Explain consciousness to the layman.

[tsig]

Or if moving pebbles in the desert is a thought.
Or if the numbers on the computer screen at the bank is my worth.
Or if the number in my passport is my identity.

All part of the same problem.

What problem is that?

 

[!Kaggen]

What problem is that?

The one when people don't get the problem.

 

[tsig]

Look, if you use a definition, then you should accept what it means. "Whatever the brain does" is a bad definition, because a lot of what the brain does is not directly concerned with consciousness. Piggy's definition - what the brain does when it's awake that it doesn't do when it's asleep - is a good definition - or at least it's better.

It seems that if you take people literally on this topic, then you're equivocating.

So because the brain does one thing it means it can't do another?

 

[tsig]

That sounds alright for groundhogs, caterpillars and elephants, etc., but they can't write poetry, paint master pieces or write songs like 'Cat Scratch Fever'.

That's true and they're the better for it.

^{zen = no mind}

 

[westprog]

Then I already gave one. I didn't place any requirements that a human or conscious entity interprets the results--just a requirement that there exists a mapping.

Then that covers a vast area - because there are a lot of states in every physical system, and they can be mapped to whatever you like.

 

[yy2bggggs]

Then that covers a vast area - because there are a lot of states in every physical system, and they can be mapped to whatever you like.

That's correct--it does cover a vast area.

 

[lupus_in_fabula]

So because the brain does one thing it means it can't do another?

Sure it can. But as a definition of consciousness it’s rather crappy.

Although it’s slightly less crappy than: consciousness = "whatever stuff does".

 

[westprog]

You see when far more advanced technology than we have at present is considered. I change sides in this debate. This technology is a long way off though.

This I don't necessarily accept. If the process isn't properly understood, then it doesn't matter how much technology you throw at it - it won't work.

It might be that the erroneous thinking and fallacies involved in the AI viewpoint happen not to prevent their approach working, given sufficient resources. If the approach is wrong, then it will still be wrong with unlimited tech.

Back in 1970, if you'd asked an AI researcher what the state of the field would be in forty years time, given Moore's Law, I wonder what he'd have said. I wonder what an AI researcher would expect the field to be like in 2050.

 

[westprog]

Then please answer my question: how do hearts and oceans perform algorithms and calculations ?

They have transitions between different states, which can be mapped to the execution of an algorithm. If any physical state will do, then any state transition is meaningful.

The number of states available - given as many molecules as there are in an ocean - is very large. Much greater than the number of states available in the bits on a digital computer.

 

[IanS]

I broadly agree with what you said above about the nature of our universe boiling down to a set of particle-field interactions (hence the descriptions I have given before of "Universe/Big-Bang from Nothing", however, I think there is a problem with the quote above -

- how will you create particles in any model/simulated world, without transferring information from the particles that presently make up our own universe?

If you have to lose information from particles in our universe every time you create or simulate a particle in some other world, then the process is self-defeating and physically impossible, because it cannot be done without changing the particles of our present world/universe.

The only way you could make a simulated world like that, would be if there was absolutely no connection between the simulated world and our universe. But in that case, by definition, there would be no longer be any way to produce the simulation, because there would be no possibility of any transfer between the two unconnected worlds.

I think that is his whole point, a simulation has nothing to do with the physical constraints. Its the principle that counts.

OK, but in that case it means it cannot actually be done. I’m referring to the quote below -

The mechanics are simple: Just simulate every single particle in their world, using rules identical to the rules of particle interaction in our own world.

- I’m suggesting that it cannot actually be done, because it’s prevented by the fact that it would mean materially changing the properties of particles and fields in the donor world. The process would be self defeating.

Anyone can of course conceive of anything as a mater purely of human imagination. But that does not mean it can actually be done in any universe that we know of. In this case, I think that it cannot work … it will not actually be possible to “ simulate every single particle in their world, using rules identical to the rules of particle interaction in our own world identical rules “.

