Too many Karoly's, Info theory

[From Rick Marken (930314.1800)]

I made the mistake of going to the library and making a copy
of Karoly's Annual Review of Psychology (1993) paper on self-
regulatory mechanisms. Why didn't I listen to Greg? I could have
just gone out and bought an emetic. Here is a representative
piece of wisdom:

"Although from a purely engineering (cybernetic) perspective, the
standard of correctness is 'physically embodied as a perfectly real
reference signal inside the control system' (Powers, 1986), in human
self-regulators the rule generation and rule following routines
are variable and subject to moderating influences."

What in the world does this mean??? That reference signals are variable
in humans? (If so, that's part of PCT and nothing in the Powers quote
imples anything different). That program level references (rule generation
"routines") are variable" (Still no problem). That the outputs resulting
from comparison of perception and reference are variable and opposed
to disturbance (moderating influences)? Again no problem. Throughout
this article Karoly implies that Powers' approach to "self regulation"
only applies to machines. I think the people working in this area
(and it now apprently includes a LOT more than Carver and Scheier) are
so far gone that learning PCT is not even a possibility. How depressing.

Bill Powers (930314.0900) --

WOW! What a wonderful antidote to the Karoly paper.

So how DOES information get into the organism from the
environment? Clearly, only through uncontrolled perceptions.

Brilliant; could this be one sideways contribution of IT to PCT?

Uncontrolled perceptions may be uncontrolled at many successive
levels, but eventually they join with controlled perceptions to
become inputs to a high-level control system. There they act as
disturbances of the perception to which they contribute, _and
their information content is therefore rejected by that system_.

This is a fascinating idea; the only "informational" perceptions
are uncontrolled perceptions -- which will tend to exist AS INFORMATION
at the levels of perception at which they are uncontrolled. I wonder
what this means phenomenologically?

Your post makes me realize that I should say that only CONTROLLED
sensory input variables are dependent variables; uncontrolled
sensory variables (like the moving target in a pursuit tracking
task) are still not independent variables; maybe a good name for
them is INFORMATIONAl VARIABLES; these are perceptions that are "just
there". A perception (like the moving target in the traking task) can
be an informational variable at one level (like the transition level;
you just see a moving line) and a disturbance (and, thus, non-
informational) at another -- like the relationship level where you are
trying to control the difference relationship (target-cursor) At this
level the information about the disturbance is cancelled out by the
control process.

Your discussion reminds me that I am going to have to put some un-
controlled perceptions in the spreadsheet model; right now the model
gives the impression that ALL sensory inputs are part of control
processes at all levels; that is obviously NOT very realistic.
Off to Excel.



PS. The "Bill" in my last signiture was Bill Clinton (I guess I
just like Bills, unless they're from Buffalo); if you don't know
who the "jerks" are then you are an incurable Republican and it's
probably not worth providing information at that level.