Simulating a brain on a computer

[David Bailey]

I am actually a bit stunned by now - I originally thought I was saying something obvious, but nobody seems to see what I am saying - so maybe the non-physical part of my brain is coming loose or something.......

The mathematical world is full of relationships - such as 22+17=39 - that are true for all time - they are, if you like micro-theorems. Now what I am trying to point out is that for specific values of input,program, and output, the relationship input=>program=>output is exactly the same sort of object!!

You could tell my mini program gave the specified output just by looking at it. I could have added a loop (say), and I would still have had a mathematical relationship, and maybe you could have figured out the output, or maybe not. Now I could add 10000 loops and recursions (or whatever) and create the gedanken simulation program, but it would still be the same kind of object!!!

If you claim a program has full consciousness, the computer is kind of irrelevant - that relationship input/P/output is out there as a mathematical fact - like Pythagoras' theorem or whatever. The entire half hour brain simulation is just one more mathematical fact.

Looked at from this point of view, running the computer program is just a way to check the the theorem.

Where does your machine consciousness lie - in the computer, or in the corresponding mathematical relationship that was true in the time of the dinosaurs and will still be true when the sun goes red giant.

If anyone else sees what I am getting at, please, please say so!

David

[Paul C. Anagnostopoulos]

Originally Posted by David Bailey

The mathematical world is full of relationships - such as 22+17=39 - that are true for all time - they are, if you like micro-theorems. Now what I am trying to point out is that for specific values of input,program, and output, the relationship input=>program=>output is exactly the same sort of object!!

Agreed. Of course, to state the theorem we need to know the output beforehand.

If you claim a program has full consciousness, the computer is kind of irrelevant - that relationship input/P/output is out there as a mathematical fact - like Pythagoras' theorem or whatever. The entire half hour brain simulation is just one more mathematical fact.

Assuming no truly random inputs, then the input/P/output is a mathematical fact, yes.

Looked at from this point of view, running the computer program is just a way to check the the theorem.

If we know the output beforehand, and all we want to do is check the theorem, then yes.

Where does your machine consciousness lie - in the computer, or in the corresponding mathematical relationship that was true in the time of the dinosaurs and will still be true when the sun goes red giant.

In the execution of the program.

~~ Paul

[David Bailey]

Originally Posted by Paul C. Anagnostopoulos

Agreed. Of course, to state the theorem we need to know the output beforehand.


Assuming no truly random inputs, then the input/P/output is a mathematical fact, yes.


If we know the output beforehand, and all we want to do is check the theorem, then yes.


In the execution of the program.

~~ Paul

Great - I think at last we are talking on the same wavelength!

Is the question as to whether the theorem is known beforehand very important? If it is, how does that alter the status of the re-run of the computer simulation of 30 mins of pain - when the outcome is already known?

So only the execution of the program can actually experience the pain - but how do you show that? What about computers made of clockwork, or even bureaucratic computers (run by people) - do they feel pain? Why exactly does a gadget running through all the steps that prove the 'pain theorem' cause that pain to be felt?

Of the various issues you have raised, which are vital to the question of whether the 30 mins of pain is actually endured?

David

[Interesting Ian]

Originally Posted by David Bailey

If anyone else sees what I am getting at, please, please say so!

David

I think I do -- more or less.

[Interesting Ian]

Originally Posted by David Bailey

Great - I think at last we are talking on the same wavelength!

Is the question as to whether the theorem is known beforehand very important?

It would be completely irrelevant wouldn't it?

So only the execution of the program can actually experience the pain - but how do you show that?

I suppose they would say that the brain is essentially a super sophisticated computer. The brain experiences pain. Therefore the execution of an appropriate program must experience pain.

What about computers made of clockwork, or even bureaucratic computers (run by people) - do they feel pain?

Yes, absolutely anything that can reproduce the function associated with pain will experience that pain. It could be some device consisting of tin cans, or whatever.

Why exactly does a gadget running through all the steps that prove the 'pain theorem' cause that pain to be felt?

It's just a brute fact about the Universe.

[Interesting Ian]

Originally Posted by Interesting Ian

DB
Why exactly does a gadget running through all the steps that prove the 'pain theorem' cause that pain to be felt?

