the observation of the same actions performed by a subject with similar motor strategies, due to a similar form of hemiplegia.

The present project is a randomized controlled clinical trial to verify the influence of the observed model on the effectiveness of AOT in the rehabilitation of the affected upper limb in children with UCP. In particular, the study will verify whether the rehabilitation through AOT based on a pathological model (improving the patient's current abilities) gives more results than AOT based on a typical (healthy) development model, as reported by all the documented researches in literature.

Furthermore, to investigate the functional reorganization of the sensorimotor system after rehabilitative treatment, a subgroup of participants will be subjected to a fMRI session (Functional Magnetic Resonance), to verify functional changes, comparing data before and after AOT.

The study will compare two rehabilitation models through AOT to assess feasibility and effectiveness. In the preliminary phase of the study a single rehabilitation treatment protocol will be planned, based on the observation and execution of unimanual and bimanual actions integrated in activities that are significant for the subject's age group. The recorded actions will be performed by a child with typical development and by two hemiplegic children:

1. upper limb disability with grades between 4-5 according to House Functional Classification (HFC) system (corresponding to synergic hand of Ferrari and Cioni's Kinematic Hand Classification).
2. upper limb disability with grades between 6-7 according to HFC system (corresponding to subfunctional hand of Ferrari and Cioni's Kinematic Hand Classification).

The participants (N. 26) randomly assigned to the experimental group will be subjected to the rehabilitation protocol through AOT based on the observation of the "Pathological Model". The model, while showing a manipulation strategy similar to that of the study participant, will show an improvement with respect to the patient's current abilities. The children randomly assigned to the control group will be subjected to the rehabilitation treatment protocol through AOT based on the "Typical developmental model".

Both experimental and control treatment will consist of a minimum of 10 hours (min. 10, max 15) for a period of 15 consecutive working days (3 weeks). Each video sequence will show a specific action repeated several times for 3 minutes. Subsequently, participants will be asked to perform the action observed with their hemiplegic upper limb or with both upper limbs for a maximum of 3 repetitions.

A subgroup of study participants (N. 8+8) from both groups will undergo two fMRI sessions, before and immediately after the AOT.

The fMRI session consists of two experimental tasks:

1. action observation (functional session 1-2), in which patients will observe short videos in which actions are performed with the right hand from a subjective perspective.
2. motor task (session 3), in which actions consist in reaching-grasping an object (a cube, a sphere or a cylinder) and putting it into boxes, placed on the right and on the left of the object.

The study consists of two experimental conditions:

1. Observation of the paretic hand, in which the actions will be performed by another patient with UCP.
2. Observation of the healthy hand, in which the action will be reproduced by a healthy model.

The aim of the study is to investigate the functional reorganization of sensorimotor system comparing functional activation before and after rehabilitation with AOT.

Inclusion Criteria:

• confirmed diagnosis of UCP according to definition (MRI and clinical history)
• age 6-16 years at the time of recruitment;
• mild or moderate disability of the upper limb: score between 4 and 7 according to the House Functional Classification (HFC) system;
• sufficient cooperation and understanding to participate in the activities, according to the doctor/physiotherapist who is treating the child and to the neuropsychological evaluation.

Exclusion Criteria:

• cognitive, visual or auditory impairments;
• history of seizures or seizures not well controlled by therapy;
• upper limb surgery within 8 months prior to enrollment;
• upper limb botulinum toxin injection (BoNT-A) within 6 months prior to enrollment.

Additional exclusion criteria for participants in fMRI sessions:

• insufficient cooperation to participation in neuroimaging studies lasting about 30 minutes;
• presence of contraindications to instrumental investigations using 3 Tesla Magnetic Resonance system (implants, metal prostheses, shunts etc.). The presence of these contraindications will be verified by the use of a preliminary questionnaire administered to the parents.
• < 9 years at the time of recruitment.

Derived data supporting the findings of this study will be available from the corresponding author upon reasonable request.

Single