In message <199705062111.WAA21416@typhoon.dial.pipex.net>, Dave
Pape <davepape@dial.pipex.com> writes
>Hey, now I reckon that self-referential thought-processes giving rise to
>self-replicating emergent phenomena could explain what happens when you
>learn a new concept.
>
>Say... I was learning how to highlight a stretch of text in bold type.
>
>At the start of the learn, I'd be... pressing the mouse button at the start
>of the stretch, dragging the mouse (without letting go of the button) over
>the text, letting go of the button, then clicking the Format menu, choosing
>Format Text, and choosing "Bold".
>
>At the end of the learn, I'd just be MAKING THE TEXT BOLD. (I think...
>because everything else is automatic... I'm not conscious of the little
>ideas...)
I see. So the learning phase consists of the subconcepts
triggering each other, presumably sometimes misfiring (learning errors)
This takes conscious effort to tighten the autocatalytic loop between
the subconcepts and the perception of what is happening to the text.
Yes I think an important difference between the learning phase
and the learned actions is to do with the engagement of the perceptual
system to give feedback.
>
>Now what I'm wondering is... at the start of learning a new concept, is it
>the case that:
>
>The new concept isn't autocatalytic.
Yes I would say the attention/perception/text on screen are
involved in the learning phase to tighten the autocatalytic loop so it
can self execute (which in this example would mean being able to block
text even if you couldnt see what was hapening.) I agree that this is
the difference before and after something has been learned.
Tony Hindle.