Are You Conscious?

[PixyMisa]

No. You. Did. Not.

What. You. Are. Describing. Is. Impossible.

Show. Some. Specific. Code. If. You. Disagree.
(or to be precise you can only do it in the cases where you already know what the last instruction is)

Yes, I saw this point in my last response.

There's two problems, though: First, it is actually possible; second, if you recall where this point came from, you're grasping at red herrings.

What different method? What will find the last instruction even if it was a return or a jump?

For example, a register that is set with the previous PC value when the PC value is changed by an instruction, and cleared when it isn't.

Now, you may object that this is changing the computer.

Yes, it is. But you can do that. It's not like the halting problem, where there is provably no general solution. A Turing-equivalent machine can be reprogrammed so that it's a Turing-equivalent machine that knows what the last instruction was.

Now, knowing what the next instruction will be on a machine that can test-and-branch in a single instruction that can refer to the value of the next instruction - that can lead to contradictions. (Location 9: If the next instruction is at location 10, go to location 20. Of course, a real computer would evaluate this once, ignore the implicit contradiction, and continue on from location 20.)

You cite this as an all purpose proof for every different occasion don't you?

No. But it is the key theorem underlying this argument, so it wouldn't hurt you to understand how important it is, and what its implications are.

But it does not even come remotely close to proving what you claim.

What it proves is that if I can show something can be done on one computing device (of a long list of types), even if I had to change the details of the instruction set or registers or stack or some such, then it can be done on any of the other types. There are computing devices less general, i.e. less powerful, than a Turing machine. There are definitions of computers that are more powerful, that can execute supertasks, but they all involve physical infinities and cannot be built. (So the brain cannot be such in any case.)

And by the Church-Turing-Deutsch thesis, we can extend this to physical systems. Anything that the brain, a physical system, can do, we can do on a Turing machine. (There are a couple of subtle issues where we can't, again, relating to infinities, but no argument has been presented that these issues are relevant to the discussion, and in any case they run smack into the Planck scale nature of the Universe and may be provably physically irrelevant if not provable mathematically irrelevant.)

Anyway, remember how we got into this discussion? You were insisting that each step of a computer program was isolated and couldn't know about other steps or other data. This is true as a tautology: Each step of the program is deterministic and refers only to the values to which it refers. But those values can be anything, and which values an instruction refers to can change as the program executes.

So, for example, a generalised neural net can be implemented on any Turing-equivalent device. Do you put forward the same objection with respect to a neural net, and if so, how do you apply it?

 

[westprog]

Anything that the brain, a physical system, can do, we can do on a Turing machine.

Wrong.

See, anyone can do it.

 

[PixyMisa]

However, this doesn't apply to everything that happens on a computer. Certain programs are time dependent, because they interact with the real world in such a way that if they take too long, they will not produce the same outcome.

Yes. But this is provably irrelevant.

How does this relate to consciousness?

How does this relate to consciousness?

Well, partly because the AI people in the seventies were not involved with the real-time programmers. They were using the Turing model, where time was not an issue. So we have theories in which any two implementations of the same Turing machine are functionally equivalent - as in the rhetorical question that Pixy asked Aku. Is a simulation of a computer equivalent to a computer? Well, the programs will produce the same output. But if you want to play a YouTube video, you'll quickly find that the simulation does produce functionally - and qualitatively - different output.

Again, you are simply grasping at red herrings.

First, you certainly haven't presented any evidence that consciousness is time-dependent, merely hand-waving. The question of time dilation that you dismiss as irrelevant in fact cuts right to the point, because it allows us in principle to change the time factor.

We can do the same thing more simply: We place you in an isolation tank. Then we show you a video. Then we gradually speed up the video. At what point are you no longer conscious?

In any case, the entire argument is irrelevant. A simulated brain would, by your argument, be conscious if its environment were simulated at the appropriate pace. The simulated brain could still communicate with us - it would be running slower, or faster, but we could exchange emails just fine.

