Subjective Testing for Controlled Variables (was g(d) = - h (o))

[From Rick Marken (981030.0810)]


Subjective experiments. Just look around. I find that, for _me_,
some perceptions (like the perception of a Republican) are associated
with unpleasant emotions while others (like the perception of the
Mozart C major Piano Quartet) are associated with pleasant emotions

Bruce Gregory (981030.0615 EDT)]

To prevent someone from interpreting these examples in S-R terms,

They are _subjective_ experiments. The only one interpreting them
is _you_ (the subject). So you will just have to be trusted not to
interpret the results in S-R terms;-)

it would be necessary to show that these emotions (or lack
thereof) are associated with efforts on your part to control.

Not true (for me, anyway). I experience these emotions whether I
am controlling the perception or not. I don't control whether or
not I meet people who espouse what Republicans humorously refer
to as their "principles" ("greed" and "hatred" and "hypocrisy"
are principles?) but I feel unpleasant emotions when I meet such
people anyway.

This. I presume, would require the application of the test.

Yes. When you are dealing with others you have to test to
determine what variables they are controlling. But I'm not
sure that it's even feasible to try to apply the test to oneself.
Remember, the test requires (among other things) the application
of a disturbance to a hypothetical controlled variable while
monitoring to see whether the hypothetical controlled variable is
influenced (as expected) by the disturbance. I don't see how a
person can apply this methodology to themselves.

The test for controlled variables is really aimed at determining
what variables others are controlling. I think a person can (if
they pay attention) get a pretty good idea of the percpetions they,
themselves, are controlling. In this case, the method of doing this
is more like introspection than testing for controlled variables.




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