> > watch toddlers playing with toys—their imaginations are boundless and they are able to imbue even the most mundane objects with a sense of wonder and magic.

> > watch toddlers playing with toys—their imaginations are boundless and they are able to imbue even the most mundane objects with a sense of wonder and magic. Also a lot of that "play" is "merely" epistemic and phenomenological research, i.e. hard work. There is a lot of creativity in designing and selecting experiments that work with what is at hand. And of course discovery is fun -- that same "wonder and magic" is still experienced in adulthood when you validate something you've believed for a while or just realized. Piaget discusses this extensively, though not in the vocabulary I used. And for the mandatory AI/CS linkage: Piaget was Papert's thesis advisor, so every RNN user or practitioner implicitly depends on Piaget's insight.