My take on why indeed the study of consciousness may not be as simple

[westprog]

[My bold.]

It is possible to say that a signal is carrying digital information. It is also possible to say that it is carrying at most X amount of digital information per Y unit of time. It is further possible to be specific with this analysis with regards to modulation and coding schemes.

I'm not familiar enough with the limits of information technology, but it seems likely that while it's possible to set the limits of a given signal's capacity wrt modulation and coding, it's very difficult to know whether it is carrying information according to some different arrangement.

The more of the signal we have, the more certain we can be with our assessment, just as is the case with your rabbit and dog. With a single base pair, you can't even guess. There is no fixed lower bound for information transfer rates - short of the heat death of the Universe - so the analogy is inexact. But for any quantitative question we can provide a quantitative answer.

 

[westprog]

So are you still stating that by architecture we can not tell digital computing from analog computing?

We can distinguish them in engineering terms, or functional terms. If we build a machine to be an analogue computer, then we can recognise it.

However, in physical terms, aside from usage (which is the point I've been making post after post), both machines will manipulate electricity, and the digital computer will be performing analogue operations, and the analogue computer can provide digital information.

The point about usage is critical. If a device is built to perform a particular function, then we identify it as being that type of device. But purpose has nothing to do with physics. If we are interested in what happens in a computer independently of our use of it, we have to step back and consider the physical processes. The physical processes of a digital signal cannot be distinguished from a non-digital signal. (I prefer not to contrast digital and analogue signals as being different things, because every digital signal must be an analogue signal).

 

[westprog]

That is what I am saying is goal moving, we can always look at margin and single cases to destroy any theory.

That's science. The exception disproves the rule.

However, in this case, we aren't dealing with odd outliers. We're dealing with fundamental physical principles.

The original statement was about devices.

We can proceed to devices if we have consensus on signals.

 

[PixyMisa]

Alright.

I have been lurking in this thread for some time, and there is something that I don't quite understand...

What on earth is the relevance of the computer vs. a pile of computer parts?

How is this any different than a brain vs. a sloppy bag of brain parts?

Please. Enlighten me!

Nobody knows but westprog, and he/she's not telling.

 

[drkitten]

Alright.

I have been lurking in this thread for some time, and there is something that I don't quite understand...

What on earth is the relevance of the computer vs. a pile of computer parts?

Penrose has suggested that the secret to consciousness occurs in a particular physical construct (the microtubules of the human brain).

The Strong AI hypothesis suggests that the secret to consciousness occurs in a particular functional construct (a particular type of program running on a Turing machine or equivalent).

A bag of computer parts (or a bag of brain parts, for that matter), contains the same physical constructs as a functioning system, but is functionally very different.

Two different instantiations of a Turing machine will contain the same functional constructs, but are physically very different.

If the secret to consciousness is lies purely in the microtubules, then a computer can never be "conscious." But, of course, a bag of brain parts, which contains the same microtubules, would be conscious.

 

[PixyMisa]

I'm not familiar enough with the limits of information technology, but it seems likely that while it's possible to set the limits of a given signal's capacity wrt modulation and coding, it's very difficult to know whether it is carrying information according to some different arrangement.

Some different arrangement?

So there's modulation and coding and... Some different arrangement?

Shannon and Nyquist just added a few hundred posthumous RPM.

 

[PixyMisa]

We can distinguish them in engineering terms, or functional terms. If we build a machine to be an analogue computer, then we can recognise it.

No you wouldn't.

 

[Wudang]

The physical processes of a digital signal cannot be distinguished from a non-digital signal. (I prefer not to contrast digital and analogue signals as being different things, because every digital signal must be an analogue signal).

I think I see where you're coming from. But, assuming I'm right, doesn't that only apply in the case of a signal where we do not observe the intended recipient and hence do not observe the marshalling or demodulation or xxxxxx that the signal then undergoes?

 

[westprog]

Updating of the program status word?

And in what chapter of your physics textbook will you find that defined?

When I say physical process, that's what I mean. In engineering and functional terms the computer is entirely different from the parts - because a human being uses it.

This is only relevant if computation is claimed to be a physical process. If it isn't then we are in a different realm.

 

[Gate2501]

Penrose has suggested that the secret to consciousness occurs in a particular physical construct (the microtubules of the human brain).

The Strong AI hypothesis suggests that the secret to consciousness occurs in a particular functional construct (a particular type of program running on a Turing machine or equivalent).

A bag of computer parts (or a bag of brain parts, for that matter), contains the same physical constructs as a functioning system, but is functionally very different.

Two different instantiations of a Turing machine will contain the same functional constructs, but are physically very different.

If the secret to consciousness is lies purely in the microtubules, then a computer can never be "conscious." But, of course, a bag of brain parts, which contains the same microtubules, would be conscious.

Isn't this damning to the argument for this "microtube consciousness" then? Clearly a sloppy bag of disconnected brain parts, if hooked up to a body, would not be able to demonstrate conscious behavior(we see this in braindead patients). Neither would a bag of computer parts hooked up to a power supply and the proper peripherals, of course, but then again, we wouldn't expect it to.

