My question (930125) was: "Is offering sympathy "Positive Feedback?"
Bill's comment was (930126): "Sure it is. The more you get, the more you
want."
Bruce also responded with, "Is an answering groan from another part of the
avalanche positive feedback?" Thanks again, Bruce.
What are the Controlled Perceptual Variables involved here -- of which you
"want more?" I had several possibilities in mind in asking the question,
and the two responses suggest additional variables -- and now I am
considering whether to follow up on these questions (which tend to lead to
discussion of higher order variables - personality, character, etc), or
turn to related topics.
If you examine your procedure in forming your responses, I think you will
find that you considered several alternatives, applied various criteria
anticipating possible further interactions, selected your response and put
it into effect. Here I say, "you," but this presents your own DME at work.
As remarked from time to time, positive feedback tends to result in
divergence or oscillation of the relevant variable. For lower order
systems, this is generally intolerable. Hence the importance of stability
in electro-mechanical systems and Hierarchical Feedback Systems. This must
be considered when "anticipation" or "feed-forward" elements are added. It
can also be true for a two order system if the operation of the lower order
system is not complete (or very nearly so) before the higher order system
acts. This is discussed in BCP, p 52 ff.
For illustration, our early reaction time experiments are described, ibid p
245-6. In addition, the two stage reaction of the "Portable Demonstrator"
is included. However, I once had the experience of trying this demo and
found the subject moved his hand together with my "push signal." Although I
made my signal as sharp as I could, his hand still began an immediate
downward movement, following mine. His response could be described as
"highly damped." The demo worked after I revised my instructions to
emphasize that he was to keep his hand firmly in place until he received
the signal. This experience further illustrates the critical importance of
the details of the instructions. With our reaction time measurements we
found it important to examine the instructions.
After all, the experimenter is part of the experiment!
With a somewhat different emphasis, the participation of the experimenter
as part of the feedback loop was discussed in "Integrated Data Collecting
and Processing Systems in Psychophysiology," R K Clark, R L McFarland, M
Bassan, Annals of The New York Academy of Sciences, Vol 115, July, 1964.
Such mixed timing of levels of control was the theoretical basis of the
approach McFarland and I took in our work with adult stutterers. A speaker
divides his attention among several variables. He monitors his speech for
its general characteristics (speed, tempo, loudness, etc), he plans his
next statements, and he selects ideas for presentation. Noticing a flaw
the exact accuracy of forming a phoneme can be very disruptive when the
error is noticed at the wrong time. The lower order actions, even if
flawed, must be completed before higher order systems begin their
operation. The non-stutterer pretty much lets his grammar, pronunciation,
etc take care of themselves. In early stages of learning to speak,
problems of timing, etc are usually worked out. Remembered procedures
together with their effects are used to establish patterns that are quickly
available when needed. Of course, at each decision point the DME is at
work.
At a meeting of speech therapists where I had outlined these ideas I was
asked whether I had once been a stutterer (in fact, I never was). The
questioner was a "reformed stutterer" who thought my understanding of the
stuttering might have resulted from personal experience.
Combining memory with the operation of lower order systems can be
illustrated by extending Bill's music box to the dimensions of a player
piano, I'm thinking of the old pneumatic version. This device may have
controlled air pressure and speed (controllable from the bench), but needed
no other controls. If a new selection was desired, there was no need for a
new complex computational system to operate the individual keys -- a new
music roll was inserted. But higher orders selected the roll.
Similarly, concerning the trill played on the piano. When learning the
instrument, much time is spent building a repertoire of remembered
combinations, sequences, etc etc. Among them is the "trill." When needed,
the remembered reference signals are "plugged in," and the lower order
systems operate accordingly. Consider the much greater complexity of
building the player of a violin!
These are among the skills that people learn, Fifth Order. When using such
instruments is under discussion (Sixth Order, interpersonal communication),
the skills would be considered Fifth Mode of Sixth Order. "Mode" refers to
the topic of discussion. "Mode" provides terminology for analyzing
communication in terms of levels of control system. "Modes" do not control
communication, they are tools used in the process of communication.
Enough for now.
Regards,
Bob Clark