[SPAM] Re: Discouraged

[Bruce_Gregory4]

[From Bruce Gregory (2003.12.10.20.50)]

[Martin Taylor 2003.12.10.1734 EST]

The big perceptual control hierarchy, according to this view of
reorganization, developed because successive refinements of the
criteria for perceptual control improved the precision with which
perceptual control side-effects had the appropriate influence on the
intrinsic variables. One lives longer if one stops when confronted
with a patch of red, if one can identify that patch as belonging to a
traffic light rather than to an autumnal maple tree.

So, if there are many intrinsic variables, presumably survival (or
rather, preferential propagation of the genes) involves many
interactions and interaction loops among them. The complexity of
these interactions requires corresponding complexity in the
perceptual control hierarchy that acts on the environment so as to
maintain the intrinsic variables near their (genetically determined)
reference values.

What, then, does it mean to say "how closely the perception is
coupled to intrinsic variables"? And how would the answer relate to
the case under discussion?

I believe that my view of the hierarchy is essentially the same as
yours. The reference level for the highest levels of the hierarchy are
determined by reorganization. Maintaining these highest levels at their
associated reference levels controls intrinsic variables as side
effects. It was in this sense that I said that the highest levels "are
closely coupled to intrinsic variables. That is, the reference levels
of these perceptions are not established by still higher levels in the
hierarchy. I hope this is a bit clearer now.

In this picture you can 'voluntarily reorganize' in the way you
described _only_ if the perception is _not_ at the highest level. So
discovering that you can effect such a change tells you at once that
you are acting somewhere below the highest level.

For some individuals, religious beliefs seem to be at the highest
levels. The reference level associated with these beliefs cannot change
without having a negative effect on the control of intrinsic variables.
I should say, without _likely_ having a negative effect on intrinsic
variables. Each 'fundamental belief' is a local maximum on a fitness
landscape. Small changes tend to reduce fitness, and so are very
difficult to make.

Other individuals seem to have no such religious beliefs. Or, I should
say, it is not clear what replaces religious beliefs in these
individuals. They must have controlled perceptions at the highest level
of their hierarchies, but it is not obvious what they are.

For example, I have a standard that says that we are all entitled to
our own beliefs, not not to our own evidence. If you want me to believe
something, i want to know the evidence that supports it and the
evidence that would count against it. I hold to this standard because
it is the only way I know to make progress. I can perfectly well
imagine adopting an arbitrary belief such as, "There is no God but
Allah, and Mohammed is His prophet." But I decline to do so because I
find no reason to prefer this particular belief to dozens of other
beliefs such as, "Salvation only comes to those who accept Shiva as
their personal saviour."

I am a pragmatist for pragmatic reasons. I know that the Mormons could
be right, but I know of no way to discover wither this is the case.
What are my highest level perceptions -- the ones I am unwilling to
change because to do so would move me off a local maximum on the
fitness landscape?

Bruce Gregory

"Everything that needs to be said has already been said. But since no
one was listening, everything must be said again."

                                                                                Andre Gide

[Bill_Powers]

[From Bill Powers (2003.12.11.0703 MST)]

Bruce Gregory (2003.12.10.20.50)–

I believe that my view of the
hierarchy is essentially the same as

yours. The reference level for the highest levels of the hierarchy
are

determined by reorganization.

This is a bit tricky. What you say agrees with part of my
proposal, in that one of the possible sources of highest-order reference
levels is reorganization. But I also allow for the possibility of genetic
sources (not for things like religion, of course), experience and
experiment, and “zero”: don’t forget that zero is a point on a
scale of perception and is not the same thing as absence of a perception.
The exterior angle at the elbow may range from zero (as sensed) to some
maximum, with a zero reference signal meaning fully extended.

Also, since reorganization can change physical circuitry, it can change
the amount of contribution of a higher-level system’s output to a
lower-level system’s net reference signal. Thus with a constant
higher-level output, reorganization can change the lower reference
signal.

But the main trickiness is that reorganization can work anywhere in the
hierarchy, so it’s not like a higher-level system setting reference
levels for the next lower one. That would imply a systematic relationship
between intrinsic error and the highest-order reference signals, but as
Martin’s post made clear, there is no such systematic relationship in
this theory.

Each ‘fundamental belief’ is
a local maximum on a fitness

landscape. Small changes tend to reduce fitness, and so are very

difficult to make.

I don’t think that “fitness” is the best term for something
that pertains to a single lifetime, since the only test of fitness (in
evolutionary theory) is survival and reproduction. A local minimum of
intrinsic error would seem a better term for the same phenomenon,
especially since intrinsic error is suppose to be an indication of how
well the life-support systems are functioning, or something like that.
The relationship to evolutionary fitness would come through the settings
of intrinsic reference levels, which must change slowly over evolutionary
time, and must be partly determined by natural selection.

"Everything that needs to be
said has already been said. But since no

one was listening, everything must be said
again."

I noticed that in one of yesterday’s posts, this epigram was attached
twice. Was that a recursive illustration of the truth of the
statement?

···

Andre Gide