[From Rick Marken (951130.1330)]
Martin Taylor (951129 14:00)--
I have a bone to pick on one issue that tends to recur on CSG-L.
I agree with you completely. You are right to take me to task. I did not
express myself clearly when I said:
For example, you conclude that differential responding to the pictures
reflects the organism's ability to discriminate the pictures; different o's
result from different values of s because the bird can disciminate different
values of s.
By putting it this way I just muddled the issue .
I was TRYING to say that, in a conventional experiment, the stimulus (s)
variable seems to cause the response (o) variable via the organism. That
is, it appears that o = f(s). In the Herrnstein experiment, for example, the
picture variable (s) seems to cause the pecking variable (o) via the pigeon;
different values of picture ("girlfriend" or "not girlfriend") seem to cause
different values if pecking ("pecking" or "not pecking").
But it is likely that s is actually a disturbance to a controlled variable
(q). q is protected from the effects of the disturbance by o. When this
is true, the relationship between s and o reflects characteristics of the
feedback function, not the behaving system. That is, o= 1/g(s). In the
Herrnstein experiment, the pigeon is probably controlling a logical
variable (( girlfriend AND peck) OR (no girlfried AND no peck)), keeping it
"true" and the observed relationship between s and o is probably not much
different than the actual one.
But the important point is this: Regardless of whether the relationship
between o and s is o = f(s) or o = 1/g(s), the organism has to be able to
discriminate s (or the variable affected by s, if s itself is not visible, as
in compensatory tracking tasks). In the case of Herrnstein's girlfriend, it
is clear that the pigeon can discriminate different values of s (girlfriend
from no girlfriend); the bird has to be able to do that even if it is
controlling a logical variable. All Bruce Abbott has been arguing is that
the Herrnstein experiment shows that pigeons can make that discrimination,
and I conceded, last night, that that is the case (Rick Marken
(951129.1845)). So I (reluctantly;-)) agree that conventional research can
tell you some things; it just can't tell you what perceptions are being
controlled -- which some of us think is important to know and others of us
apparently don't.
Bestest
Rick