The applicability of motor learning to neurorehabilitation

Statements to the effect that recovery is a form of learning or relearning are commonplace in the field of neurorehabilitation. In this chapter, motor training will refer to what is done to the patient and motor learning will refer to what the patient may do in response. This distinction is important—just because training is happening does not mean that anything is being learned. The relearning premise for neurorehabilitation is based on three other a priori assumptions. First, that the nature of the deficit to be rehabilitated through learning is known. Second, that the kind of motor learning that should be targeted by training is known. Third, that patients after stroke have an intact learning capacity despite impaired performance. In this chapter the focus will be mainly on rehabilitation of arm paresis after stroke, which results from damage to motor cortical areas and/ or their descending pathways. This narrower focus is essential if the topic of learning and neurorehabilitation is to remain within the bounds of a single chapter. That said it is hoped that the general principles introduced here, which will be emphasized over details, are broadly applicable across the range of post-stroke impairments and to other neurological conditions. Arm paresis after stroke refers to loss of strength and motor control, along with changes in phasic and tonic muscle tone [1] . Non-neural peripheral changes in muscle, joint and tendon properties can also contribute to the paresis phenotype. In this chapter it will be assumed that treatments for strength, tone (spasticity) and contractures are not based on motor learning principles and so will not be addressed further. Note again that one can train for strength but this is not motor learning. Thus, the starting point for this chapter is that when learning is invoked it implies either improving motor control or finding alternative compensatory strategies with effectors/joints/muscles in which motor control remains relatively intact; in either case, response to training is assumed to have mechanistic commonalities with motor learning in healthy subjects. It will become apparent after reading this chapter that the assumption that one can equate recovery and motor learning is subject to several fundamental caveats.