Hacker Newsnew | past | comments | ask | show | jobs | submitlogin

> That's hard to put into words in the middle of the night. In the end, I tend to recall my dreams quite well, many of which have a permanent place among other episodic real life events.

Yeah I found that when I start losing the ability to recall that if I just try to remember. That's the point of the journal anyways. But yeah, it is crazy the dreams I had and how real they were. Entire lives. And even the lessons you learn from choices you make. The emotions and how things even feel real to the though.

> and the dream noticing what you're up to, eventually booting you from the matrix.

This kept happening to me! The first time I did it started to fly but then just kept going up and up and it got brighter and I woke up haha.

> Take a moment to really appreciate the extreme level of graphics, and other senses the mind can render.

It really is insane and makes me think of the computation that the brain can do so efficiently. Like it doesn't get everything right -- which is why you're really able to get lucid (e.g. you can read something 2 times and it'll be different) -- but it is very real regardless. I'm sure there is a lot of compression going on and some tricks being played but it is often convincing. But what is also insane is that we know even pretty small animals dream in at least some realistic way. My cat tries to run in her sleep, meows, chirps like she does at birds, and wakes up with different emotions. I've had a pet rat do some similar things, at least the feet. It begs the question why this evolved, how important it is (essentially simulating your environment. Like learning from synthetic data, but shower thoughts on steroids), and how this relates to consciousness. And I wonder if we'll push harder to make machines do these incredible feats in much smaller packages and without nearly the same training or energy requirements. And I wonder how convincing these simulations are to the smaller creatures.



It seems like an extension of memory and map-making to me. And so many animals behave according to some map: squirrels and dogs digging up stored food; birds migrating; any animal that hunts or fights; any animal that brings food back to kids in a nest. All these actions require comparison of current sensory data against an ideal or memory of some benefit. The comparison requires some kind of model-forming/holding.

Now I write it out like this, it almost seems that the more interesting questions lie not in “can they do it” and more in “why do people have variations of ability; why does it stay active during sleep?




Guidelines | FAQ | Lists | API | Security | Legal | Apply to YC | Contact

Search: