Incentive

[From Rick Marken (970613.0900 PDT)]

Me:

PCT shows that there is no such thing as an "incentive"...food is
a disturbance to the perception of hunger.

Tim Carey (970613.2040)

Could you explain this paragraph a little more in terms of
feedback functions... my understanding was that if a person is
using food to bring their perception of hunger to zero then
food would be part of their feedback function.

Let me try to diagram what I was thinking.

                      r
                      >
                p---->C-----
                > >
                ^ v
                i o
     My Food-->fis<--Food<--

I am assuming here that the controlled variable is "food in
stomach" (fis). The perception of this variable (p) could be
called "fullness"; this perception is kept at the reference for
fullness, r, by actions (o) of the organism. These actions are actually
the operation of many lower level systems controlling perceptions of the
sight of food, taste of food, etc. The result of these actions is to
influence environmental events (food) so that
food gets into the stomach.

So "Food" (in the diagram) is part of the environmental feedback
connection between the actions of the organism (o) and the
controlled input variable, fis. "Food" can be thought of as a
funtion that determines the effect of actions (o) on the controlled
variable (fis). If there is a lot of easily accesible food around,
then the Food function is "strong"; a little action produces a large
increase in fis. If there is a drought, then even a lot of action
may have little effect on fis

I have also added a disturbance to the diagram that I call "My Food".
This is food that I can make available to the organism "independent"
of its own food getting actions. This is the food I can use as an
apparent "incentive" to increase (and decrease) behavior. For
example, I can use My Food as a negative incentive, to decrease the food
getting actions (o) of the organism, by _increasing_ the food I give to
the organism. I can also use My Food as a positive incentive to increase
the same actions by taking away my food. This is
basically what Bill Powers and Bruce Abbott did in their study of
food input control in rats; when rats had plenty of food in their
home cage (My Food was high) they would not work very hard to supply
themselves with food in the experimental cage; when the rats were
deprived of food in their home cage, the rats worked much harder to feed
themselves in the experimental cage.

So availability of food acted like an incentive to feeding behavior;
high availability was a low incentive; low availability was a high
incentive. This "incentive" effect is, of course, illusory; what
we are seeing is simply the relationship between the simultaneous
effects of a disturbance (My Food) and an output (experimental cage
feeding, o) on a controlled variable (fis).

It is also possible to see such illusory "incentive" effects if
you can control the function ("Food" in the diagram) that relates output
(o) to the controlled variable (fis) . This is probably what Tom Bourbon
was thinking of when he said that food is part of the feedback function
because this is the typical way the "incentive" illusion is created in
operant conditioning experiments. All you
have to do is control how much effect actions can have on food delivery
to the stomach. In operant conditioning experiments this
is done by varying the number of times the animal has to "press" or
"peck" to get one unit of food. On very demanding schedules, food
seems to have a very high incentive value because the animal presses or
pecks like mad to get it; on non-demanding schedules the food
seems to have less incentive; the animals work less hard for it.
Of course, the food has no incentive value in either case; it's
just food. The appearance of "incentive" occurs becuase the animal
is doing what it has to do to keep its perception of fis under control.

The concept of "incentive" is similar to that of "stimulus"; the
idea is that something outside of the organism is _causing_ its
behavior. Unfortunately, because PCT explains why organisms
_appear_ to respond to stimuli and why they _appear_ to get
motivatied by incentives, some people have mistakenly concluded
that PCT shows that organisms _do_ respond to stimuli and _do_
get motivated by incentives. This is an unfoirtunate misunderstanding of
PCT. In PCT, it's all control of perception.

PS. I will bring an extra copy of _Mind Readings_ to the meeting
but it would sure be nice if people could actually order copies
from New View. I think the book is now listed on Amazon (the web
book seller) so maybe you could get a copy through them?

Best

Rick

···

--

Richard S. Marken Phone or Fax: 310 474-0313
Life Learning Associates e-mail: rmarken@earthlink.net

Tim Carey (970619.1915)

Hi all,

Rick Marken mentioned the following study in a recent post ...

basically what Bill Powers and Bruce Abbott did in their study of
food input control in rats; when rats had plenty of food in their
home cage (My Food was high) they would not work very hard to supply
themselves with food in the experimental cage; when the rats were
deprived of food in their home cage, the rats worked much harder to feed
themselves in the experimental cage.

I'm interested in learning some more about this particular study. Is it
available anywhere to read? How might I be able to get information about
it?

Regards,

Tim

[From Bill Powers (970619.1056 MDT)]

Tim Carey (970619.1915) --

Rick Marken mentioned the following study in a recent post ...

I'm interested in learning some more about this particular study. Is it
available anywhere to read? How might I be able to get information about
it?

Bruce Abbott is working on a paper about the experiment, which was competed
only in mid-winter. I'll have the data and some programs for looking at
them at the meeting.

Best,

Bill P.

[From Rick Marken (970620.0840 PDT)]

Bill Powers (970619.1056 MDT)--

Bruce Abbott is working on a paper about the experiment, which
was comp[l]eted only in mid-winter.

That's great!

I'll have the data and some programs for looking at them at
the meeting.

That's great, too. But I sure wish Bruce could make it to the
meeting to present this himself. Ah well. Maybe next year.

Best

Rick

···

--
Richard S. Marken Phone or Fax: 310 474-0313
Life Learning Associates e-mail: rmarken@earthlink.net
http://home.earthlink.net/~rmarken/