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

[rocketdodger]

And so according to the theory a real unified human consiousness would result across many light years with no communication whatsoever between the components creating this consciousness.

No -- you haven't been paying attention.

The consciousness occurs when you program the devices, not when they execute their instructions.

In other words, the "recheck" doesn't mean anything, only the initial run. And everything else is simply a remapping of the initial run. The initial run is the consciousness.

 

[Paul C. Anagnostopoulos]

And so a brain might not be an algorithm and still not magic.

In short, the brain might just be physical.

Right. I'm just waiting for some evidence that we need anything nonalgorithmic. It is certainly a possibility.

Give a different output for the same input.

I am not saying there is anything special about that, I am just saying that it is not a TM and therefore it is not true to say that a non-TM cannot implement a TM.

A Turing machine with a pseudo-RNG can give a different output for the same input as long as you don't insist that the seed is part of the input (which for a long-term simulation is fine). A TM with a true RNG can give a different output for the same input no matter what. The concensus appears to be that a TM with a true RNG has no more power than a regular TM. So I think that if there is going to be something about the brain that makes it TM incompatible, it wont' be an RNG.

A clincher for what?

Is a TM+RNG more powerful than a TM alone? That is an important question.

Again you are missing the point.

The entire computationalist position here has hinged on drkittens claim that a non TM cannot implement a TM. But if a system can use non discrete values to implement a TM then the claim is wrong.

Sorry, I don't understand what you're saying. What is a "non TM"?

And I haven't heard any coherent suggestion as to why it should be necessary in the first place.

It doesn't lead to a verifiable, falsifiable hypothesis. It does explain anything. It is not simpler than the other hypothesis, in fact it is more extravagant.

It is not the obvious position or the default position - it seems more probable that the brain has non-discrete processes and would therefore not classify as an algorithm.

So I say the mind may or may not be an algorithm. I don't know. What is wrong with that? Why assume a position with no explanatory power?

I'm not sure what other people are saying, but I'm not saying the brain is a Turing machine. I'm simply suggesting that it is doing nothing that cannot be simulated on a Turing machine and thus a brain simulator can be built for a conventional computer. I very much doubt that it is using noncomputable values to more than a few decimal places. I also doubt that it is doing parallel processing that cannot be done with multiple computers.

No I am suggesting that there is something that is no more powerful than a Turing machine that is not a Turing machine.

Well sure. It's pretty clear that the brain is not actually a tape and a read/write head. Neither is a conventional computer. The question is whether it's Turing machine compatible.

And I am just waiting for a compelling argument that the brain must be a Turing Machine.

I am not rooting for any team.

Fair enough. The burden of proof is on both sides.

But when there is an extravagant, unverifiable, unfalsifiable metaphysical claim infallibly declared (and I dont mean you) on the basis of a questionable interpretation of a well-known mathematical result, then that seems the very definition of woo to me.

The very best knowledge we will ever have on the subject will be in the form of falsifiable, verifiable hypotheses.

And I don't think there is one for either position at the moment. In fact we will probably have great advances in neuroscience without even caring about the answer to this question any more than we care about whether or not there really is a smeared out dead/alive cat in Schroedinger's box.

Yet I daresay that there is one question that will never be settled. No matter how much we learn about consciousness and no matter how clever the simulations become, there will always be people who say that we have only discovered the neural correlates of consciousness and not actual consciousness.

~~ Paul

 

[PixyMisa]

I'm not sure what other people are saying, but I'm not saying the brain is a Turing machine. I'm simply suggesting that it is doing nothing that cannot be simulated on a Turing machine and thus a brain simulator can be built for a conventional computer.

This is one of a number of points that needs to be posted in bold red 40 point all caps and probably blinking. We should make a list.

 

[Dancing David]

But again, that is fine. It is something that could apply to anything in the universe.

What we are looking at is the properties of pain that might differentiate it from, say, the Thames.

By the defintion of the behaviors of 'pain'. therefore if you demonstrate that the Thames has the behavioral criteria of pain, then it experiences pain.

Unless we are headed to Franko's 'how do you knwo a rock is not conscious" question.

There are no p-zombies under behaviorism, if something acts as though it is conscious, then it is. Because it is only by denitional usage that terms have meaning in behaviorism, radical behaviorism^WP i can't speak on. I agree with it, but don't really understand it. I thought i did but according to Jeff Corey I am a methodologist as a behaviorist.

