Are concepts learned? Or more precisely, "why do we feel we have to think of what we learn as things or concepts? Why must we `thingify' everything?" "Thingifying" isn't always bad, but Minsky feels that in general it is disastrous for a science of mind.
"Instead of assuming that our children come to crystallize a single `concept of quantity', we must try to discover how our children accumulate and classify their many methods for comparing things." Minsky's rule for learning is to always try to combine related agents first.
(NB: Two points. First, societies-of-more are learned. Second, what does it mean for agents to be related? Minsky hints that it has to do with shared brain resources, and therefore possibly similar brain locations.)
Pearl Street | Society of Mind Home Page | Dawson Home Page |