Why to such a deep degree. If I have a computation that runs on a PA-RISC architecture then I can run an equivalent computation of an Intel chip without simulating the subatomic particles of the PA-RISC processor. So if the brain is running an algorithm then I ought to be able to run an equivalent algorithm on a register machine without doing it down to sub-atomic particle level, or even to neural level.
You could certainly do it with less, but I don't know if it is known exactly what effect quantum-level interactions have on the brain, so it's
conceptually simpler to go for a complete physical simulation than a simulation of a more abstract but much simpler neural-net that processes information in the same way.
(Conceptually simpler, but practically much more difficult. But for a thought experiment, physical difficulty is irrelevant.)
I think that describing the brain as a computer is a little misleading, because when people use the term "computer" they usually mean a
stored-program computer or
Turing machine, but the brain is nothing like these things.
It's more like an
analogue computer, in that the processing of information is hardwired, not the result of following a set of programmed instructions, so you can't just use an interpreter or recompile the code to run the program on a different machine, because there is no program to interpret or recompile.
You have to either simulate or emulate the machine instead. In the same way, you have to simulate or emulate the brain to reproduce it on a digital computer.
But somehow it is going to participate with a series of similar events which will happen in the future to be a sensation of nausea. What it in nature that will link this event in the past to events in the future to produce this sensation?
The number in the register will influence the future steps of the instructions.
It's my understanding that consciousness is a
process performed by the brain. It is the process that experiences sensations. The sensations are variations in how the process proceeds.
There's no sensation in the units of information being processed or the physical structure performing the processing, but the process
itself is the sensation.
Why should this process do any more than we see it doing, placing a collection of encoded numbers on storage?
From an external viewpoint, it doesn't.
But the process itself doesn't "see" the encoded numbers (or electric charges, or chemical states) that it is based on. It only "sees" what it is processing, such as input from sensory apparatus (such as eyes, ears, skin) and internal abstractions.
This sensation of nausea which will be produced, somehow, maybe over millions of years by this processing being completed by others, maybe being moved to different computers - what is it exactly? Is it the output of the computation? Or something to do with the details of the computation itself?
It's a variation in the process being performed by the computation.
I still don't really understand the claim.
It is a rather confusing and tenuous concept when you look into it.
I'll admit I only barely understand it myself, and I don't think I could even come close to explaining it in a way that can be easily understood by someone who hasn't already managed to come up with an idea of how it works themselves.
Maybe it might help if you start wondering how consciousness can arise in the brain, if all the brain is doing is moving chemicals and electrons around.