This is not what computing is like now, when most people interact continuously with their programs. And it's the interaction with the environment that defines human consciousness. Can a human mind be modelled by a pure Turing machine? Not if it is time dependent.

Sorry, this is simply untrue. Yes, any time-dependent function can still be modelled by a Turing machine; you simply internalise the time-dependence. How do you think weather forecasting works? We take a time-dependent process, simulate it, and increase the computing power until the simulation runs substantially faster than the real thing.

It would all be simple enough if RD could just bring himself to say that a pure Turing implementation of consciousness might well be lacking in some vital element - viz, time dependence.

It might be simple for you. It would also be untrue.

 

[PixyMisa]

Wrong.

See, anyone can do it.

Wrong.

When I do it, it's because you're repeating an argument already demonstrated to be incorrect.

Please show your disproof of the Church-Turing-Deutsch thesis if you want to continue. And, as already noted, you can't use physical infinities in your proof.

 

[PixyMisa]

When I asked you what was the explanatory power you answered "it depends upon the question you are asking".

Yes. What question are you asking?

Hmmm.... So computationalism predicts that we ought to be conscious. And we are conscious. Wow! That is so impressive.

It is correct. It doesn't need to be impressive.

But some neurons compute. And consciousness does not result from those computations.

So?

Well no, you are predicting that consciousness will always be associated with an equivalent algorithm whether or not it is associated with a functional brain.

More generally, yes.

And again, your "falsification" says - if this theory is false then we are not conscious. Again - not very impressive.

No, it says nothing of the sort.

Which is why I said that you didn't understand the concept of falsification - which has nothing to do with alternate hypotheses.

That's not the point.

The point is that the computational model is extremely broad, and it makes extremely broad predictions, and the ones that we can readily test are useless for falsification because we already know that they are true.

If you want to come up with something narrower, then you need to propose an alternative theory, and a test that distinguishes between the two. I did jump a step there in my response, which lead to some confusion, but I already cleared that up.

This "if the other theory is wrong then mine must be, by default, right" is woo, not science.

That's bears no relation to anything I said either.

So are you saying that an equivalent algorithm evaluated on any substrate will not necessarily produce this conscious experience that I have?

Does a sorting routing produce the same results regardless of what data you feed it?

But earlier you claimed very specifically that it did.

I said nothing of the sort, nor did I ever imply anything of the sort.

Er no, you rather specifically claimed that a set of calculations - even the desk check of a program, even if there were no physical connection between calculations - could result in the moment of consciousness you you are experiencing right now.

Another instantiation of the same moment of consciousness, certainly.

A set of calculations is more specific than an algorithm. An algorithm is a generalised method for solving a problem; a set of calculations is an instantiation of an algorithm against a specific instance of the problem.

An equivalent algorithm to what your brain does could produce anything or nothing, depending on how you apply it.

It is based on a perfect understanding of your position - if I had it wrong in any way I gave you more than ample time to correct me. I questioned you quite closely on this.

I have corrected you.

You're welcome.

 

[westprog]

Wrong.

When I do it, it's because you're repeating an argument already demonstrated to be incorrect.

Please show your disproof of the Church-Turing-Deutsch thesis if you want to continue. And, as already noted, you can't use physical infinities in your proof.

The Church-Turing thesis claims that any effective computation can be carried out on a Turing machine. It does not claim that anything can be done on a Turing machine. Indeed, the definition of effective computation is quite restrictive.

Equating "effective computation" with "anything" is not Church-Turing - that's pure Pixy. As an example of what cannot be done using a Turing machine - try tossing a ball in the air and catching it. A Turing machine can simulate this, of course - but a human brain can actually do it. An inability to tell the difference between the two is not a healthy thing.

Incidentally, I found the Stanford account of Church-Turing helpful, especially the section on misunderstandings.

The literature on the computational theory of the mind contains numerous endorsements of propositions ...that are supported by nothing more than a nod toward Turing or Church ... Perhaps some writers are simply misled by the terminological practice whereby a thesis concerning which there is little real doubt, the Church-Turing thesis properly so called, and a different thesis of unknown truth-value, are referred to indiscriminately as Church's thesis or the Church-Turing thesis -- albeit with accompanying hedges like ‘strong form’ and ‘physical version’....An error which, unfortunately, is common in modern writing on computability and the brain is to hold that Turing's results somehow entail that the brain, and indeed any biological or physical system whatever, can be simulated by a Turing machine.

 

[nescafe]

As an example of what cannot be done using a Turing machine - try tossing a ball in the air and catching it. A Turing machine can simulate this, of course - but a human brain can actually do it.

And an appropriately programmed computer with the correct sensors can actually do it too. Not exactly a Turing machine, and I am not claiming that the sensor/hand/computer ensemble is conscious, but it does emphasize the point that Pixy and I have been trying to get across w.r.t the boundaries of simulation.

ETA: First link was for robotic dribbling instead. Here is their page on dynamic regrasping -- throwing a cylinder in the air and catching it.

 

[rocketdodger]

This is quite an interesting post, because RocketDodger has previously given indications that he understands some things about computing from a practical point of view. His theory is pretty vague, but he seems to know what programs do, in a general sense.

So he should know - because it's pretty fundamental - that the steps of a computation that can be carried out on a Turing Machine are not time dependent. It doesn't matter if each step takes a nano-second or a year - the outcome will be the same.

However, this doesn't apply to everything that happens on a computer. Certain programs are time dependent, because they interact with the real world in such a way that if they take too long, they will not produce the same outcome. That's one reason why in the '70's and '80's DEC became a highly successful company - because they produced computers - and more importantly, operating systems - that were capable of producing guaranteed responses in a given time.

That is not to say, of course, that the mainframes being produced in that era - in fact, all computers ever built - didn't have time dependencies. It's impossible to build a computing device in the real world that doesn't have time dependencies. However, the intention of the bulk processing mainframes was to provide a Turing type environment for programs where timing issues would not apply. A COBOL payroll program would not be written with timing issues in mind. Indeed, there would be no COBOL keywords related to timing. The instructions would, as in the Turing model, be ordered, but the programmer would simply submit the program, and await the output, without concerning himself with precisely when any given instruction was executed. Indeed, the function of the multi-tasking operating systems was to conceal issues of timing from the programmer and provide a pure Turing environment.

In the more practical world of process control and monitoring, this abstraction was not possible. The DEC systems provided the programmer with tools that allowed guaranteed response times. There were also microprocessor based systems and microcontroller based systems. Languages such as FORTRAN were extended with system calls to ensure precise timing. Specialist languages such as RTL/2 were written. I've designed such languages myself.

Programs written in such languages, for such computing systems cannot be modelled on Turing Machines because Turing Machines don't concern themselves with issues of timing. Hence the real-time world was quite vulgar for a while, outside the mainstream of computer theory.

But the problems of real-time programming were also the problems of computing as a whole. The issue of synchronising processes running in parallel was important for operating system design as well as real-time control. And as computers became cheaper, more and more real-time applications arose.

It's now the case that most of the programs run on a normal PC have real-time issues. Play an MP3 and it can't be interrupted for half a second without noticing. Hit a key in Microsoft Word and you expect the corresponding letter to appear on screen. Microsoft Windows has the designs of those DEC real-time operating systems under the hood.

How does this relate to consciousness? Well, partly because the AI people in the seventies were not involved with the real-time programmers. They were using the Turing model, where time was not an issue. So we have theories in which any two implementations of the same Turing machine are functionally equivalent - as in the rhetorical question that Pixy asked Aku. Is a simulation of a computer equivalent to a computer? Well, the programs will produce the same output. But if you want to play a YouTube video, you'll quickly find that the simulation does produce functionally - and qualitatively - different output.

The AI view of computing seems stuck in it's world of the 1970's - stacks of punch cards ready for processing, according to the whim of the computer operater - each producing its output when it's finished. All modelled as Turing machines, and every run equivalent if it produces the same output.

