It is also the basis for past memories being happy - it is still "you" that did those pleasant things, which gives a form of comfort (as opposed to "I" want that, and can't have it, which is also true, but a much more miserable way to reflect on your past).
Once, years ago, I broke this fourth wall by accident. It was during one such happy time that I knew I wasn't likely to be able to repeat. For some reason, out of the blue, the thought occurred to me "Sucks for you, future Shylock, that you won’t get to do this, and I do!”
As soon as this thought passed, it felt like I had accidentally opened a door to a strange and dark portal, through which lay a basilisk of second-guessing, unhappiness and intrusive thoughts. I was determined not to ponder it too long, and not to encourage this thought to persist. I didn’t think I would actually start making impulsive decisions based on a philosophical conviction that amounted to “lol, discount rates should be infinity”. But the various versions of me were, up to that point, always linked by a strong sense of mudita, or sympathetic joy. If the me of the past experienced something, happy days. If the me of the future gets to experience it, happy days. A world where this base psychic cooperation doesn’t exist seemed like a world of madness and deformity.
You might think that nobody could possibly operate this way. But there are good reasons to suspect that other people’s internal psychic states can be radically, wildly different from your own, and you might not know it. We know that people have enormously different abilities to form mental images, with one end being eidetic imaging where mental imagery pictures are completely crystal clear and photographic, and the other being aphantasia where they are completely absent. While this fact could have been known any time in human history, as the tech needed for it involves the complicated process of “asking people some basic but unusual questions”, as far as we can tell it wasn’t discovered until Francis Galton figured it out in 1880.
We know that people also vary in whether they have an internal monologue, a voice in their head talking to them consciously like a conversation. I have seem numbers on this for the population placed at 30-50%, from lazy googling of the subject. Like before, both the people with it, and the people without it, can’t quite believe that the other group exists, until this is pointed out.
And even here, if you believe the context described in Julian Jaynes’ “Origin of Consciousness in the Breakdown of the Bicameral Mind”, it is not always obvious that this monologue, if it exists, was conceived the same way. That is, if you have one, you probably think that the voice is “you”. Jaynes argued that in the Bronze Age, people thought that the voice was the gods, commanding them to do things. Which is to say, you probably never wondered whether the voice is really you, and just assumed, like all this stuff, that your assumptions on the matter were universal. But they probably aren’t.
I wonder whether people who are impulsive and sufficiently low IQ actually lack this instinct of psychic unity across time, or have a very underdeveloped sense of it. I imagine they also just don't think about it much. But it is one possible answer to what "very high time preference" might mean as an underlying foundation.
At a minimum, a sufficiently high time preference becomes had to distinguish from the outside from failing to incorporate futures selves as the same as present self. If I were to guess, I suspect that this basilisk, rather like Roko’s Basilisk, only applies to a certain kind of abstract thinker. Which is to say, most people with low time preference think that the two selves are the same instinctively. Most people who have high time preference just don’t think about the matter at all. Not many people would be interested enough to wonder, and philosophically inclined enough to seriously get concerned. But the question of what exactly it is that makes “you” you is quite thorny. Indeed, one simple answer to Roko’s Basilisk is the view that an atomically identical version of you is actually better thought of as a twin, and not you. In that case, if a future superintelligent AI tortures my twin, yeah that sucks for him, but that’s life sometimes. Ironically, this means that there’s no value in torturing said twin, because it’s not going to affect me today. Just take the one box, bro.
Incidentally, there’s a revealed preference test for whether you really believe that an atomically identical you is really you. Suppose someone had a cloning machine. They had cloned you against your will, and were threatening to make an exact copy on Mars. You can press a button, and transfer half your net wealth to a bank account controllable only by that person. Do you press it? What if they make another copy? And another?
Like so many things, when you ask people to actually pay for their beliefs, you might be surprised by how many people take the practical, basilisk-proof answer.
But if the atomically identical version isn’t you, why is the much less similar version from 20 years ago, or from 20 years in the future? It is surprisingly hard to say.
Good, consistent decision-making turns out to be supported by a giant Newcomb’s Box rock, overhanging a cliff. Take the second box, and it vanishes.
I hope I haven’t seeded anything too bad here.
And with that, I strongly encourage you to go back to the usual movie. It is an occasional humorous incongruity when movies break the fourth wall. But it is nearly always better when the fourth wall is respected, without which, the whole concept of the movie quickly breaks down.