It seems like a good argument AGAINST the point that Westprog seeks to make...

Gate is still confused.

 

[Wudang]

And in what chapter of your physics textbook will you find that defined?

Chapter 34. "Physics" Paul A Tipler. 1976 - where he discusses semi-conductors.

When I say physical process, that's what I mean. In engineering and functional terms the computer is entirely different from the parts - because a human being uses it.

This is only relevant if computation is claimed to be a physical process. If it isn't then we are in a different realm.

It's instantiated in a physical process. All computation in a computer is carried out by updating the PSW which is a defined piece of carefully-poisoned silicon.

 

[drkitten]

When I say physical process, that's what I mean. In engineering and functional terms the computer is entirely different from the parts - because a human being uses it.

This is only relevant if computation is claimed to be a physical process. If it isn't then we are in a different realm.

So everything you've written after post #12 is based on a completemisunderstanding, then.

Figures.

I agree. I'm not sure where the pessimism comes from among some philosophers / researchers, that consciousness may be an insoluble paradox. The computational model for the mind is so suggestive -- all knowledge based on nets of comparison functions (x is A, y is not-A), consciousness the highest-level comparison function, likely recursive (I was I, now I'm not) -- that even if it's not the whole story, the gaps that remain should only offer even more knowledge about the nature of the brain and body and their relation to consciousness, not insuperable barriers to understanding.

This is explicitly a claim that computation (and consciousness, for that matter) is defined mathematically, not physically. (Which, of course, is completely compatible with the lambda calculus and whatnot as defined by Church and his friends.)

 

[drkitten]

Isn't this damning to the argument for this "microtube consciousness" then?

Um, yes.

No that there's much there to damn. Penrose is so clearly wrong that arguing against his position is beating a greasy smear on the ground where a dead horse used to be.

It seems like a good argument AGAINST the point that Westprog seeks to make...

Gate is still confused.

No, Gate is not confused. Westprog is confused, and you have correctly identified a major problem with his argument.

 

[westprog]

I think I see where you're coming from. But, assuming I'm right, doesn't that only apply in the case of a signal where we do not observe the intended recipient and hence do not observe the marshalling or demodulation or xxxxxx that the signal then undergoes?

But that is usage - which is how I said we can make the distinction.

 

[westprog]

So everything you've written after post #12 is based on a complete misunderstanding, then.

Figures.



This is explicitly a claim that computation (and consciousness, for that matter) is defined mathematically, not physically. (Which, of course, is completely compatible with the lambda calculus and whatnot as defined by Church and his friends.)

I suggest you take it up with the people who think it's a physical process then. I tend to think it isn't.

If computation is not physical, and consciousness arises from computation, that seems to imply that consciousness isn't physical either.

 

[westprog]

Sauce for the goose, then.



How people use the computer IS a physical process. Also, electrical current doesn't flow in the dissassembled parts, does it ?

If the power supply is plugged in it does.

 

[westprog]

And there are the same contraints on digital vs. analog signals.



So are you still saying that we can not physically an analog computation from a digital computation?

I'm saying that "computation" is not something that relates to physics, AFAIAA. I asked for a definition in physical terms some time ago, and I'm still waiting.

 

[Gate2501]

I suggest you take it up with the people who think it's a physical process then. I tend to think it isn't.

If computation is not physical, and consciousness arises from computation, that seems to imply that consciousness isn't physical either.

Ok, another question...

I personally think that computation is a series of physical processes. EDIT: Of course we can mathematically model these physical processes, could this be why you are confused?

If you think that computation is not a physical process(or processes), does that mean that there is not a thermodynamic limitation on computing? *hint: this question is a trap!*

 

[westprog]

Repeating an incorrect statement will not make it correct, no matter how many times you trye.

So what does Penrose do to qualify as being a physicist in your eyes? Since writing papers published in physics journals, devising new theories in physics, being a professor of physics, teaching physics and writing a huge book on physics clearly don't make him a physicist, just what would?

Or is it that you disagree with him, so you decide that he's not a physicist after all?

 

[westprog]

And you can't do this same sort of thing with a signal?

We can't, for instance, look at a radio signal and determine whether or not it encodes a particular television program in analog or digital, given that we know exactly what each of the signals encoding those programs will look like?

Of course we can look at a given signal with a given encoding scheme, and say that data is encoded in it according to that scheme. We cannot, even in that specific case, where the signal carries analogue information, state with certainty that the signal does not also carry additional digital data, except via usage.

I think you are ignorant of the fact that we only know what the DNA of a rabbit or a dog looks like because we have sequenced the DNA of existing rabbits and dogs. Have you gone from being 30 years out of date to thinking we can determine the exact phenotype of a mammal just by glancing at a sequence?

I don't know what your point is here. The DNA of rabbits and dogs can be identified and distinguished. If we don't have enough of the DNA, we can't tell anything.