John Anderson (950629.0630 EDT) wrote:
Suppose that the results of an action, like choosing a certain golf
club, reduce the error between some desired (reference) perception,
like getting the ball in the hole, and the actual perception.
But an action cannot reliably produce an outcome when disturbances are present.
Now if
some process strengthens the synaptic connections in the control
systems whose activity produced the action that led to those "good"
results, then when those control systems become active again, a
similar neural response will be likely to occur. If the global set of
active control systems in which it occurs is similar to the one
preceding the first "good" result, which I think is a reasonable
expectation given a similar situation, then the overall action of the
organism (the golfer) will probably be similar, and hopefully (for the
golfer) will lead to a similar result.
It seems to be that much of the S-R vs. PCT debate could be cleared away by
simply keeping in mind that a given action cannot reliably produce a
desired outcome. However, controlling for some combination of lower-level
perceptions CAN reliably produce an outcome, that is, a higher-level
perception. Isn't that what HPCT is all about?
So learning is not strenghtening connections between stimuli (perceptions)
and responses (actions), but rather strengthening connections between
different types of perceptions. For Calvin to learn to ride his bike, he
needs to figure out what controlled lower-level perceptions will result in
the higer-level perception of success at bike riding. S-R learning will
not do this. P-P learning will.--Gary