I asked this question on Yahoo! Answers recently, but want to ask again here:
The “Best Answer” was:
Anyone else have any more info/details? Can anyone elaborate on the hormone production problem, for instance? But really, any info would be good. The more info, and the more details, the better.
They would die. You need to cycle through the deepest stages, esp. stage 3 in order to experience whatever regenerative processes occur during sleep.
New Scientist had a series (3) articles on sleep in one of their Feb. issues. Here are the links but you’ll need a subscription to read the full articles
Sleep and dreaming: Where do our minds go at night?
Sleep and dreaming: Slumber at the flick of a switch
Sleep and dreaming: Why can’t we stay awake 24/7?
btw, I know someone who has had multiple sleep studies done and many have shown her to spend much of her nights in REM sleep. She has night terrors. During the day she suffers from narcolepsy and a host of other psychiatric issues.
How long would it take for them to die, and what would happen to them in the meantime? I know in fatal familial insomnia it takes around 7 to 18 months for death to occur, and that’s without getting any sleep beyond stage 1 for the last two stages of the disease (which are said to last around three months and six months, respectively).
I’ve read that REM sleep is the most important for health, and that babies spend around 80% of their time in REM sleep. But on the other hand, I’ve also read that neuronal activity during REM sleep is very similar to that during waking hours. So I’m curious as to how it would affect someone who spends most or all of their sleep time in it.
Are the night terrors and narcolepsy the cause or the result of her spending too much time in REM? Are you able to give any more details about her psychiatric issues? (If you can’t or don’t want to, I understand.) Do you know how long she’s had this REM sleep problem? (e.g. her whole life, the past X years, etc.) And do you know just how much of her sleep she spends in REM, and whether it’s in one long bout or in phases?
The main thing to understand, and even the links I gave say this, is that we understand almost nothing about sleep - not it’s purpose, it’s mechanisms, it’s biochemistry, and on and on. So I don’t think you’re going to get solid answers to a lot of your questions but some clues are starting to emerge that seem to make a sort of sense.
As far as the importance of deep stage 2 and 3 sleep, I don’t think there is any dispute about that. The less you get, the more tenuous your existence - as in the case of the friend I spoke of. I can’t really go into much specific detail since I would run the risk of identifying her to anyone here who might know her, so it will suffice to say that it is possible that there were other factors that were more important and what one might consider causative. However it’s my personal opinion that the sleep issues could have been the lynch pin for the bulk of the serious issues at least.
In terms of the types of clues that seem to be emerging, one of those is that it’s not just the sleep stages that are important but what is often referred to as sleep architecture as well. Someone can spend more or less the “right” amount of time in each stage of sleep, but if they aren’t transitioning from one phase to another smoothly the way the process is supposed to work, then those people tend to have problems regardless.
This was true of my friend, well sort of. She didn’t get much deep sleep, but even when she did get some, her sleep architecture was completely boned. This was clear from the sleep studies but she also had a monitoring device that she wore to bed. I think it was something like this. Sleep From her descriptions, it’s able to tell when you fall asleep, how long you spend in each stage and some other information as well. You can even upload the information for analysis but I think that’s extra after a free trial period.
I know the device isn’t cheap, but she absolutely swears by it so if you know someone who needs to assess the efficacy of their treatment, something like this is de rigueur.
Thanks, deltasigma. This is actually for a story I’m writing, so this is all entirely hypothetical.
Basically, two characters now have a psychic link, and for the first little while they can only communicate when they’re both asleep — and when they’re asleep, they can only communicate while they’re in REM. During this period (which lasts a few weeks I guess?), they can’t really control their link, so it means they spend almost their whole time sleeping each night in REM sleep. They do get some time out of REM (when one of them is asleep and the other is not), but not much.
After some weeks, they will sort of grow into their ability and their sleep will return to normal, with them communicating psychically in their sleep only if they want to.
Look into transcranial magnetic stimulation. Also transcranial direct . . . something.
TMS uses a magnetic field to induce a current in the brain like a very controlled and sophisticated form of ECT - in fact it’s so far beyond that I hate to even put it in the same camp.
People have had mystical experiences on TMS and it is now FDA approved for depression and possibly OCD (obsessive-compulsive). It is also referenced in those links. They use either TMS or the other one to manipulate the sleep stages. So this might work well with your story.
brb
Your comment about sleep architecture is actually quite helpful. With my two characters, say one of them is asleep for an hour or two, and then the other one falls asleep. The first one to fall asleep would find themselves yanked very abruptly from whatever phase they were in, into REM sleep. While the second one to fall asleep would go almost straight into REM. Meanwhile, the first one to wake up would be awaking right out of REM sleep (yuck), while the second one would either return to normal sleep (mercifully for them) or find themselves jerked awake (again, right out of REM) by the shock of their partner leaving.
Also, do you know anything about sleep inertia? Wikipedia says waking up during slow-wave sleep yields greater sleep inertia than waking during other phases. What about REM sleep? Would that produce a fair bit of sleep inertia? Does it have a high arousal threshold (difficulty of being woken from), the way slow-wave sleep does?
Ah, according to Wikipedia, REM sleep has an even higher arousal threshold, which I guess also answers my question about sleep inertia.
Thanks for all your help, deltasigma.
The links really are your best resources. Here is just a snippet:
Thanks. I can’t really subscribe to NewScientist right now though, so I can’t really get anything useful out of the articles. 
Why “yuck”? It always seems much better to me to wake up during a dream, and isn’t that when REM occurs? What am I missing here?
You can have dreams during any stage of sleep I believe, it’s just that they are most commonly seen during REM sleep.
The friend I mentioned earlier, her night terrors tended not to occur during the REM phase IIRC. She would also have dream-like experiences during hypnogogic and hypnopompic episodes.
REM has the highest arousal threshold, therefore you probably have the greatest sleep inertia if you’re waking from it. Plus, even if you perceive yourself to be waking up during a dream, that doesn’t mean that’s actually the case. Not to mention that everyone’s different.
Also, you can have dreams during non-REM stages of sleep, most commonly in N3. They’re just less vivid, less memorable, and more disconnected.