A degree of motor recovery is typically seen after acquired brain injury in children. The extent to which rehabilitation efforts can claim credit for this is disputed. Strong correlations between late impairment outcomes and early severity and impairment indices are seen both in adults and children. These correlations have been interpreted by some as evidence that recovery is largely intrinsic and that any additional rehabilitation effects are small. Such views are belied by published animal studies demonstrating the possibility of large rehabilitation effects. Animal models suggest that to achieve similar rehabilitation treatment effect sizes in clinical practice, rehabilitation ‘doses’ should be greater, rehabilitation efforts should start sooner, and premature accommodation of impairment should be avoided.