You could take an idea into another world as a "set of rules", but that alone won’t produce any particles at all. You would need to actually transport real properties of the fields and particles into that other world ... but doing that would destroy the very same particles and fields in the donor world. That’s why it won’t work.

 

[westprog]

Where does the "other world separate from us" come in between the kid pressing the button and the image on the screen changing in ways that make the kid imagine a person jumping?

That's the important point. If the kid isn't pressing the button, or watching the screen, what does the game seem like to Mario?

Let's cut this right down to a very simple virtual world - the word "pear". In order to communicate this information to the people reading this message, I need just 32 bits. Four eight-bit ASCII characters. However, the people reading it will extract an enormous amount of information from those 32 bits. Is that information concealed in that small number? Obviously not. The number is just a way that my concept of "pear" can be transferred. What does the pear seem like to the number? The very question is meaningless. There is no virtual pear inside the number. The pear exists in my consciousness, and I'm passing that message to the people reading this via a number which decodes into the symbols that appear on the screen as "pear". This triggers associations. While someone else's association will not be exactly the same as mine, it will still be close enough for meaningful communication to take place.

How about the Mario game? It's effectively no different. The designers of the game are passing information to the child playing the game. The information is passed as a succession of numbers, which, just as the representation of the word "pear" on someone's browser, are represented as sensory data - visual and auditory. Where is the "virtual world"? Inside the child's mind. There is no virtual world on the games console, any more than there is a virtual pear hidden in the TCP/IP packet that carried the word "pear" across the internet.

The relative complexity of the Mario game can fool us into thinking that there's something going on inside it. Generally this involves a certain level of computer sophistication, but falls short of really knowing how the thing works. There might be dozens of independent data structures representing "Mario", effectively unconnected. Much of the information doesn't even lie inherently in the program. The colours that appear on the screen are arbitrary numbers passed to a different module.

It doesn't matter how complicated the interactions are - the only place the virtual world exists is in the mind of a human being interacting with it. Look at it in the context of the program running on a computer, and it is just a succession of bit patterns. No matter how many bits, and how many different patterns, it has no meaning until a person comes along, any more than the 32-bit number being passed through a router thinks that it's a pear.

 

[westprog]

Only in the sense that "real tornado" idiomatically refers to a particular weather pattern. It is not true that "simulated tornado" is not a real thing, and is also not true that "simulated tornado" does not have particular kinds of effects.

Yes, it has effects on the consciousness of the climate scientist interpreting the results. That's the essential element that make it a tornado simulation. If nobody looks at it, it isn't a tornado simulation - it's a computer emitting heat as it changes the patterns on its RAM chips.

 

[westprog]

So because the brain does one thing it means it can't do another?

I'm looking at what I wrote and what you wrote and trying to find some kind of connection between them and I'm unable to do so.

 

[westprog]

Sure it can. But as a definition of consciousness it’s rather crappy.

Although it’s slightly less crappy than: consciousness = "whatever stuff does".

It's a good basis for neurological research. It's a bad basis for creating consciousness on a computer. As I said before, it's not really a "definition" - more an indication of where to look to find a definition. It seems to make sense to look where we know it happens.

 

[!Kaggen]

OK, but in that case it means it cannot actually be done. I’m referring to the quote below -




- I’m suggesting that it cannot actually be done, because it’s prevented by the fact that it would mean materially changing the properties of particles and fields in the donor world. The process would be self defeating.

Anyone can of course conceive of anything as a mater purely of human imagination. But that does not mean it can actually be done in any universe that we know of. In this case, I think that it cannot work … it will not actually be possible to “ simulate every single particle in their world, using rules identical to the rules of particle interaction in our own world identical rules “.

You could take an idea into another world as a "set of rules", but that alone won’t produce any particles at all. You would need to actually transport real properties of the fields and particles into that other world ... but doing that would destroy the very same particles and fields in the donor world. That’s why it won’t work.

RD will call dualism.

