> This idea, so far as PCP is concerned, relates to the core ideas
>one uses when speaking of MEMORY PROCESSES. What is stored for use by a
>system of constructs. I have proposed that we should think of storage
>in terms of TWO POLED CONSTRUCTS. Thus, each anticipatory construction
>-- whether it be a motor movement by which we anticipate a valid body
>position, position of a putative object in space, or our self as text
>for another to process -- represents a NEW constructions. That is,
>constructions should not be taken as the unit of storage.
> I have come to believe that this is a central issue in working
>with PCP.
I have been thinking about the issue of what is "stored" in "memory"
recently. I would ask Jim to elaborate further on just what he means when
he says that a "two poled construct" is stored. In what form is the
construct represented? How is it stored?
I say this because I would like to evaluate the extent to which Jim's
ideas are consistent with the following line of thinking: The issue of
"storage" is a very problematic one. When we construct a reprsentation,
image, idea, concept, etc, the represetntation is just that -- a
construction. The representation is a product of constructive -activity-
and thus is not simply "retrieved." The best treatments that I can think
of that pertain to the issue of storage comes from connectionist and
neural network thinking. The idea would be that what is "stored" is not
a representation, a concept, or even a "construct." Rather, "knowledge" is
distributed throughout a neural network, and is not stored in any one place
. When we construct a representation or construct, we put together on-line
knowledge elements that are distributed throughout a neural network. As
such, we construct representations differently each time they are invoked.
As such, knowledge isn't stored intact. Rather, the building blocks
for creating knowledge is "stored", and we don't simply retrieve, we
reconstruct in action. I would think that this is compatible with Jim's
line of thinking....
Best,
Mike Mascolo
%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%