 

[Dancing David]

But what about all the stuff that happens inside your brain that you are not conscious of?

Do you call that internal or external?

Private, even from yourself.

 

[AkuManiMani]

I'm not sure what other people are saying, but I'm not saying the brain is a Turing machine. I'm simply suggesting that it is doing nothing that cannot be simulated on a Turing machine and thus a brain simulator can be built for a conventional computer.

I don't think that there is anyone here saying that the brain cannot be simulated with a TM. But keep in mind that the operative word here is "simulation". A simulation is just a dynamic representation; in other words, it just the description of a thing and not the thing itself. I think many here are making a serious categorization error.

For instance...

A detailed simulation of photosynthesis gives us a helpful model of the process, but it will not fix carbon or provide chemical energy. A simulation of the solar system would be an invaluable tool for understanding the one we live it, but it will not produce gravitational effects. Like wise, a comprehensive simulation of a brain may give us insights on it's functioning but that does not mean that it will actually produce consciousness.

Each of the simulations in those examples are descriptive analogues of physical processes. But its important to remember that a description of a physical process, no matter how detailed, is NOT a reproduction.

 

[!Kaggen]

I don't think that there is anyone here saying that the brain cannot be simulated with a TM. But keep in mind that the operative word here is "simulation". A simulation is just a dynamic representation; in other words, it just the description of a thing and not the thing itself. I think many here are making a serious categorization error.

For instance...

A detailed simulation of photosynthesis gives us a helpful model of the process, but it will not fix carbon or provide chemical energy. A simulation of the solar system would be an invaluable tool for understanding the one we live it, but it will not produce gravitational effects. Like wise, a comprehensive simulation of a brain may give us insights on it's functioning but that does not mean that it will actually produce consciousness.

Each of the simulations in those examples are descriptive analogues of physical processes. But its important to remember that a description of a physical process, no matter how detailed, is NOT a reproduction.

Whilst you see a difference some will point out that you cannot explain the difference and therefore others will deny a difference. Whilst I see all these views as equally wrong and equally right depending on the time line of cognition.

 

[Paul C. Anagnostopoulos]

I don't think that there is anyone here saying that the brain cannot be simulated with a TM. But keep in mind that the operative word here is "simulation". A simulation is just a dynamic representation; in other words, it just the description of a thing and not the thing itself. I think many here are making a serious categorization error.

For instance...

A detailed simulation of photosynthesis gives us a helpful model of the process, but it will not fix carbon or provide chemical energy. A simulation of the solar system would be an invaluable tool for understanding the one we live it, but it will not produce gravitational effects. Like wise, a comprehensive simulation of a brain may give us insights on it's functioning but that does not mean that it will actually produce consciousness.

I think that a simulation of the brain is one of those simulations that does produce a functioning brain, for all intents and purposes. Just like the simulation of arithmetic produces the correct answers and the graphical simulation of a book produces readable text. This is an empirical question, of course.

~~ Paul

 

[Paul C. Anagnostopoulos]

Whilst you see a difference some will point out that you cannot explain the difference and therefore others will deny a difference.

As I said above, some people will deny to their graves that we have produced machine consciousness. No matter how much it appears that the machine is having phenomenal experiences, they will say that it is not. These people have a vested interest in keeping the definition of phenomenal experience vague so that the goalposts are on wheels. I am not saying that the definition is easy to come by, only that some folks want to keep it slippery.

~~ Paul

 

[!Kaggen]

I think that a simulation of the brain is one of those simulations that does produce a functioning brain, for all intents and purposes. Just like the simulation of arithmetic produces the correct answers and the graphical simulation of a book produces readable text. This is an empirical question, of course.

~~ Paul

Aren't all relevant questions ultimately empirical ?

 

[westprog]

As I said above, some people will deny to their graves that we have produced machine consciousness. No matter how much it appears that the machine is having phenomenal experiences, they will say that it is not. These people have a vested interest in keeping the definition of phenomenal experience vague so that the goalposts are on wheels. I am not saying that the definition is easy to come by, only that some folks want to keep it slippery.

~~ Paul

This might be true, but I don't think that there's any doubt that some people will insist to their graves that we have produced machine consciousness. No matter how little evidence there is that the machine is having phenomenal experience, they will say that it is. And, yes, they do have a vested interest in making phenomenal experience well defined - even if it means excluding whatever we understand by "experience".

