[From Rick Marken (950502.0830)]
John E. Anderson (950501.2245 EDT) --
not being a psychologist and so not having been steeped in the S-R
controversy, I'm not even sure if I CARE about the S-R controversy.
"S-R" is just my shorthand wany of saying "lineal cause-effect". Almost all
models in the life sciences (other than control models) are, thus, S-R
models from my perspective. Your description of the NSD model suggests that
it is a cause-effect model of how the brain produces behavior.
For example, in your earlier post [John E. Anderson (950429.1530 EDT)] you
said:
initial perceptual input representing the surroundings will be followed
within milliseconds by activity induced in adjacent neurons by new input
from the surroundings. Because the adjacent neurons are also sensitized by
the neuroactive envelope from the initial activity, their output will occur
sooner. The net effect will be that the animal will be able to decide what
to do more quickly
This is a description of behavior as a cause-effect process. Perceptual
inputs cause behavioral outputs ("deciding" is part of this causal process;
control systems don't "decide" what to do). The NSD model makes it possible
for this cause- effect system to produce effects (responses) more rapidly
(sooner).
Whatever the nature of the NSD model, it is a model of an open-loop system.
Prediction and the ability to "decide what to do more quickly" are irrelevant
to the input-output relationships in a closed loop.
The model you describe makes sense only if outputs are in some way separated
from the inputs that generate them. That is, the NSD model is relevant if
organisms work the way life scientists have always assumed they work. For
example, the NSD model is relevant if an organism works this way:
sight of predator --> running movement --> avoidance of predator
A stimulus (sight of a predactor) causes a response (running) that causes an
adaptive result (increased distance from predactor).
This is a cause-effect model of behavior. The best way to find out what's
wrong with it is to read everything Bill Powers ever wrote. I will take the
liberty of summerizing Bill's life work in five sentences: The cause-effect
model does not explain why responses (running movements) reliably produce an
adaptive result. Disturbances (such as the changing location of the
predator) always contribute to the effect of responses. The explanation for
the reliability of the results of responses is that the results themselves
are under control by the organism. Responses are part of a control loop so
they are not caused by inputs. Rather, responses are varied, as necessary,
to produce the controlled result while compensating for unpredictable (and
usually undetectable) disturbances.
PCT suggests that the nervous system acts, not to cause efferent neural
responses but to control afferent (perceptual) neural signals. The nervous
system must do this by comparing perceptual siganls to reference signals
generated by the brain itself; this comparison must be something like a
subtraction that generates an error signal that drives the physical output
that affects the environmental cause of the perceptual signal. A biological
model of control would (I presume) explain the neural basis of perceptual
computation, reference/perception comparison and the efferent (error)-based
cause of physical output. Such a model would be based on an understanding
of the implications of the fact that the NS is part of a causal loop,
NOT a causal chain. That is:
->NS->
> > NOT ->NS->
<-----
Best
Rick