This is not what computing is like now, when most people interact continuously with their programs. And it's the interaction with the environment that defines human consciousness. Can a human mind be modelled by a pure Turing machine? Not if it is time dependent.

Wtf are you blabbing about?

First, you really need to stop this whole "addressing a poster in the third person" thing, as if you are speaking to an audience about me -- it makes your posts sound like they were made by a tool.

Second, abstract computation is order dependent. That is kind of how we get the notion of an algorithm -- a series of steps to get something done. You can't just do any step you like in any order you like.

And what does order dependence imply in reality? Hmmm, maybe ... time dependence?

Because if I have a machine that takes X seconds to compute step n of an algorithm, we must wait those X seconds before we can proceed to step n + 1.

Do you disagree with this?

This is like when WW2 bombers used to jettison aluminium strips to avoid radar. Does general relativity and time dilation have anything to do with what we're talking about? No, but create some noise and confusion.

The only reason you think general relativity has nothing to do with this is that you don't know much about either.

Let me ask you a question: If you are on a starship travelling 0.9999 c relative to Earth, are you going to notice your own consciousness slowing down?

I predict your answer will be "we haven't been in that situation, so I couldn't say." And you accuse me of dodging the issue ...

So since you are going to dodge -- like you do every time -- let me break it down for you.

Assuming you would not notice your consciousness slowing down, that means your perception of your own consciousness would be normal while my perception here on Earth was that every particle in your brain slowed down immensely.

So how could consciousness be "time dependent" in the way that you seem to misunderstand if your consciousness functions just fine on the starship, from your point of view?

Hint -- it has to do with the time dependence of any real system being reducible to order dependence in a mathematical sense. Kind of like I explained in the previous post which you proceeded to mock.

EDIT -- Oh, and BTW, if you want to claim that your consciousness might actually slow down from your own POV, it means you are a dualist, because there isn't a shred of scientific evidence that supports such a claim.

 

[westprog]

And an appropriately programmed computer with the correct sensors can actually do it too. Not exactly a Turing machine, and I am not claiming that the sensor/hand/computer ensemble is conscious, but it does emphasize the point that Pixy and I have been trying to get across w.r.t the boundaries of simulation.

Not a Turing machine at all, in fact.

The point I was making was that a Turing machine cannot, in principle, interact with the world in real time. A computer can. It can do things that a Turing machine cannot. This is important when we're thinking about whether consciousness is necessarily equivalent to the operation of a pure Turing machine. Pixy still seems to be asserting that it is - and that Church-Turing proves this. I've provided contrary references.

ETA: First link was for robotic dribbling instead. Here is their page on dynamic regrasping -- throwing a cylinder in the air and catching it.

 

[westprog]

Wtf are you blabbing about?

First, you really need to stop this whole "addressing a poster in the third person" thing, as if you are speaking to an audience about me -- it makes your posts sound like they were made by a tool.

Second, abstract computation is order dependent. That is kind of how we get the notion of an algorithm -- a series of steps to get something done. You can't just do any step you like in any order you like.

And what does order dependence imply in reality? Hmmm, maybe ... time dependence?

Because if I have a machine that takes X seconds to compute step n of an algorithm, we must wait those X seconds before we can proceed to step n + 1.

Do you disagree with this?

Turing-equivalence implies that the time an algorithm takes to complete is irrelevant. It is not an issue. Hence anyone putting forward a pure Turing explanation of consciousness is claiming that the timing of the machine will not effect the experience of consciousness.

I am saying that if consciousness is timing dependent, then it is not a pure Turing phenomenon. IOW, if an operation taking X seconds gives a different result to one taking Y seconds, then it is not compatible with a Turing machine, where the duration of operations is deliberately not specified, and cannot effect the result.

Chaff about relativity deleted. If anyone else wants to discuss it, go ahead.

 

[rocketdodger]

The Church-Turing thesis claims that any effective computation can be carried out on a Turing machine. It does not claim that anything can be done on a Turing machine. Indeed, the definition of effective computation is quite restrictive.

