Dreams made a whole lot more sense to me when I learnt about idea of the brain as a predictive processing machine. It asserts that the brain works as a sort of recursive two-sided process, with theory creation on one side and evidence collecting on the other. Recursive in that the visual processing system might get a "is that a door?" theory from a higher level system, which it'll then answer by collecting and trying to match evidence from multiple lower level systems ("is that an edge?" "is that a shadow?" etc.) until it eventually reaches the level of processing stimuli directly, with each system on the way acting as both questioner and answerer.
In this framework dreaming is when the evidence collecting side becomes a lot less rigorous than it usually is, so it says "sure, why not" without reference to a stimulus (i.e. a way to be falsified). In other words it's the brain's predictions continuously coming true.
> brain as a predictive processing machine. It asserts that the brain works as a sort of recursive two-sided process, with theory creation on one side and evidence collecting on the other.
SlateStarCodex has a nice article[1] about this.
> The key insight: the brain is a multi-layer prediction machine. All neural processing consists of two streams: a bottom-up stream of sense data, and a top-down stream of predictions. These streams interface at each level of processing, comparing themselves to each other and adjusting themselves as necessary.
> ...
> You’re not seeing the world as it is, exactly. You’re seeing your predictions about the world, cashed out as expected sensations, then shaped/constrained by the actual sense data.
In this framework dreaming is when the evidence collecting side becomes a lot less rigorous than it usually is, so it says "sure, why not" without reference to a stimulus (i.e. a way to be falsified). In other words it's the brain's predictions continuously coming true.