Based on the Glasgow Coma Scale, people with Locked-In Syndrome _are also in comas, technically_. Figuring it out seems remarkably hard, and is almost certainly wildly under-detected. I'd be interested to see if we are able to use tools like this to discover whether our assumptions about comatose states are at all accurate.
Also, of course, the obvious Inception-oriented ideas would be good to test out as well.
You know, I've always wondered why, in edge cases like this, it's not standard to run the patients through fMRI to detect whether or not the patient is actually conscious before just consigning them to a bed with no interaction. Or am I mistaken, and this is routinely done?
For every 1,000 people who undergo general anesthesia, there will be one or two who are not as unconscious as they seem — people who remember their doctors talking, and who are aware of the surgeon’s knife, even while their bodies remain catatonic and passive. For the unlucky 0.13 percent for whom anesthesia goes awry, there’s not really a good preventive. That’s because successful anesthetization requires complete unconsciousness, and consciousness isn’t something we can measure.
Here we have a much more common problem than locked-in syndrome, a problem which anesthesiologists have been studying since before their job had a name, but we can't even solve that because we don't even have a decent working definition of what we mean by "consciousness" in these contexts, let alone what we can measure to determine someone's "consciousness". It's an active field of research, which is a fine thing to remember whenever we start getting too high on science fiction:
Also, of course, the obvious Inception-oriented ideas would be good to test out as well.