Equating "effective computation" with "anything" is not Church-Turing - that's pure Pixy. As an example of what cannot be done using a Turing machine - try tossing a ball in the air and catching it. A Turing machine can simulate this, of course - but a human brain can actually do it. An inability to tell the difference between the two is not a healthy thing.

Incidentally, I found the Stanford account of Church-Turing helpful, especially the section on misunderstandings.

Round and round we go on the merry-go-round of ignorance!

drkitten has already explained to you -- and I know, because I read the posts (although you most likely didn't, despite the fact that they were addressed to you) -- that although the CT thesis is formally restricted to "effective methods" it is just fine and dandy for the computational model because people have been trying to find instances of real things that don't satisfy the notion of "effective method" for 50 years and nobody has done so yet.

So saying with glee "the notion of effective computation is quite restricting" and then running around the room thumbing your nose at everyone is pretty stupid, because all the actual systems anyone has ever observed fall under that "restrictive" notion.

 

[PixyMisa]

Not a Turing machine at all, in fact.

The point I was making was that a Turing machine cannot, in principle, interact with the world in real time.

Because Turing machines don't exist.

A computer can.

Computers are finite instantiations of Turing machines.

It can do things that a Turing machine cannot.

Only because Turing machines don't exist.

Computationally, no, it can't.

This is important when we're thinking about whether consciousness is necessarily equivalent to the operation of a pure Turing machine.

No, it's completely irrelevant.

Pixy still seems to be asserting that it is - and that Church-Turing proves this.

The Church-Turing thesis alone doesn't prove this. However, it leaves only certain, very specific, reasons why it wouldn't work. And you have never addressed any of those reasons.

I've provided contrary references.

No you haven't.

 

[rocketdodger]

I am saying that if consciousness is timing dependent, then it is not a pure Turing phenomenon. IOW, if an operation taking X seconds gives a different result to one taking Y seconds, then it is not compatible with a Turing machine, where the duration of operations is deliberately not specified, and cannot effect the result.

Except for the fact that operations are defined according to their result...

Really, I would like you to give an actual example of something that you have in mind, because I think this is nonsense and you will only realize it once you try and then fail to give such an example.

Chaff about relativity deleted. If anyone else wants to discuss it, go ahead.

Thank you for confirming my predictions.

 

[PixyMisa]

Turing-equivalence implies that the time an algorithm takes to complete is irrelevant.

Right.

It is not an issue.

Right.

Hence anyone putting forward a pure Turing explanation of consciousness is claiming that the timing of the machine will not effect the experience of consciousness.

Wrong.

What you are doing - which is a pure red herring - is taking a non-time-dependent simulation of the brain, and the natural world. You will of course find differences in the way the simulation experiences the world - it may run faster and notice more details than we do, or run slower and notice fewer details.

You have never put forward any reason why it would not be conscious.

I am saying that if consciousness is timing dependent, then it is not a pure Turing phenomenon.

But this does not follow. It's just flat wrong.

You can model time-dependent processes on Turing-equivalent machines, and we do this all the time.

IOW, if an operation taking X seconds gives a different result to one taking Y seconds, then it is not compatible with a Turing machine

Wrong.

where the duration of operations is deliberately not specified, and cannot effect the result.

The timing of the operations of the Turing machine aren't specified, because Turing machines don't exist.

But you can model any physical process on a Turing machine, time dependencies and all.

Chaff about relativity deleted. If anyone else wants to discuss it, go ahead.

Westprog, your post was entirely chaff. Every sentence was a falsehood already refuted.

 

[westprog]

Round and round we go on the merry-go-round of ignorance!

drkitten has already explained to you -- and I know, because I read the posts (although you most likely didn't, despite the fact that they were addressed to you) -- that although the CT thesis is formally restricted to "effective methods" it is just fine and dandy for the computational model because people have been trying to find instances of real things that don't satisfy the notion of "effective method" for 50 years and nobody has done so yet.

