The most extraordinary claim in this post — that improvements persist after the computer is off — has no research behind it.

There's no research at all on whether the improvements stick around after the computer is turned off, so that part of the claim is unsupported. The specific strength percentages also have no research backing them. However, other studies do show that brain-computer interface training can help stroke patients move their arms and hands better, which would include actions like scratching your nose or wiping your mouth. But those studies didn't always test those exact movements — some looked at finger or wrist motion instead — so the claim is a bit more specific than what the research directly shows.