 

[!Kaggen]

As I said above, some people will deny to their graves that we have produced machine consciousness. No matter how much it appears that the machine is having phenomenal experiences, they will say that it is not. These people have a vested interest in keeping the definition of phenomenal experience vague so that the goalposts are on wheels. I am not saying that the definition is easy to come by, only that some folks want to keep it slippery.

~~ Paul

Its a matter of motivation.
When we want to build a machine with consciousness we want a model.
When we want to express our consciousness we want poetry.

 

[westprog]

Maybe in about ten years we will be able to model a person playing chess.

AI is like nuclear fusion in that sense; it's always a fixed interval short of reaching completion.

 

[westprog]

I don't understand this point. Computers use approximations of real numbers all the time. They are Turing compatible. You don't think the brain needs noncomputable reals to work, do you?

The brain needs real quantities in order to work. It's a real object. Whether those quantities equate to the real numbers of mathematical theory is a fairly deep question.

 

[westprog]

It appears that randomness does not confer additional power.

We don't know that. It depends, of course, what you mean by "power". If you mean "ability to perform calculations on the natural numbers", then yes, it seems unlikely that adding a random element to the machine would make any difference. If it's a matter of how people have experiences, then randomness might well be the element providing that capacity.

There's nothing in the theory of Turing machines or Turing machines with a random capacity that indicates which of them would be able to produce experience. The word "powerful" is irrelevant in that context.

 

[westprog]

The conclusion. How do you know that conscious experience isn't simply a matter of processing information?

~~ Paul

I still don't know what is meant by "processing information". Information is passing through every part of the universe all the time. What does it mean to "process" it?

I'd be interested to see if this one can be answered without rhetorical questions in the form of "So you think that...".

 

[rocketdodger]

We don't know that. It depends, of course, what you mean by "power". If you mean "ability to perform calculations on the natural numbers", then yes, it seems unlikely that adding a random element to the machine would make any difference. If it's a matter of how people have experiences, then randomness might well be the element providing that capacity.

There's nothing in the theory of Turing machines or Turing machines with a random capacity that indicates which of them would be able to produce experience. The word "powerful" is irrelevant in that context.

Ah, the westprog admission of incorrectness.

Simple pattern, we see it all the time:

1) It depends on the definition you are using (nevermind that everyone but myself is using the commonly understood definition) and yes, if you use that definition I might be wrong.

2) And anyway, it is irrelevant, because we haven't gathered enough data on the issue.

 

[westprog]

In other words, you have to have some basic knowledge of how the system works to begin with. So you weren't trolling; you just misspoke. Gotcha.

And in any system, it's necessary to know what is the essential component and what is peripheral. One can observe some aspect of behaviour and implement it, and not realise that some vital element is missing.

Thus the numerous wing-flapping attempts at flight. It's also noteworthy that for the people wedding to the flapping model, the failure of their attempts just indicated the need to flap harder.

 

[rocketdodger]

I still don't know what is meant by "processing information". Information is passing through every part of the universe all the time. What does it mean to "process" it?

I'd be interested to see if this one can be answered without rhetorical questions in the form of "So you think that...".

OOoh, look, another common westprog pattern:

1) This has been explained to me by others, but since you might not be aware of that, I will act naive in the hopes that you might fall for my incorrect arguments without noticing how incorrect they are.

Seriously ?

Are you going to continue playing dumb until every last forum member shows why you are wrong?

Off the top of my head, here is a list of individuals who have explicitly shown why your claim is wrong -- to the point of you dropping the issue with them, I might add.

1) Me.
2) yy2bggggs.
3) Belz.
4) Randfan.
5) drkitten.

That is five thus far. Paul soon to be added to the list, I suppose...

 

[rocketdodger]

I don't think that there is anyone here saying that the brain cannot be simulated with a TM. But keep in mind that the operative word here is "simulation". A simulation is just a dynamic representation; in other words, it just the description of a thing and not the thing itself. I think many here are making a serious categorization error.

For instance...

A detailed simulation of photosynthesis gives us a helpful model of the process, but it will not fix carbon or provide chemical energy. A simulation of the solar system would be an invaluable tool for understanding the one we live it, but it will not produce gravitational effects. Like wise, a comprehensive simulation of a brain may give us insights on it's functioning but that does not mean that it will actually produce consciousness.

Each of the simulations in those examples are descriptive analogues of physical processes. But its important to remember that a description of a physical process, no matter how detailed, is NOT a reproduction.

Can you offer any evidence that we are not currently inhabiting a simulation?

If not, then what are the implications of this?