[From Bruce Abbott (970227.1135 EST)]
Rick Marken (970226.2245) --
Rick Marken (970226.1430 PST):
In the line judgment case the DV (judgment) is not caused by
the line (IV) but by sensory variables influenced by the line --
sensory variables that are themselves influenced by the DV
(judgments) or variables related to the DV.
Bruce Abbott (970226.2005 EST) --
Yes, the judgment is not caused by the line, but by sensory
variables influenced by the line. I fail to see how the judgments
influence the sensory variables, and I doubt that you do either,
but it is required that they do for your argument to hold, so
you simply assert that they do.
Lines do not ordinarily lead to judgments of length. When the image
of a line falls on my retina I don't suddenly say "long" or "short"
or whatever the judgments are in a psychophysical experiment.
I agree.
The sensory variables that lead one to make length judgments are
quite complex; they include other people, the words they speak,
their body movements, the lines shown, contextual features of the
situation (whether it occurs at a college or a police station), etc. The
sensory variables that lead you to make line length judgments
are complex perceptual functions of what is happening at your sensory
organs: perceptions of things like your own cooperativeness, the
goals of the experimenert, the length of the lines relative to other
objects, etc.
Quite true.
Your length judgment doesn't depend on line length any more than
your mouse movements in a compensatory tracking task depend on the
retinal position of the cursor. Like the cursor in the tracking task,
the actual length of the line is only a _part_ of one or more of the
perceptions your are controlling when you make your judgment. So the
judgment you make is not an open-loop response to line length; the
judgment is part of a control loop. What variables are actually being
controlled, I don't know. Surely one variable is the relationship
between the judgments (symbols like numbers or words) and line
length. But, as you know, if this relationship is a controlled
perception then the observed value of this relationship depends on
the subject's reference for this perception.
Also correct. (By the way, how did the judgement turn into one of line
_length_? Oh, well, it doesn't change the argument.) Consider the following:
retinal image of line -->[function relating image to judgement]-->judgement
(I am oversimplifying the function, which would have other inputs in
addition to the retinal image of the line, but this should be sufficient for
this discussion.)
Assume that the function was developed by having the participant make
judgements of lines of various lengths, and after each judgement, telling
the participant the actual line length in millimeters. This places the
judgement in a closed loop whereby the participant can vary her
judgement-function parameters as necessary until the judgements generally
conform rather closely to the actual line length:
+------->[measurement]-->length in mm------------+ r
^ |
> v e
line length-->[judgement function]--->judgement----->------>[JF params]
^ |
i> >
ยทยทยท
+---------------------------------------------+
After the judgement-function parameters have been set, we can just ask our
participant to provide a line-length judgement, open loop (i.e., without
providing feedback). For this purpose she is now functioning essentially as
a talking ruler -- show her a line of 50 mm and she says "50 mm".
The fact that she is providing these judgements shows that she is
controlling something, but it is not the length of the line and it is not
the output of the judgement function. If she is controlling for giving the
experimenter what he has asked for (her honest appraisal of the length of
the line), then her judgements and the measured lengths of the lines will
continue to agree within the limits of her accuracy in making the judgements.
Anyway, I am sorry that I succumbed to this conflict. I certainly can't
browbeat you into doing psychological research the way I
think it should be done
I don't know why you think it's necessary. I'm already _doing_
psychological research the way you think it should be done.
and you can't get me to think much of the
psychological research that is done using standard methodology.
Well, I can see that offering strong disturbances to that controlled opinion
hasn't even begun to trigger any reorganization. But as I stated at the
outset, I didn't hold out much hope for that.
The fact
of the matter is that if you understand PCT then you will do
the research properly; standard behavioral research methods will simply
not make sense any more.
The real fact of the matter is that certain applications of standard
behavioral research methods already don't make sense to me any more. I have
argued that they continue to make good sense for attacking certain kinds of
problems, not that they are appropriate tools for the analysis of functions
whose outputs feed back into their inputs. Alas, you are unable or
unwilling to perceive this distinction.
Regards,
Bruce