Could this also be used to ask them yes or no questions? I wonder if that would be helpful or not. Perhaps "would you like us to pull the plug?"

Something like, count to 10 on the "yes" screen if your answer is "yes" or count to 10 on the "no" screen and so on.

This technique can be used to operate a speech synthesis device. It would then but possible to more completely evaluate the patients.

Given facilitated communications, I'd give this a BS until proven otherwise.

“The patients’ families were very happy when they found out that the patient could do maths”

A Chinese study in New Scientist where they admit to giving the family false hope, a lot of alarm bells.

Facilitated communication is an obvious scam, built on complete nonsense, with no empirical support. This is completely different.

I'm sceptical, because some people in (what was presumed to be) a vegetative state also showed those "signs of awareness". Not read the paper yet, though.