Call me lazy, but I’m going to point you to a post I made last year about the idea of distributed representation. I violated one of my own self-imposed rules and created a little flash presentation to try to visualize how this all worked.
One of the things I was trying to wrap my melon around was how this all related to Zone of Proximal Development and there doesn’t seem to be any conflict. If we stipulate that an knowledge domain has a set of potential connects then ZPD is represented by the relative proportion of those connects that have been realized. If an individual has realized all potential connections, that would define the condition of “known.” If an individual has realized none of the potential connections, that would represent “unknown.” Someplace in between would represent the zone of proximal development, that is, a place where new knowledge could be linked in. For the purposes of the network representation, that would be “the edge.”
Now one of the things that really appeals to me about this is that it gives me a handle on the notion of “known” that I have been troubled by in the past.
Remember that we stipulated that you can’t learn something you already know. It’s as if there’s no potential there for the “learning” to happen once it’s already taken place. I know the alphabet so I can’t learn it again. But does knowing the letters of the alphabet — the shapes, the sounds, the patterns — represent the full “knowledge” of the cognate we’re calling “the alphabet” and can anybody ever consider that they know all there is to know about anything? Are there always potential connections? If so, does that mean that my diagram of ZPD really has no solid center? No region of the knowledge base that is absolutely completely learned?