Harri Besceli's Shortform

though plausibly a minute of dream experience feels longer than a minute of waking experience, eg. sometimes I go to sleep for 15 minutes and it feels like it's been hours

The argument from LaBerge and other psychologists is that when it 'feels like it's been hours', it's just an illusion or narrative fiction, which you can't detect because the critical thinking has been disabled, and no more real than when in a play, the curtain drops and then rises 'hours later'.

"You may be wondering, then, how you could have a dream that seems to last for years or lifetimes. I believe this effect is achieved in dreams by the same stage trick that causes the illusion of the passage of time in the movies or theater. If, on screen or stage or dream, we see someone turning out the light as the clock strikes midnight, and after a few dark moments, we see him turn off an alarm as morning shines in, we'll accept (pretend, without being aware we pretend) that many hours have passed..." [pg 22, Exploring the World of Lucid Dreaming (1990); Stephen LaBerge & Howard Rheingold.]

You can't remember or produce hours of experience corresponding to that; you simply remember having remembered, etc, shortcircuiting to the final mental state without having actually gone through the apparently preceding states, like the skeptical argument that the world could have been created 5s ago by God in the exact same state as it would have been if it had actually been created billions of years ago naturally. Similar to how you can spend decades or 'billions of years' in a psychedelic state, or be 'trapped forever in a timeless crystal' like Scott Alexander mentions - don't worry, time passed anyway and he got better. (See also false insights, retrocognition.) And when you try to intervene by waking people up in lucid dreams or doing tasks, they seem to still be processing time at a normal 1:1 rate. An additional piece of evidence is how people say "time slows down" in an extreme event like a car crash, but if you have people go bungiejumping and you test their perception of time using clocks, they aren't able to perceive anything they normally wouldn't. Quantity-wise, it seems like 1 minute of dreaming ~ 1 minute IRL.

Now, quality-wise, they may be able to remember more, there's some followup research on that, but that is more about motivation/valence intensifying things than ability or unlocking some 'overclocked mode' in the brain. (They always could, they just didn't bother.)

So the question devolves back to the intensity of the experience. If you want to optimize it for some sense of utility, then if dreaming has near-zero intensity, optimization is a lot less appealing than the naive toting up of"I spend X minutes a day in REM sleep, so that's Y hours a lifetime" would suggest.

My own observation is that while my dreams are not infrequently quite unpleasant and nightmarish in content, probably the majority of my dreams are negative, the corresponding emotional reaction is not there. Many dreams ought to have one waking up in a cold sweat, heart pounding, unable to focus for the rest of the day, but they don't. (And when they do, we call them 'night terrors'. Why aren't all bad dreams "night terrors"? Why does it decrease with age, and correlate with psychiatric problems?) But even if one were to write them down (and I did for a while in a dream journal) to block forgetting, I feel back to normal practically upon waking, and definitely within a short time. I've surely had thousands of terrible, terrible experiences in my dreams (like last night, I had a long and vivid dream about waking up to my father having suddenly died overnight, where I was "very sad" - but upon waking, I was much more cheerful than one would hope I would be if my dad had abruptly died), and they still have not added up to any major PTSD or depression or even noticeably affected my waking life. So that is pointing towards the experience of dreaming being far, far less intense than it seems.

We could also point to sleepwalkers of various sorts: even when executing complex actions (like murdering someone), I've never seen any accounts which mention deeply felt emotions. (WP emphasizes their dullness and apathetic affect.) Or if I watch a cat, paw twitching as they clearly are hunting something in a dream, they seem excited, but not nearly as hyper-excited and intense as when I watch the same cat in a real hunt. (We could also point to anesthesia awareness survivors, who do [LW · GW] seem often deeply traumatized by a single awakening during surgery - even ones that they don't remember afterwards - and in fact triggering endless nightmares as a side-effect.)

I think this makes sense from a functional perspective. Why would dreams have any more than a shadow of waking life's impact, any more than daydreaming about yourself as an action hero or writing a comic book script feels like the real thing would? If dreaming is for some sort of model-based learning process, it's probably best to decouple it from the autonomic nervous system in the same way you're definitely decoupled from your muscles and limbs. If you are imagining yourself being shot a dozen times a night to explore your imagined responses or experiences, you don't want to physically react like that a dozen times a night! That would be extremely bad for you physically and psychically: if you reacted to dreams even a small fraction as much as to the same event in real-life, then after decades of dreaming many times per night, night after night, you would have a dozen different PTSD disorders and half the DSM. A full emotional reaction/qualia would appear to be wasteful and unnecessary for the learning. A shadow suffices for simulation.

So that's a physical reason why dream-experiences would be far more impoverished (and thus of less moral weight) than real-life experiences.

There is a missing factor from this, which is anticipation and retrospection, in terms of weighting the relative value/importance of experiences.

Personally, I heavily discount the importance of imaginary experiences. A real papercut is hundreds of times more impactful in my decision-making and valuation than a dreamed amputation.