Re: Behaviorism vs. evolutionary psychology

On Jul 14, 3:01 am, Don Geddis <d...@xxxxxxxxxx> wrote:
JGCASEY <jgkjca...@xxxxxxxxxxxx> wrote on Thu, 12 Jul 2007:


DG:
Can you try to describe again what you think remains
unknown about the process?

JC:

There are a lot of unknowns. The only reason we know
evolution was possible is because we have observed it.
You cannot deduce it from what you know about the
molecules.

Science may know a lot about molecules and the atoms that
make them up but not enough to see any incremental pathway
even to the first self replicating system of molecules.

Imagine you were given the rules of Conway's cellular
world could you derive from those what patterns would
result when they were implemented? Of course we can
run the rules on a computer and observe what happens.
In the case of atoms we have real entities to play with
so we don't have to simulate them in a computer although
on a small scale they are doing just that.

The posts were to Curt who is interested in something
similar which is the incremental pathways from simple
random behaviors which he imagines can be shaped via
reinforcement feedback from a "critic" into intelligent
behaviors. We know this is possible with evolved brains
with unknown inherent learning mechanisms and some
innate unconditioned behaviors. It is unknown exactly
what must be built in for learning to take place.

Curt figures nothing has to built in if his network can
learn everything, sort of like evolution evolved the
current brains of new borns. I think Jeff Hawkins of
Numenta fame hopes for the same possibility. Unlike Curt
Hawkins had managed to make a demo that does something
interesting,

Hawkins Picture demo:

http://www.numenta.com/about-numenta/technology/pictures-demo.php

So to sum up, the mystery is in the unknown. For me it is
not sufficient to say it is "nothing but" as an explanation
without spelling out in what way it is "nothing but".

If you can precisely define what you are talking about, that
is, no unknowns, you can simulate it on a computer. You can
try and model a process that has unlimited potential pathways
that lead to complex survival machines via incremental steps.

My suggestion however is that there may not be unlimited
potential pathways. [Life's other secret by Ian Stewart]
Given a 100 earth-like planets on which life evolved you
would see some common forms appear again and again due to
the constraints present in the real world. An example of
this on our world may be examples of convergent evolution.
Same problem, similar solutions.

--
JC


.

Relevant Pages

  • Re: So whats null then if its not nothing?
    ... >> The problem is that evolution in the world external to the academic ... We deal with RDBMS vendor software. ... opinion in the end, so make peace with that fact, and ... >> Some people believe that unknown and unknowable values ...
    (comp.databases.theory)
  • Re: How many times must we see the same drivel?
    ... > authors I cite below are not religious advocates trying to make a point; ... except for Behe and Denton. ... > would have suppressed evolution not only in some primordial soup, ... > be named or described at some unknown time in the past did something unknown ...
    (talk.origins)
  • Re: How many times must we see the same drivel?
    ... authors I cite below are not religious advocates trying to make a point; ... If evolution requires gradual change, how could the cell contain several ... Science proceeds by narrowing the questions down a bit to ... be named or described at some unknown time in the past did something unknown ...
    (talk.origins)
  • Re: Genomic data - common ancestry or common designer?
    ... evolution in this statement) cannot make this statement and remain ... The main problem with the `common designer' argument is that it matches ... Sunday by genetic evidence contradicting the ancestry shown in the fossil ... It also requires an unknown Designer that designs things for unknown reasons ...
    (talk.origins)
  • Re: Chimpanzee an Man LCA, a banana?
    ... and whether infant fat offers any protection ... requires that evolution create a feature ... needed' feature -- one for a might-sometimes- ... unknown in nature -- and nearly all those ...
    (sci.anthropology.paleo)