Fred's Home Heating System, operants

[From Chris Cherpas (2000.11.26.2030 PT)]

Bill Powers (2000.11.26.01255 MST)--

I seem to recall that an operant is not any particular act, but only the
class of all acts that can produce the same observable change in the
environment. Thus any manner of depressing a lever from using a paw-press
to sitting on the lever would be called the same operant, as long as the
lever goes down. Lighting a fire in a fireplace on a cold day and closing a
window would be the same operant in that they both can raise the
temperature of the room. And changing from one injector to another would
not change the operant, since they all are supposed to produce heat
flashes. Is this correct?

Correct. Note that subclasses (sub-operants?) could be defined by
various differences in context.

Of course, when you say "observable change in the environment," I assume
you mean from the experimenter's perspective. But a change observed by
the experimenter has only a _putative_ role in the subject's operants
-- at least that's the "radical" behaviorist rap. Then, again, what we
really find is inconsistency: Skinner was "objective" when he wanted
to claim the advantage of being non-spooky and scientific, but was
"subjective" when he wanted to distinguish himself from previous
("methodological") behaviorisms and to talk about private events.

Mary Powers (2000.11.26)--

...But the recognition that living systems are control systems has
now been in the world for something like sixty years. Isn't it time
for EABers (and any other behavioral, social or life scientists)
to deal with that?

Yes it is. I was trained as an EABer and find the control system
perspective a huge improvement. I'm cautiously optimistic about
others.

Best regards,
cc

[From Bruce Abbott (2000.11.27.1200 EST)]

Chris Cherpas (2000.11.26.2030 PT) --

Bill Powers (2000.11.26.01255 MST)

I seem to recall that an operant is not any particular act, but only the
class of all acts that can produce the same observable change in the
environment. Thus any manner of depressing a lever from using a paw-press
to sitting on the lever would be called the same operant, as long as the
lever goes down. Lighting a fire in a fireplace on a cold day and closing a
window would be the same operant in that they both can raise the
temperature of the room. And changing from one injector to another would
not change the operant, since they all are supposed to produce heat
flashes. Is this correct?

Correct. Note that subclasses (sub-operants?) could be defined by
various differences in context.

For purposes of determining how some particular bit of behavior comes to
occur more frequently than others under given conditions, one can define
some arbitrary bit of behavior according to some observable consequence of
that behavior -- such as raising the temperature of a room. But if your
interest is in assessing the possible effect of a particular consequence on
the bit of behavior that produced it, then it is best not to define the bit
of behavior in terms of the consequence whose possible effect you are
investigating. If you think that the behavior is being engaged in because,
in the past, such behavior has resulted in the warming of the room, you want
to define the behavior in terms of those actions that in the past have
produced this consequence, without including the consequence itself. In the
rat studies we may define a lever-press as an operant -- it has the easily
recorded consequence of closing a switch, and all actions that have this
same consequence (different ways of pressing the lever) are functionally
equivalent. If we defined the operant as "producing a food pellet," then
our analysis runs into problems, because as soon as we stop following
lever-presses with food pellets, we no longer have the operant as we defined it!

So if for the purpose of analysis we wish to separate the act from some
consequence of the act, then that consequence cannot be included in the
definition of the act.

We can see this situation expressed in the sort of experiment with which you
are so very familiar, Chris, namely concurrent schedules. Pecking the left
key and pecking the right key are treated analytically as different
operants, even though either behavior may eventually produce a brief access
to the grain magazine. The same is true in my furnace example, because the
consequence of commanding injector 1 to fire may not be the same as that of
commanding injector 2 to fire. Injector 1 may be clogged, and injector 2
not, so one fails to produce a flame, whereas the other succeeds. If we are
trying to understand why a particular injector continues to be operated
rather than some other, and we think it has to do with a certain normal
consequence of injector operation (a burst of heat follows), then we want to
define our operant (the behavior we are studying) in terms of that
particular injector.

Bruce A.