[From Bruce Abbott (941016.1710 EST)]
Me (931014.1440)
Rick "Mr. Irritating" Marken (941015.1530)
What is missing here is an appreciation of the fact that variable
interval schedules typically generate response rates far in excess of
those required to collect all reinforcers as they become available.
Schedules (the environment) _generate_ response rates (behavior)??
I thought behavior was the control of perceptual representations of
the environment?
It would not be incorrect even in PCT terms to say that a disturbance to the
controlled variable (or some surrogate of it) GENERATES a response (output
that counters the effects of the disturbance). Given that I wish to stay
comfortable (reference), a cold draft in my office may cause me to put on a
sweater. Given that the pigeon is hungry (error signal resulting from food
deprivation), and keypecking is the only available way to correct it, a given
schedule will generate a given rate of keypecking (output). There is a causal
chain from disturbance to the system's response, whether you like to think of
it that way or not, even in servomechanisms!
They represent empirical
relationships that an actual THEORY of behavior must explain. (Rick
Marken, are you listening?)
Yep. Afraid so. And I'm the one who picks the empirical relationships
that I "must" explain, pardner;-).
Fine. I don't believe I called you a theory of behavior. Since you are not
one, you are free to try to explain whatever you like.
The actual problem illuminated by all that research conducted to
investigate the limits of the matching law is to identify what
perceptual variables the subjects of these experiments are attempting
to control.
Well, once again I'm sure glad to see that all this PCT research has been
and is being done. And here I am, complaining that no one is doing
PCT research. Makes me feel like an old kvetch. How could I have
missed all this stuff? Maybe this research has happened since I stopped
being a "real" psychologist. Perhaps you could explain, Bruce, how all
this research is aimed at identifying controlled variables?
This exchange illustrates a general pattern to your responses. You read into
a statement something other than what was actually said and then respond to
this misinterpretation. Perhaps I have not been as clear as I would like to
believe, but like Martin Taylor, I just can't understand how you can construe
my words the way you do. I did NOT say that these researchers of the matching
law were trying to identify perceptual variables as they are understood within
PCT. I was suggesting that this research has suggested possible perceptual
variables that PCT research should investigate from its own perspective.
I know you like to raise this complaint about the lack of true PCT research
whenever you can, but please, don't look for excuses to bring it up. I've
heard you, believe me. Substantive replies to the assertions actually made,
issues discussed, and questions raised would hold my interest far better, and
I invite you to make them. As you did here:
However, there are more interesting questions here. Why, for
example, when confronted with a concurrent VI-FI schedule, do
pigeons stubbornly respond at higher rates on the VI key when such
behavior leads to a higher ratio of responses to reinforcement than
could otherwise be attained?
I think Bill P. already answered that in the "Analysis of a rat experiment"
paper. It probably happens because there are temporal variations in the
rate at which an organism responds.
The fact that maximizing rate of reinforcement on VI schedules requires excess
responding because of temporal variations in response rates is -- and was --
nothing new. However, it would not appear to answer the question why there is
this strong bias. Although I presented the matching law in the context of
concurrent schedules, the VI-FI problem appears in its clearest form in
concurrent CHAIN schedules. In these schedules, identical concurrent VI-VI
schedules (the "initial links") set up access to a second schedule (a
"terminal link") rather than to grain. One key leads from VI 1-min to, say,
VI 10-sec and the other from VI 1-min to, say, FI 10-sec. Although responding
equally on the keys during the initial links leads to equal rates of
reinforcement in the terminal links, pigeons consistently respond at higher
rates on the VI 1-min schedule leading to the VI terminal link. This then
leads directly to:
What is the perceptual variable being controlled here?
Perhaps the perceptual variable being controlled here IS rate of
reinforcement, as you suggest, but then the question becomes why do FI
schedules lead to a different perception of reinforcement rate than VI
schedule schedules do? How is the pigeon integrating these events over time
to create this perceptual variable? And, of course, what PCT experiments can
I devise that would supply the answer?
Any suggestions? (:->]
Bruce
p.s. Were you aware of the e. coli "tumble-and-swim" computer simulation in
Staddon & Ettinger's learning text? The graphs look almost identical to
yours, although they used somewhat different rules.