I used to have a fear when I was a kid that when you were asleep you were actually wide awake but unable to move, and just forgot your panicked attempts to move each morning.
I don't sleep well.
I wouldn't expect to remember dreams if I'd actually been awake, paralyzed, and panicked.
I've also watched other people sleep, and they generally have facial expressions that change over time. My prior on "paralyzed in a way that causes dynamic and pleasant facial expressions despite actually being tortured" is very low.
I'd expect people who suffer from sleep paralysis to have much different perspectives, of course. I'd also expect sleep walkers (and talkers) to have much different perspectives.
Sleep paralysis is a real thing, as are waking instances of it. I have distinct memories of being awake and paralyzed in bed,lasting for up to several minutes. Weird, but not utterly terrifying.
Mine, when I was a teenager, was that you actually died every night, and another mind with your memories woke up instead. I figured that there was not much point in worrying since I couldn't do anything about even if it was true, and I needed to sleep besides. Still pretty scary!
If not remembering it makes it not matter, nobody can remember anything after they die, and everyone's going to die, wouldn't that mean nothing matters?
My answer is that there's no persistent her. The past is every bit as real as the present, and the her from the surgery is just as real as the present her.
Would you pay $10 to have [extremely pleasant experience of your choice] for one hour, but have all the memories of it erased immediately afterwards?
Like I said, all the memories being erased afterwards is a given.
From the point of view of amnesiac future me, it would be just like I had payed for someone else to have a pleasant experience, which I wouldn't remember for obvious reasons.
Kahneman puts it similarly in Thinking Fast And Slow -- he describes "two selves" -- the experiencing self, and the remembering self.
Please elaborate. My immediate reaction is Occam's Razor makes the idea of multiple distinct selves pretty unlikely. But then I think maybe you're defining self differently than I would, and it's certain that the brain and mind have a certain amount of modularity. Even if our self can be divided into several functioning parts, and even if we are using/ experiencing some part stronger than others, that doesn't seem to be evidence of multiple individual selves. Even when you have modularity that results in some weird behavior, that doesn't suggest multiple selves would be split between the past and present. From science, I think about the fact that say, we can see things in the present and imagine things in the past and I've never heard about them being done by distinct parts of the brain. But the strongest intuitive reason and evidence I'd put forth, would just be my experiences with friends who have PTSD. When undergoing a flashback they don't experience things as if it were a memory of the past. That combined with the lack of evidence that people with/ who get PTSD don't seem to be any different than nuero typicals. (There's some evidence that some people are more likely to get PTSD, but no evidence that anyone is immune.)
In the naive model of the self, where "you" make a decision, and then your body executes that decision, and it's the same "you" every time. And it's true that this is a very simple model, and if we didn't have any other evidence, that's what we would probably go with.
But that model that doesn't explain the Peak-end rule very well -- how is it that a certain experience is worse while experienced, but better when remembered? Similarly, it's hard to explain hyperbolic discounting in terms of a continuous self. And how would self-talk be effective (and it is) if the self were indivisible -- whom would you be trying to convince? Finally, how can we make a decision before becoming conscious of it?.
A better mode of the self would say that there's not one thing that you can call your "self" -- instead, your mind is basically bodged together from a bunch of random parts. So when we say that there are different experiencing vs remembering selves, what we're really saying is that reasoning about your preferences as though you only had one set of preferences, doesn't work. At the time, you really prefer to avoid the additional bit of small pain at the end. But when you remember the experience, you really do remember it as being better than you otherwise would. So what do "you" prefer? There's no single sensible answer to that.
"But that model that doesn't explain the Peak-end rule very well -- how is it that a certain experience is worse while experienced, but better when remembered?"
Anchoring bias. Forgetting. Contemplation of past events in a broader context. Rewriting memories. All these things happen.
"And how would self-talk be effective (and it is) if the self were indivisible -- whom would you be trying to convince?"
That's a good point. Although, I already think that the brain is modular, and it could be very possible for different parts of your brain to be experiencing different things, the biggest problem is the point at which one self ends and another begins. You explain this a little bit in the next paragraph, which makes it seem like you basically agree with me, although it seems very vaguely defined. Even if they're like a baseball team, trading modules ocassionally, there's a lot of vagueness that seems not to form many expectations given specifics, so experimentation would be very hard.
How many selves are there? Do they vary with time? Is there really any reason to suspect there's only two? Is the peak end rule the best evidence for how they're divided? What are the expectations based of of this? What experiments are being done?
I agree there isn't a sensible utility function for individual humans. I think as far as that goes, they must be on the right track. I just wish there was more evidence, and more expectations we could derive from the model.
My immediate reaction is Occam's Razor makes the idea of multiple distinct selves pretty unlikely.
That's not the way Occam's Razor works. The quantity to minimize is the number of bits needed to specify a theory, not the number of objects the theory describes.
There are multiple humans. Taking the arbitrary number of humans alive at anyone point, and saying that census data must be faked, and that there are only as many humans as I have direct experience of, would be shitty application of Occam's razor.
Saying I have a second "self' without giving evidence of the self, without any specifications of what that self would be like, and how that's different, behaviorwise, from a human with one self, and how that gets us more explanatory power for our postulated entities, is exactly what Occam's razor is for,
I mean, this whole thing about the person you are "dying" every time you change state seems like empty navel gazing, seems to apply the word "death" in a way that is meaningless, and seems way more like the "wave function collapses and no worlds arise" interpretation of QM than the many worlds one.
Brains are very complex. If we end up with a grand theory of nuerology, and that incorporates the idea of two selves as the simplest explanation, I'll listen. But my understanding was that we really aren't there yet. Until then, iy's just an arbitrary postulation, like saying we have 3 selves or 4 selves or 9 epistemological gremlins.
http://www.smbc-comics.com/index.php?db=comics&id=2705
Addresses questions like "If I don't remember, but it definitely happened... who suffered?" in a rather non-obvious way (non-obvious to me, anyway).