[From: Bruce Nevin (Fri 92048 08:04:11)]
(Rick Marken (920506 17:30) ) --
I have read and enjoyed your blind men paper. I have a few comments on
substance and presentation. I have quoted sections of your text where I
have some ideas for changes. My suggestions are almost entirely in
lines not marked with > on the left.
···
-=+=-=+=-=+=-=+=-=+=-=+=-=+=-=+=-=+=-=+=-=+=-=+=-=+=-=+=-=+=-=+=-=+=-=+=-
The existence and ubiquity of control is plain to me but more is needed
to get it across to a non-PCT audience. The example that is given (head
turning wrt seeing) is weakened by the phrase "at least sometimes,"
which to an audience who haven't yet grasped control might seem to
contradict the claim of universality. So here's the first suggestion:
-=+=-=+=-=+=-=+=-=+=-=+=-=+=-=+=-=+=-=+=-=+=-=+=-=+=-=+=-=+=-=+=-=+=-=+=-
It is hard to
imagine an organism that does not exist in such a closed-
loop situation because all organisms are built in such a way
that what they do affects what they sense. Eyes, for
example, are located on heads that move so that what the
eyes see depends on what the head does. To the extent that
what the head does depends on what the eyes see
(for example,
when the head turns in response
to an attractive passer-by) there is a closed loop; sensory
input causes responding (head movement) which affects the
cause of responding (sensory input).
(Closed feedback loops concurrently exist for other actions affecting
seeing, such as rotating the eyeballs, closing the eyelids, and dilating
the pupils.) The fact that behavior is stable shows that the
feedback in this
loop must be negative.
Organisms do not
normally exhibit the "run away" behavior that characterizes
positive feedback loops (such as the "feedback" from a
microphone that amplifies its own output).
[Can you provide an example of the rare type of organism behavior
resulting from runaway feedback, for clarity?]
-=+=-=+=-=+=-=+=-=+=-=+=-=+=-=+=-=+=-=+=-=+=-=+=-=+=-=+=-=+=-=+=-=+=-=+=-
The equations would bear a clearer relationship to text if they stand
off as self-contained entities, rather than being embedded in the
punctuation of the text. Perhaps the convention in the journals you
have in mind is to put equation numbers in the right margin; I find it
clearer on the left. Also, a summary of the meanings of terms used in
the series of equations would be helpful to the non-initiate, or at
least to me. Thus:
-=+=-=+=-=+=-=+=-=+=-=+=-=+=-=+=-=+=-=+=-=+=-=+=-=+=-=+=-=+=-=+=-=+=-=+=-
The fact that organisms exist in a closed negative
feedback loop means that two simultaneous equations are
needed to describe their relationship to the environment.
These are given as equation (1) and equation (2), below.
Equation (1)
describes the effect of sensory input on responding.
(1) r = k.o (s*-s)
For
simplicity we will assume that all functions are linear and
that all variables are measured in the same units.
Here, r is the response variable and s is the sensory input
variable (expressed as deviation from s* which is the value
of the input that produces no response -- or no change in
response -- from the organism). The multiplier, k.o, is the
linear organism function that transforms sensory input into
responding.
It
transforms a small amount of
sensory energy into a huge amount of response energy (such
as when patterns of light on the retina are transformed into
the forces that move the head). In control engineering, k.o
is called the "system amplification factor"
or "gain"
and it can be quite
a large number.
The second equation, too often ignored by
psychologists, describes the effect of responding on sensory
input. For simplicity it is assumed that responding,
r,
adds to
the effect of the environment,
d, as in equation (2):
(2) s = k.f (r)+ k.e (d)
The variables
r and d
have independent (additive)
effects on the sensory input, s. The
effect of the environmental variable, d,
on sensory input,
s,
is determined by the
environmental function, k.e.
The
feedback effect that
the organism's responding has
on the sensory cause of that responding is
represented in equation (2)
by the feedback function, k.f.
The terms in these equations are summarized as follows for reference
in the discussion that follows:
s = sensory input
s* = reference value such that s = s* produces no response
r = response
d = environmental variable
k.o = organism multiplier whereby the organism transforms a
small s into a much larger r
k.e = environmental multiplier making d commensurate with s
k.f = feedback multiplier making r commensurate with s
Equation (1) and equation (2)
must be solved as a
simultaneous pair in order to determine the relationship
between stimulus and response variables in the closed loop.
The result is
equation (3):
(3) r = k.o/(1+k.o k.f) s* - (k.o k.e)/(1+k.o k.f) d
<Text moved from here to a point after equation (1)>
With sufficient amplification (such that
k.o >> k.f and k.o >> 1) equation (3) simplifies to
equation (4):
(4) r = s* - (k.e/k.f) d
Equation 4 is an input/output equation. It
describes the relationship between
environmental (stimulus) and response variables when
an
organism is in a closed-loop, negative feedback situation
with respect to its environment. The result of being in such
-=+=-=+=-=+=-=+=-=+=-=+=-=+=-=+=-=+=-=+=-=+=-=+=-=+=-=+=-=+=-=+=-=+=-=+=-
I think that around this point it would be helpful to describe modelling
and refer to models that are available and that anyone can examine that
approximate the observed behavioral outputs of organisms with 95-99%
accuracy, contrasting this with the track record of the three views
presented in the next section. This would be an effective place to
motivate the equations, saying that they are the conceptual core of
the programs used in this modelling. Might not hurt to mention that
these programs are essentially surprisingly simple, with
most of the complexity in modelling the physics of the environment
yielding d (and maybe k.e) -- or do I misunderstand some prior posts of
Bill's?
-=+=-=+=-=+=-=+=-=+=-=+=-=+=-=+=-=+=-=+=-=+=-=+=-=+=-=+=-=+=-=+=-=+=-=+=-
would be hard to imagine why someone would even try to
make such a measurement unless he or she knew that the
organism was controlling its sensory input. In fact, just the
opposite is the typical assumption.
Researchers typically assume
that the organism is
controlled by its sensory input.
It can be surprisingly difficult to grasp this simple reversal of
perspective: the organism is doing whatever it takes to maintain each
controlled sensory input s at its respective reference value s* -- it is
controlling its sensory inputs.
-=+=-=+=-=+=-=+=-=+=-=+=-=+=-=+=-=+=-=+=-=+=-=+=-=+=-=+=-=+=-=+=-=+=-=+=-
equation (4) can be used to show what control might look
like if,
like most researchers,
one did not know that
control was in fact what
was occurring. It turns out
-=+=-=+=-=+=-=+=-=+=-=+=-=+=-=+=-=+=-=+=-=+=-=+=-=+=-=+=-=+=-=+=-=+=-=+=-
Aside from similar dis-embeddings of subsequent equations, those are all
the changes I come up with. I think it's a very good paper. I can't
evaluate prospective publications.
Be well,
Bruce
bn@bbn.com