 

[westprog]

RD will call dualism.

I'm trying to get my head around the idea that if you don't believe in these virtual worlds, then you're a dualist.

 

[!Kaggen]

That's the important point. If the kid isn't pressing the button, or watching the screen, what does the game seem like to Mario?

Let's cut this right down to a very simple virtual world - the word "pear". In order to communicate this information to the people reading this message, I need just 32 bits. Four eight-bit ASCII characters. However, the people reading it will extract an enormous amount of information from those 32 bits. Is that information concealed in that small number? Obviously not. The number is just a way that my concept of "pear" can be transferred. What does the pear seem like to the number? The very question is meaningless. There is no virtual pear inside the number. The pear exists in my consciousness, and I'm passing that message to the people reading this via a number which decodes into the symbols that appear on the screen as "pear". This triggers associations. While someone else's association will not be exactly the same as mine, it will still be close enough for meaningful communication to take place.

How about the Mario game? It's effectively no different. The designers of the game are passing information to the child playing the game. The information is passed as a succession of numbers, which, just as the representation of the word "pear" on someone's browser, are represented as sensory data - visual and auditory. Where is the "virtual world"? Inside the child's mind. There is no virtual world on the games console, any more than there is a virtual pear hidden in the TCP/IP packet that carried the word "pear" across the internet.

The relative complexity of the Mario game can fool us into thinking that there's something going on inside it. Generally this involves a certain level of computer sophistication, but falls short of really knowing how the thing works. There might be dozens of independent data structures representing "Mario", effectively unconnected. Much of the information doesn't even lie inherently in the program. The colours that appear on the screen are arbitrary numbers passed to a different module.

It doesn't matter how complicated the interactions are - the only place the virtual world exists is in the mind of a human being interacting with it. Look at it in the context of the program running on a computer, and it is just a succession of bit patterns. No matter how many bits, and how many different patterns, it has no meaning until a person comes along, any more than the 32-bit number being passed through a router thinks that it's a pear.

Hmmm.. don't spoil RD's vision of manipulating little kids minds through the computer screen into believing what they see is real.

 

[westprog]

Hmmm.. don't spoil RD's vision of manipulating little kids minds through the computer screen into believing what they see is real.

Very, very small children might think that Mario lives inside their Wii, but they pretty soon figure it out.

 

[!Kaggen]

Very, very small children might think that Mario lives inside their Wii, but they pretty soon figure it out.

So that explains Pixy Misa's thoughts

 

[Belz...]

I repeated and highlighted the things it contradicts. The simulation itself is coherent. The "world of simulation" does not exist only in the mind of the perceiver of output. And it is not imaginary.
You're trying to say that a simulated tornado doesn't exist because it cannot blow down your house. That's wrong. It exists simply because you're really simulating it. It is true that it cannot blow down your house, but I would hope that if the real tornado could, then your simulation would really blow down your simulated house.

It's also true that if you add a simulation of your simulation of a tornado to your simulation of a tornado, then the simulation of your simulation of a tornado cannot blow down your simulation of a house; if this weren't the case, you wouldn't be simulating your simulation. So the way you're making your point doesn't entail what you think it does either.

Finally, a tornado really will blow down your house, most likely. But this tornado we're talking about is a hypothetical tornado, and your house is probably still standing while we're talking about it. The hypothetical tornado itself is a type of simulation. This waters down your point. Furthermore, any voluntary action you perform entails an intent, and a prediction--specifically, you're predicting that the thing you do will result in particular "controlled" body movements which will accomplish your intent; your sense of control in a voluntary act is derived from the correlation between the observed results and this prediction (in particular, you tend to note immediately that "something is wrong" on a conscious level if these things diverge significantly, especially if your body doesn't react the way you predict it should). And this sort of prediction is not only a form of simulation as well, but it's the main point of having concepts of reality in the first place. Your notions of reality are simulations. So you're not even properly contrasting simulations of the brain with the brain.

An interesting point.