What you call "conscious attention" seems to be a sort of cognitive process. What the article calls unconscious state is a state. The idea is that in some states, some processes (like predictive coding) don't take place.
Comatose patients show EEG activity related to sound and language processing. That has been known for at least 30 years (a quick search turns up e.g. https://www.thelancet.com/pdfs/journals/lancet/PIIS0140-6736...). The article also speaks about attention. I even quoted that bit.