II
It's just a brute fact about the Universe.

No that's wrong. Functionalism is similar to behaviourism. In behaviourism the pain literally is saying ouch, ones face contorting etc. In functionalism it is the carrying out of the process itself which is pain. It is the execution of the program that literally is the pain.

[davidsmith73]

Originally Posted by Paul C. Anagnostopoulos

Now my question to you: If you think that a mind cannot be simulated on a computer with appropriate random inputs, then I believe you are saying that a mind is more than some combination of deterministic and random events. Could you please describe the third thing that is included? If it is easier, you can describe the third thing that is involved in libertarian free will.

Ok, this is just stuff I haven't really thought through properly but let me know what you think.

We shouldn't be focussing on where the mind fits into the physical world, but how physical things fit into the mind. It's not that the mind is more than a combination of deterministic and random events as if it is something extraneous to those things. It could be that all physical events are relationships completely contained and constructed from qualitative experiences. Consider any "physical thing". Surely, it must consist of a quantitative relationship between two or more variables, ie, how two or more "things" vary with respect to each other. So after we have exhaustively described this relationship in terms of quantities, does it change the nature of the quantities if we conceive of the relationship as being between this quality or that quality? In other words, do the dimensions of a square care whether they are red or blue? I say not. But they have to have some quality to them otherwise there is no relationship to observe; there are no "things" that actually vary between one another. You may be thinking that the variables in any physical relationship are just physical entities. If so, then what are the variables that constitute the relationship of those physical things? And so on, ad infinitum.

So if there can't be physical reality without qualitative reality then the computer simulation thought experiment doesn't even get off the ground. In other words, imagining there are circumstances where a string of logical commands running through a bunch of computer chips are not conscious is an illusion. No physical thing suddenly "gets to be consicous". Physical things are contructed from consciousness.

How I don't know

[David Bailey]

Originally Posted by Interesting Ian

No that's wrong. Functionalism is similar to behaviourism. In behaviourism the pain literally is saying ouch, ones face contorting etc. In functionalism it is the carrying out of the process itself which is pain. It is the execution of the program that literally is the pain.

This is exactly why I feel this computer gedanken experiment has so much to tell us! Thinking about computers can clean up a lot of fuzzy thinking. For example, it would be completely trivial to write a program that printed "Ouch!" every time the left mouse button was pressed. Would in be feeling pain - I doubt it!

More generally, is it reasonable to expect a computer program to have any side effects - like feeling pain - as it grinds its way to an answer.

Ian, can I take it that when you replied "Yes, absolutely anything that can reproduce the function associated with pain will experience that pain. It could be some device consisting of tin cans, or whatever.", you were answering on behalf of the materialist viewpoint - not your own belief?

People have to choose between accepting the idea that a computer simulation of a brain (or entire body) would be equivalent to the real thing, or rejecting the idea (my position). It is interesting to note that a brain based on the TV receiver analogy would not be capable of simulation, because presumably the computer would be unable to link up to the mental components.

David

[Paul C. Anagnostopoulos]

Originally Posted by Interesting Ian

Yes, absolutely anything that can reproduce the function associated with pain will experience that pain. It could be some device consisting of tin cans, or whatever.

Not necessarily. I don't think it has to be the case that any mechanism for computing the outputs from the inputs would be conscious. It might be that only certain mechanisms, with certain physical properties, would be conscious.

~~ Paul

[Paul C. Anagnostopoulos]

Originally Posted by davidsmith

So if there can't be physical reality without qualitative reality then the computer simulation thought experiment doesn't even get off the ground. In other words, imagining there are circumstances where a string of logical commands running through a bunch of computer chips are not conscious is an illusion. No physical thing suddenly "gets to be consicous". Physical things are contructed from consciousness.

Constructed from what consciousness? It can't be my consciousness,* because thee is clear evidence that the coherency of physical things is maintained across my being conscious of them (this is why solipsism is silly). So it has to be some kind of universal consciousness metamind sort of thing. In that case, how can we tell the difference between everything being consciousness, being physical, or being pink jello?

~~ Paul

* In other words, physical things aren't simply figments of my awareness.