Introduction

Significant growth and interest in the field of rehabilitation medicine has been fueled in part by advances in rehabilitation science within an interdisciplinary research model (DeLisa, 2004). More importantly, research in rehabilitation has witnessed the application of the scientific method to specific functional problems such as the recovery of walking in neurologic populations (Barbeau and Fung, 2001) and the recovery of upper extremity (UE) use after stroke-hemiparesis (Taub and Uswatte, 2003). Recently, this translational research has spawned various protocolbased treatments, for example, to enhance walking in individuals with spinal cord injury (Field-Fote, 2001) and chronic stroke (Sullivan et al., 2002), and to enable use of the hemiparetic UE in adults with sub-acute stroke (Winstein et al., 2003). However, if rehabilitation medicine is to join the ranks of other evidence-based medical and pharmaceutical practices, objective treatment protocols will become a necessary component of valid efficacy and effectiveness research (Whyte and Hart, 2003). The development of specific and objective rehabilitation treatment protocols will be a clear signal of progress in the field of rehabilitation medicine. At present, the majority of published protocols in neurologic rehabilitation lack an explicit scientific rationale for the intensity, duration, and content (e.g., task-specific versus muscle-specific) of training used within the rehabilitation treatments. Without an explicit rationale (or even hypothesis), the precise parameters of training for a given rehabilitation treatment can take on a mythical quality with hidden meaning at worst, and lead to “blind” following at best (Dromerick, 2003).