So saying with glee "the notion of effective computation is quite restricting" and then running around the room thumbing your nose at everyone is pretty stupid, because all the actual systems anyone has ever observed fall under that "restrictive" notion.

Tell it to Stanford university. They very clearly, in the article I quoted, point out the mistakes in the computational view, explain why they are mistakes, explain what Church-Turing actually says, and so on, etc etc etc. You can continue to bluster about it, but the article is quite clear.

 

[PixyMisa]

The Church-Turing thesis claims that any effective computation can be carried out on a Turing machine. It does not claim that anything can be done on a Turing machine. Indeed, the definition of effective computation is quite restrictive.

All of which is perfectly true.

So what is it that you are claiming is not an effective computation?

Equating "effective computation" with "anything" is not Church-Turing - that's pure Pixy.

I never said anything of the sort.

As an example of what cannot be done using a Turing machine - try tossing a ball in the air and catching it. A Turing machine can simulate this, of course - but a human brain can actually do it.

Red herring - false equivalence. And a non-sequitur through and through.

A Turing machine can do it in its domain. A human can do it in the human's domain. Turing machines are not something that physically exists, so they can't catch balls that physically exist. The number 42 can't play the violin, and the concept of justice can't ride a horse.

An inability to tell the difference between the two is not a healthy thing.

Damn straight.

Incidentally, I found the Stanford account of Church-Turing helpful, especially the section on misunderstandings.

I've read that page. It plays up the limitations of the Church-Turing and Church-Turing-Deutsch theses beyond what's really supportable. The limitations are well-defined and quite specific. Yes, a Turing-equivalent machine can't execute a supertask, and Church-Turing-Deutsch doesn't handle physical infinities.

So, where's the supertask? Where's the physical infinity? What is your actual objection?

 

[PixyMisa]

Tell it to Stanford university. They very clearly, in the article I quoted, point out the mistakes in the computational view

No they don't.

explain why they are mistakes

No they don't.

explain what Church-Turing actually says

It actually does this fairly well.

and so on, etc etc etc. You can continue to bluster about it, but the article is quite clear.

And not applicable.

Supertask, Westprog?

Physical infinities?

Some other specific process you can demonstrate happening in the brain that is not computable? Evidence, please, that it happens in the brain, and that it is not computable. Once you've provided that, then we can discuss whether that represents a problem for the computational model.

 

[westprog]

Except for the fact that operations are defined according to their result...

Really, I would like you to give an actual example of something that you have in mind, because I think this is nonsense and you will only realize it once you try and then fail to give such an example.

Playing an MP3 file. Do it in X seconds and it's fine. Do it in Y and it doesn't work. I explained this to you already. I showed the difference between computation and what a computer can actually do, and how time dependence affects certain operations to the extent that they will not take place correctly if they don't complete within a given time.

I explained all this and you said "WTF are you blabbing about". Then you tried to bring in relativity. Sometimes I prefer the monosyllabic Pixy approach. It's easier to ignore.

 

[PixyMisa]

Playing an MP3 file. Do it in X seconds and it's fine. Do it in Y and it doesn't work.

Define "work".

I explained this to you already.

Yes, and you were wrong then too.

I showed the difference between computation and what a computer can actually do

And you were wrong about that.

and how time dependence affects certain operations to the extent that they will not take place correctly if they don't complete within a given time.

Which can of course be modelled by any Turing-equivalent system, and indeed is something we do all the time.

I explained all this and you said "WTF are you blabbing about".

Yes, because you were wrong.

Then you tried to bring in relativity.

Which is one of many things that demosntrate that you are wrong.

Sometimes I prefer the monosyllabic Pixy approach. It's easier to ignore.

Maybe so. Maybe so. You're still wrong.

 

[westprog]

I never said anything of the sort.

Anything that the brain, a physical system, can do, we can do on a Turing machine.

But it's all just circular reasoning over and over. The brain is a computational device, therefore it's a computational device.