Action observation has a positive impact on rehabilitation of motor deficits after stroke

Abstract

Evidence exists that the observation of actions activates the same cortical motor areas that are involved in the performance of the observed actions. The neural substrate for this is the mirror neuron system. We harness this neuronal system and its ability to re-enact stored motor representations as a means for rehabilitating motor control.

We combined observation of daily actions with concomitant physical training of the observed actions in a new neurorehabilitative program (action observation therapy). Eight stroke patients with moderate, chronic motor deficit of the upper limb as a consequence of medial artery infarction participated. A significant improvement of motor functions in the course of a 4-week treatment, as compared to the stable pre-treatment baseline, and compared with a control group have been found. The improvement lasted for at least 8 weeks after the end of the intervention.

Additionally, the effects of action observation therapy on the reorganization of the motor system were investigated by functional magnetic resonance imaging (fMRI), using an independent sensorimotor task consisting of object manipulation. The direct comparison of neural activations between experimental and control groups after training with those elicited by the same task before training yielded a significant rise in activity in the bilateral ventral premotor cortex, bilateral superior temporal gyrus, the supplementary motor area (SMA) and the contralateral supramarginal gyrus. Our results provide pieces of evidence that action observation has a positive additional impact on recovery of motor functions after stroke by reactivation of motor areas, which contain the action observation/action execution matching system.

Introduction

Motor deficiency is the leading cause of disability following stroke (Duncan et al., 1992). The mechanisms underlying recovery are multifold including active relearning strategies and passive processes of lesion adaptation (Seitz et al., 2002). The traditional neurorehabilitative approach to treat motor deficits after stroke is mainly based on techniques aiming at stimulating the use of the paretic limb during supervised training sessions (Rossetti et al., 2005). The main tenet is that practice of repetitive, active movements by a paretic limb leads to effects induced by positive neuronal plasticity. The effects of repetitive training are empirically well proven (Shepherd, 2001, Aichner et al., 2002, Jang et al., 2003, Byl et al., 2003, Dombovy, 2004): this approach was demonstrated to be more effective than conventional physiotherapeutic approaches (Beer, 2000, Hesse et al., 2002), with a smaller learning time of manual arm movements (Beer, 2000), a faster recovery and smaller therapy lasts (Kwakkel et al., 1999). Examples include the “constraint-induced movement therapy” (Taub et al., 1993, Duncan, 1997, Elbert et al., 2003), the motor relearning program (Langhammer and Stanghelle, 2003), and repetitive arm training (Buetefisch et al., 1995). These neurorehabilitative trainings are supposed to have attained the best rehabilitative outcomes so far.

There is increasing experimental evidence that motor areas are recruited not only when actions are actually executed, but also when they are mentally rehearsed or simply observed (for a review see Jeannerod, 2001). The neurophysiological basis for this recruitment relies on the discovery of mirror neurons, first described in the monkey (Rizzolatti et al., 1996, Gallese et al., 1996). These neurons discharge when the animal performs an object-related action with the hand or the mouth and when it observes the same or a similar action done by another individual. It has been shown, also in humans, that the observation of actions recruits the same motor representations active during the actual execution of those same actions (Fadiga et al., 1995, Hari et al., 1998; for review see Rizzolatti and Craighero, 2004, Buccino et al., 2004a), and this in a somatotopic manner (Buccino et al., 2001). The mirror neuron system has been shown to be involved in imitation (Iacoboni et al., 1999) and imitation learning (Buccino et al., 2004a).

While motor imagery has been shown useful in sport practice and rehabilitation especially when combined with conventional rehabilitation treatment (Page et al., 2001), observation of actions has not yet been tested as a potential tool in neurorehabilitation.

The major aim of this study was to assess whether action observation therapy may lead to clinical improvement of motor impairment in chronic stroke patients, as measured by standard functional scales. Based on the findings of previous studies concerning the use of mental techniques in neurorehabilitation (Page et al., 2001), we combined action observation with the direct effects of action execution. Our hypothesis was that the activation of motor areas by action observation becomes reinforced by the concomitant active execution of the observed actions (Binkofski et al., 2004, Buccino et al., 2006). Additionally, by means of fMRI we studied the reorganization within the motor areas following this treatment, using an independent sensorimotor task consisting of object manipulation.