Post #3087527

Why on god's green earth would you believe meta-beliefs to be true "by definition"? Historically there have been so many changes to meta-beliefs, from divine inspiration to bayes' law. I have memories of my meta-beliefs changing as a child, and honestly my meta-beliefs even change depending on my mood. So why the heck would I not expect the same to be possible in the future? As far as I can tell, subjective belief probabilities are a sort of integration over forward propagation through lines of reasoning, world models, and meta-beliefs that occur in one's brain. This belief finds practical application in stuff like me making sure I eat if my meta-reasoning is cranky (assigning too little value to new information, expecting hostility to be more productive than empathy, etc.). Though of course I am not super confident in any precise meta-meta-statements because human brains aren't well understood yet by anyone, let alone me. I suppose you could reduce all that complexity to a single line of reasoning that you try to trust with utmost confidence, but then I would expect you to get akrasia as the rest of your brain sees all the things you are missing and starts ignoring your "certain meta-beliefs".