[Martin Taylor 2017.01.01.15.32]
Happy New Year to all.
[From Rick Marken (2016.12.31.1420)]
…
RM: But maybe next year (like tomorrow;-)you could answer your exercise on collectively controlled
stability – the one that I was unable to answer.
I think I will wait until January 20, to give you a little more time
to consider the possibilities. You do have all the background
information you need, and you have been linked to a much greater
store of information than you actually need. If nothing else helps,
meditate on the gedanken experiment, substituting a wide variety of
things both tangible and intangible for the Puck and pebbles.
But this message is on a different matter that is of some PCT
interest – the technique of creating a pissing match. That is
something I never considered to be an actual reference value for a
controlled perception until Bruce said he was interested in the
gedanken experiment and its result, but not in a pissing match about
it.
Unfortunately, that ought to be a joke, but probably is just a
statement of fact despite the emoticon. Let’s follow a couple of the
techniques by way of the gedanken experiment example.
Technique A: Dispute the premise, misunderstanding what it was to
show, and present an alternative result that was actually the result
originally of interest. When I agree with that result, assert that
it’s not PCT, and is therefore wrong or irrelevant.
Example:
A1. I propose an experiment that seems to suggest that one
cannot tell by testing the control of a single variable whether the
controller is controlling a single perceptual value at one reference
value or is the result of many individual controllers all
controlling their perceptions to their differing reference values. I
suggest that the hypothetical experimenter “Dee”, being a good PCT
student, might look at her results and realize that she could not
know whether there is a unitary controller directing the individual
actors.
A2. Rick says that for Dee to hypothesise the existence of a unitary
controller was “cockamamie”, and that he would not have hypothesized
a unitary actor and would have considered only the performance of
the distributed players.
A3. I say that Rick's acceptance of the possibility of distributed
controllers was all I needed to make the point of the experiment,
which was that when you do the TCV, you are not necessarily
determining a single perception that is controlled, but might be a
virtual perceptual signal that is explained as an averaging of
signals in the individual control elements. I refrained from
commenting on whatever else Rick said, since given his agreement,
everything else was moot.
A4. Rick contradicts himself. He says he has no idea what a virtual
perceptual signal is, and that PCT “explains the existence of
observed reference states for controlled variables in terms of the
existence of neural reference signals that specify the state of
perceptual signals that correspond to variations in those controlled
variable.” I read this as saying that PCT is not compatible with the
point of the experiment, to show that what is apparently controlled
could be the result of a multitude of neural pathways, without the
necessary existence of any place in the brain that holds a unitary
perceptual value that is controlled to match a unitary reference
value. Rick’s answer seems a direct contradiction to Rick’s earlier
statement that for Dee to admit a hypothesised unitary controlled
perception is a “cockamamie” idea.
Technique B. Bring up as many red herrings as possible, to distract
from the original point, and keep it going by contradicting yourself
without acknowledging your change of thought.
Example:
B1. In describing the experiment initially, I say that Dee performs
the TCV and discovers that the position of the Puck is controlled
rather than being the result of a bowl-like configuration of the
table on which the Puck slides.
B2. Rick says Dee cannot deduce anything useful because she has done
the TCV incorrectly, and says that if I or someone else does not
explain why her procedure was incorrect, he will post an explanation
on Xmas Day. He says he is “infuriated” by the incorrect use of the
TCV.
B3. I cannot see any problem with the way Dee did the TCV, but I
hazard a couple of rather out-of-left-field guesses as to what he
might say on Xmas Day, and list what I think are the requisite tests
involved in the TCV. There are four, the last of which has three
independent variants (4a, 4b, and 4c), of which Dee used 4b. Rick
offers no explanation on Xmas day, but offers one a couple of days
later.
B4. Rick asserts that only variant 4a is acceptable and Dee did not
do it. He also claims that 4a and 4c are effectively the same,
despite their rather dramatic mathematical differences (which
corresponds to a pissing-match technique not used in this example
but has been used to fuel other pissing contests: “I don’t care
about math; I care about phenomena.”).
B5. I ask how Rick would apply variant 4a in the classic
“car-in-lane” example of control, given that the obstacles to doing
so are exactly those that prevent Dee from using it. (And I ask why
4b and 4c are unacceptable, given that he uses them in his own
demos).
B6. Rick says 4b and 4c are indeed sometimes acceptable, but does
not explain why they are not in this case. In respect of the
car-in-lane question, he says “Simple” and proceeds to describe just
what Dee did in the original experimental description, which is not
variant 4a at all, but a test of criterion 2. Dee’s “infuriating”
incorrect performance of the TCV turns out to be how he would
actually perform the TCV in an analogous situation.
···
Martin Taylor (2016.12.31.10.31)
MT: No, Rick, I won’t get drawn into your pissing match.
RM: Well, I tried;-)