Discussion about this post

User's avatar
Overcoming Bias Commenter's avatar

keep on trashing such stupidity ,any idiot old enough knows intuitively of that type of rationalisation.keep your energy to understand the really mysterious.the nature of mystery for a start.and stop thinking that the universe is waiting for you to know where it stands.

Expand full comment
Marc Geddes's avatar

Think of a hologram here, and consider that it can have various 'resolutions' - you can keep cutting up a hologram into pieces and still see the whole picture but the resolution degrades , the picture becomes more and more coarse grained and fuzzy. I think that (past a certain minimum threshold), intelligence (and rationality) is just like that hologram - the higher IQers can't in principle see any concepts the lower IQers can't, they just see the concepts in more and more fine-grained detail.

Suppose reasoning through categorization starts with a very few coarse grained concepts. Any probabilities derived from those will obviously also be very coarse-binned and imprecise. But now, consider, the fan of categorization can start introducing more and more fine-grained distinctions (more and more 'multi-choice questions' or more and more analogies) which have the effect of 'perturbing' the original probability estimates to a more and more fine-grained level.

Consider the hologram analogy again: truth is equivalent to the entire picture and Bayes deals in precise detail with any small area of it. Categorization is a way of adjusting the 'resolution' of the whole picture to any desired level of magnification. By making enough fine grained distinctions ('perturbations' of original prototype concepts), categorization can duplicate Bayesian Induction to any desired degree of accuracy. Ergo, Bayes is really just a special of categorization where the resolution of the picture is turned up to 100%.

Expand full comment
42 more comments...

No posts

Ready for more?