Introduction

Stroke can interfere with every single capacity of the human brain. Many patients experience transient deficits that disappear within 24 hours (Hennerici et al., 1988) or suffer from slight or moderate hemiparesis that regresses completely within a couple of days (Donnan et al., 1991; Duncan et al., 1992). Of the patients with severe completed stroke, approximately 50% recover from hemiparesis (Heinemann et al., 1987; Gresham et al., 1998). In most of these patients, recovery takes place within the first four weeks after infarction, usually starting with shoulder and synergistic arm–hand movements (Donnan et al., 1991; Duncan et al., 1992; Katrak et al., 1998). In subcortical infarctions, functional restoration may be slower, continuing over several years (Fries et al., 1993; Manto et al., 1995). This suggests that a variety of mechanisms may be involved in the restoration of motor functions after brain infarction. Also, clinically undetected white matter changes and the prevalence of silent stroke lesions in patients presenting with the symptoms of acute stroke affect the cognitive and attention state of patients and their capacity to recover (Kase et al., 1989; Sultzer et al., 1995; van Zagten et al., 1996). Therefore, the challenging aspect in each stroke patient is the question to what degree he or she will recover. To monitor recovery, the importance of an adequate assessment of the functional deficit and of the disability level of daily activities has increasingly been recognized (Duncan et al., 2000), since correlation with the tissue perfusion state, the recruitment of peri-lesional and remote areas during functional activation, and changes of cortical excitability, as assessed with transcranial magnetic stimulation (TMS), will provide the basis for an evaluation of the recovery mechanisms and of the efficacy of new therapeutic interventions that have been or will be introduced into clinical use.

Neurological recovery is thought to be the behavioral equivalent to post-lesional cerebral reorganization. However, recovery may be differentiated into spontaneous recovery occurring early after stroke and induced recovery occurring later asa result of dedicated physiotherapeutic approaches. Since most of the recovery takes place early after stroke, a major question regarding the underlying mechanisms is how far the restitution of tissue function is mediated by reorganization of the residual neural network. Today's non-invasive investigation techniques provide the means to explore the living human brain for processes related to normal behavior and cerebral reorganization.