Gait disorders represent a therapeutic challenge in Parkinson's disease (PD). This study investigated the efficacy of 4-week action observation training (AOT) on disease severity, freezing of gait and motor abilities in PD, and evaluated treatment-related brain functional changes. 25 PD patients with freezing of gait were randomized into two groups: AOT (action observation combined with practicing the observed actions) and "Landscape" (same physical training combined with landscape-videos observation). At baseline and 4-week, patients underwent clinical evaluation and fMRI. Clinical assessment was repeated at 8-week. At 4-week, both groups showed reduced freezing of gait severity, improved walking speed and quality of life. Moreover, AOT was associated with reduced motor disability and improved balance. AOT group showed a sustained positive effect on motor disability, walking speed, balance and quality of life at 8-week, with a trend toward a persisting reduced freezing of gait severity. At 4-week vs. baseline, AOT group showed increased recruitment of fronto-parietal areas during fMRI tasks, while the Landscape group showed a reduced fMRI activity of the left postcentral and inferior parietal gyri and right rolandic operculum and supramarginal gyrus. In AOT group, functional brain changes were associated with clinical improvements at 4-week and predicted clinical evolution at 8-week. AOT has a more lasting effect in improving motor function, gait and quality of life in PD patients relative to physical therapy alone. AOT-related performance gains are associated with an increased recruitment of motor regions and fronto-parietal mirror neuron and attentional control areas.

This study suggests that FoG in PD can be the result of a poor structural and functional integration between motor and extramotor (cognitive) neural systems.

A BS TRACT: Background: Action observation training and motor imagery may improve motor learning in Parkinson's disease (PD). Objectives: The objectives of this study were to assess mobility and balance (performing motor and dual tasks) and brain functional reorganization following 6 weeks of action observation training and motor imagery associated with dual-task gait/balance exercises in PD patients with postural instability and gait disorders relative to dual-task training alone. Methods: Twenty-five PD-postural instability and gait disorder patients were randomized into 2 groups: the DUAL-TASK+AOT-MI group performed a 6-week gait/balance training consisting of action observation training-motor imagery combined with practicing the observed-imagined exercises; the DUAL-TASK group performed the same exercises combined with watching landscape videos. Exercises were increasingly difficult to include the dual task. At baseline and at 6 weeks, patients underwent: mobility, gait, and balance evaluations (also repeated 2 months after training), cognitive assessment, and functional MRI, including motor and dual tasks. Results: Dual-task gait/balance training enhanced mobility, during both single-and dual-task conditions, and executive functions in PD-postural instability and gait disorders, with a long-lasting effect at 14 weeks. When exercises were preceded by action observation trainingmotor imagery, PD-postural instability and gait disorders showed greater improvement of balance and gait velocity both with and without the dual task, particularly during the turning phase. After training, the DUAL-TASK+AOT-MI group showed reduced recruitment of frontal areas and increased activity of cerebellum during functional-MRI motor and dual task, correlating with balance/turning velocity and executive improvements, respectively. The DUAL-TASK group showed reduced activity of supplementary motor area and increased recruitment of temporo-parietal areas during the dual task and decreased cerebellar activity during the motor task correlating with faster turning velocity. Functional MRI results were not corrected for multiple comparisons and should be interpreted carefully. Conclusions: Adding action observation training-motor imagery to dual-task gait/balance training promotes specific functional reorganization of brain areas involved in motor control and executive-attentive abilities and more long-lasting effects on dual-task mobility and balance in PD-postural instability and gait disorders.

Following the discovery of Mirror Neuron System (MNS), Action Observation Training (AOT) has become an emerging rehabilitation tool to improve motor functions both in neurologic and orthopedic pathologies.The aim of this study is to present the state of the art on the use of AOT in experimental studies to improve motor function recovery in any disease.The research was performed in PubMed, PEDro, Embase, CINAHL and Cochrane Central Register of Controlled Trials (last search July 2015). Randomized controlled trials (RCTs) that analyse efficacy of AOT for recovery of motor functions, regardless of the kind of disease, were retrieved. The validity of the included studies was assessed using the Cochrane Collaboration tool for evaluating risk of bias.Twenty RCTs were eligible. Four studies showed AOT efficacy in improving upper limb functional recovery in participants with chronic stroke, two studies in sub-acute ones and one in acute ones. Six articles suggested its effectiveness on walking performance in chronic stroke individuals, and three of them also suggested an efficacy in improving balance. The use of AOT was also recommended in individuals with Parkinson’s disease to improve autonomy in activities of daily living, to improve spontaneous movement rate of self-paced finger movements and to reduce freezing of gait. Other two studies also indicated that AOT improves upper limb motor function in children with cerebral palsy. The last two studies, showed the efficacy of AOT in improving motor recovery in postsurgical orthopedic participants. Overall methodological quality of the considered studies was medium.The majority of analyzed studies suggest the efficacy of AOT, in addition to conventional physiotherapy, to improve motor function recovery in individuals with neurological and orthopedic diseases. However, the application of AOT is very heterogeneous in terms of diseases and outcome measures assessed, which makes it difficult to reach, to date, any conclusion that might influence clinical practice.Electronic supplementary materialThe online version of this article (doi:10.1186/s40945-015-0013-x) contains supplementary material, which is available to authorized users.