Re: Problem for physicalist evolutionists

On Mar 27, 4:29 pm, someone2 <glenn.spig...@xxxxxxxxxxxxxx> wrote:
On 27 Mar, 15:41, leland.mcin...@xxxxxxxxx wrote:



On Mar 27, 9:48 am, someone2 <glenn.spig...@xxxxxxxxxxxxxx> wrote:

On 27 Mar, 12:05, leland.mcin...@xxxxxxxxx wrote:

<snip>

To have an FPP the nodes need to be linked because a set of neurons
does not a neural network make. Make a network and you have the
potential to have an FPP, and the FPP will be some part of that
network. Now, put a picture distributed across 100 nodes in that
network that is the FPP. Lo, they do no need to communicate anything
along any of those network connections to "assemble" the picture. The
picture is already in the FPP -- in the network. So yes, the nodes
have to be connected, but that's so you can actually have a neural
NETWORK and some hope of having a general FPP and sense of "I". Once
you have those connections the 100 nodes don't need to communicate
anything along any of those connections to "communicate the picture"
to the "I", since the "I" already has the picture.

Communication is necessary for the general FPP, the "I", to exist.
Communication is not necessary for that FPP to have an FPP of the
picture. So yes the nodes have to be connected, but no they don't have
to communicate along those connections. The reasons for connections
and (lack of) communication are different.

So in an ANN where the nodes are connected by electrical conductors,
could be generally communicating between the nodes, and have an FPP,
the "I" to exist. Though there could be a section of 100 nodes in the
network which had been fed a pixel each from a 10 x 10 picture. These
nodes wouldn't need to communicate the information they contained for
the "I" of the system to know the 10 x 10 picture, is that what you
are saying?

We're getting closer. Connect the nodes with whatever mechanism you
like, but there needs to be connections (otherwise it isn't much of a
network). The topology of those connections has a great deal to say
about the system, so having connections are important. It's also going
to matter that the system/network is connected in the sense of having
only one connected component in the graph theory sense. That is, for
any two nodes in the network, there is *some* path (potentially over
many different connections) between those nodes. Isolated components
are not "part of the network/system", but are a separate system.

Now, given a system in the sense above, and presuming that it has
whatever it takes in terms of complexity, topology of connections,
etc. to have an "I", can the "I" of that system "know" the picture if
it is scattered across 100 different nodes of the system with no
further communication? Yes, it can. It will need to be the right 100
nodes of course (according to the structure and topology of the
system), and to actually act on the picture some communication will
have to occur, but to "know" the picture takes no further effort. That
is, there is no need to "assemble" the picture into a representation
for the "I", since the 100 nodes each containing a pixel of the
picture *is* a representation of the picture, and it is already
contained in (and therefore part of and presented to) the "I" of the
system. Action based on the picture can occur without having to "bring
the picture together" because it is already "together" within the
portion of the system that is the "I". There is no more "presenting"
to be done.

You certainly sound like a bit of an expert on it. So there are
special topologies where the system will gain a first person
perspective, what it is to be like, the "I". That as long as it has
this "I" as long as the information regarding each pixel of the
picture is stored in the right place, there doesn't have to be any
communication from the node for the "I" of the system to see what the
picture is. Sorry if this seems like a repetition of what you said,
but I wanted to be sure I was understanding all the technical details.

I am woefully far from being an expert. This particular point
(expanding the concept of "I" to allow it to embrace the
representation of something) doesn't seem that hard. Otherwise, in
general, yes, your summary is reasonably accurate. Since I am not an
expert, there remains potential for my summaries of other peoples
ideas to have left out, or failed to fully specify points. If you
actually want details, rather than summaries of general understanding
from non-experts, then you could always try consulting some of the
many texts on this matter that you have been repeatedly pointed toward
by several different people here, including myself.

I am away this weekend, so I'm you won't be getting any responses from
me for a while. Given that others here seem to understand the idea I'm
trying to get across at least as well as I do, you can always look to
them.

.

Relevant Pages

  • Re: Problem for physicalist evolutionists
    ... does not a neural network make. ... potential to have an FPP, and the FPP will be some part of that ... along any of those network connections to "assemble" the picture. ...
    (talk.origins)
  • Re: Problem for physicalist evolutionists
    ... does not a neural network make. ... potential to have an FPP, and the FPP will be some part of that ... along any of those network connections to "assemble" the picture. ...
    (talk.origins)
  • Re: Problem for physicalist evolutionists
    ... does not a neural network make. ... they do no need to communicate anything ... along any of those network connections to "assemble" the picture. ...
    (talk.origins)
  • Re: Problem for physicalist evolutionists
    ... does not a neural network make. ... they do no need to communicate anything ... along any of those network connections to "assemble" the picture. ...
    (talk.origins)
  • Re: Problem for physicalist evolutionists
    ... there are connections to language centers (so I shout "hey, ... That's the physical basis of the FPP. ... sense that it could have a first person perspective of the picture. ... That some form of communication would be required. ...
    (talk.origins)