Speed of motor re-learning after experimental stroke depends on prior skill.

Many motor rehabilitation therapies are based on principles of motor learning. Motor learning depends on preliminary knowledge of the trained and other (similar) skills. This study sought to investigate the influence of prior skill knowledge on re-learning of a precision reaching skill after a cortical lesion in rat. One group of animals recovered a previously known skill (skill training, followed by stroke and re-learning training, TST, n = 8). A second group learned the skill for the first time after stroke (ST, n = 6). A control group received prolonged training without stroke (n = 6). Unilateral partial motor cortex lesions were induced photothrombotically after identifying the forelimb representation using epidural stimulation mapping. In TST animals, re-learning after stroke was slower than learning before stroke (post hoc repeated measures ANOVA P = 0.039) and learning in the control group (P = 0.033). De novo learning after stroke (ST group) was not different from healthy learning. These findings show that skill learning can be performed if the motor cortex is partially lesioned; re-learning of a skill after stroke is slowed by prior knowledge of the skill. It remains to be tested in humans whether task novelty positively influences rehabilitation therapy.