On naturalism and consciousness

[The Grand Nudger]

I call them arbitrary because you've chosen to identify particular structures as having mind, and others as not, based on your own ideas, but not necessarily on observable facts about the nature of consciousness. To be fair, I think your particularly arbitrary structure of interest is a very good candidate and one worth considering. But I think mine is, too.

The only structure that I know of that has "mind" as we seem to be using the term are human beings (and more specifically, human brains). So, I'm not identifying structures that have "mind", I'm trying to identify structures which could account for what we're talking about -when we say- "mind". I'm addressing -why- our brains(or incredibly hot cpus) have "mind", not whether or not they do have it.

In this case, the same thing: increased complexity.

From comp minds perspective "complexity" is -not- the explanation. You can make an incredibly complex and simultaneously weak system. Think Goldberg machines. Architecture is not interchangeable with complexity.

---

Which is, btw, what I think our brains are..that's why I don't think that we'll have to build a cpu that emulates the human brain to see AI, our brains weren't "purpose built" they were cobbled together. A purpose built system will be more powerful at a smaller scale, efficiency of computation.

---

But, in your model, why would increased complexity take us from "minimally conscious" to mind? Where do the functions come from to begin with, and why would they be more pronounced in one system, and not another? In comp mind, increased processing capability, not complexity, is what we would look for. Very often, in computer architecture, finding ways to -simplify- the system, and make it do more with less (and more reliably so) is what we're after.

The problem is that we already know about mind. We are not inferring it from other observations, but from direct experience of it, and none of the other observations we've made about the physical universe integrate well with what we know about mind.

I suppose you and I disagree here, in that mind seems, to me, to integrate very well with our observations about the physical universe. Mind (from ctm) appears to be a physical expression of logical principles - like a computer...literally, a logic sculpture. When we talk about "mind" a physical object can be pointed to, I can say "Well, if you wanted to do -that - you would need to build -this-". Now, does this mean that we accomplish -that- by doing -this- personally? No. What it does mean, is that it can be accounted for by -this-. So, it's difficult for me to see how mind doesn't integrate well with our observations of the physical universe.

What, precisely, doesn't integrate well?

For example, the subjectivity of mind is the only property which we fully accept as real, rather than as a "theory," but which cannot be observed. Even gravity, which is more ubiquitous than mind, is not approached in the same gnostic way in which we approach mind.

You and I may agree that there is a subjective experience - but the nature of that experience is still up for grabs if all we've done is agree that "something is happening", eh?
-More on what we accept in a moment.

Statement: mind is the subjective experience of data processing.

Demonstrate that any physical structure, including the brain, can be proven to have this property. I don't think it can be done logically or through observation, without any philosophical assumptions.

Data processing -is- a subjective experience.

Any system capable of addressing it's data or monitoring it's functions will be having what can -only be- a subjective experience. Regarding assumptions...neither you nor I - nor a computer, can prove -anything- without assumptions. All three of us state that -if a and b: c-. You state that we accept the subjectivity of mind as "real", but of course we (and I mean the royal we, you me, a computer...a beefy electronic toaster) would- if we wanted to "do work" on the statement, which we do.

I think that we have a particularly rich subjective experience (compared to any other system we might reference). A single self referential point of data (like whether or not a switch is on) is one thing, 80bil times 10k plus addressing a wide range of sensory apparatus and stored data is "rich".

This is why I think of mind as a transcendent property (looks like nobody will ever venture into that thread lol). Or you could look at it like this: in any system capable of complex enough functioning to supervene mind, no outside observer will be able to identify exactly what part of that complex functioning manifests as conscious experience.

Take pieces away until it stops acting like it has "mind", note that point...then replace them until "mind" returns and ask the subject to describe it's experience (if it doesn't...there's your bar...start looking at the last piece removed). There are plenty of ways to get false positives going this route- (as I'm talking about brute forcing mind - more sophisticated methods -could be- used) - but process of elim will get you to a point where you have an idea as to what you're looking for.

In ctm, you'd be looking for parallel architecture with direct bussing to the cpu registers (possibly, depending upon what you'd want the system to monitor and to what end that data would be used). An output of each gate that abstracts to "is this gate doing work?" or "to what group does this gate belong - what functions are possible on this input from this component?".

This applies not only to ANNs. Imagine taking a snapshot of 1 second of brain activity-- not fMRI, but every single chemical interaction that happened, and then figuring out where in that mess consciousness was created. Many are confident that mind is "in there" somewhere, but that level of complexity allows for the random supervenience of so many forms that one of those supervened forms may be mind, rather than anything the specific system does.

Of course we think that "mind" is -in there-.....all evidence suggests (very strongly) that this is the case. Where, in there, mind may be is difficult to pin to the wall, agreed.

It may be complex, but complex chemical interactions occur -everywhere on earth at every moment of every day- so the situation regarding complexity of interaction is no different in the brain than anywhere else, specifically on that point -without leading to any "mind" as we're using the term. So, why is that, where are the minds without brains? We'd have to create a special box and say that mind only occurs in human collections of unfathomably complex chemical interactions (or more troublingly, without any reliance on them).

I would need to find one and would expect to find one, under your model. Otherwise, we're talking about something that -only- manifests itself in the presence of adequate architecture...in which case it's pointless to posit that it comes from -elsewhere- (say, interactions of energy, or what have you, which occur pretty much everywhere at all times).

let me ask you this-- in a computational theory of mind, what is being "computed"? It seems to me you'd need metainformation-- i.e. information about your information, to arrive at an answer. But we don't have access to that.

Inputs. You don't need to know anything about the inputs, or what they refer to, or from whence they came. So long as the object in question is a computational system it will be capable of doing work in the presence of inputs (and needs no information about them whatsoever to do so).