• Quick note - the problem with Youtube videos not embedding on the forum appears to have been fixed, thanks to ZiprHead. If you do still see problems let me know.

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

westprog said:
Then it's internals carry out the same process of converting light into electrical signals, and converting those electrical signals into a standard format.
Click to expand...
Hey, Westprog! Guess what? You just described the retina and the optic nerve!

Isolation and insulation are the principles of computer architecture.
Click to expand...
Computer systems are generally modular, though this is out of practicality rather than any sort of intrinsic requirement.

They are not the principles of the central nervous system.
Click to expand...
Funny, then, how the retina connects to the optic nerve which connects to the primary visual cortex, which connects to the prestriate cortex, each of which performs a separate function in processing visual perception.

As always, you are wrong about how computers work, and you are wrong about how the brain works.
 
RandFan said:
What is isolation and insulation? Are you saying that the eye doesn't convert light waves into a signal?
Click to expand...
He's saying that computers are strictly, rigidly modular, with no blurring of layers (shades of the OSI 7-layer protocol stack!) whereas the brain is one big undiscovered undifferentiated blob.

The one tiny flaw is that neither of these assertions has any connection with reality.
 
westprog said:
Isolation and insulation are the principles of computer architecture.
Click to expand...

They are not. Look at the design of a Turing-machine; it's not at all modular.

Of course, it's hard to work with raw Turing machines, which is why computer architects tend to prefer systems that can be decomposed into independently designed and tested -- isolated -- modules. But the whole point of Turing's theorem was that that his design was provably "universal" and as such could and would do anything that could be done by a more complicated system, and the Turing machine was easier to prove things about.

It's rather like the design of an instruction set; it's provably possible to build a computer with only two instructions in its set -- "increment memory location" and "decrement memory location and branch if zero." Of course, no actual computer uses a set this small; even microcontrollers tend to have dozens of instructions and a Pentium has thousands. But the two-instruction computer can provably do anything the Pentium can.
 
Last edited:
westprog said:
Then it's internals carry out the same process of converting light into electrical signals, and converting those electrical signals into a standard format.
Click to expand...
The same thing happens in the eye.
Isolation and insulation are the principles of computer architecture. They are not the principles of the central nervous system.
Click to expand...
The brain doesn't receive a direct signal telling it what L is doing, but the computer receives signals telling it what the R cells in the webcam is doing.

So it would appear that the brain is more isolated than the computer.
 
PixyMisa said:
He's saying that computers are strictly, rigidly modular, with no blurring of layers (shades of the OSI 7-layer protocol stack!) whereas the brain is one big undiscovered undifferentiated blob.

The one tiny flaw is that neither of these assertions has any connection with reality.
Click to expand...

Yeah. The fundamental difference between trout and fish is that trout are strictly arboreal, whereas fish, by contrast, live in trees.
 
RandFan said:
Why is that important? What is intrinsicly special about neurons?

But we have cardio bypass.

You need to identify what it is about the human brain that can't be replicated.
Click to expand...

There's nothing about the human brain which necessarily can't be replicated. What I'm saying is that you can't abstract certain functionality and be sure that what you are leaving out doesn't matter.

You are making unwarranted assumptions. I'm not saying anything conclusively. However, you are implying that there is something intrinsicly significant about the human brain that can't be or won't likely be replicated.
Click to expand...

Any functioning system cannot be arbitrarily changed without knowing what you're doing.

There's no point in assuming that the functionality of the brain wouldn't be changed by exchanging computers for neurons. We'd need to actually do it. The reason that we are able to do heart bypass is because

  1. We fully understand the process that is going on
  2. We've actually done it

neither of which apply to the brain.
 
Last edited:
westprog said:
There's no point in assuming that the functionality of the brain wouldn't be changed by exchanging computers for neurons. We'd need to actually do it. The reason that we are able to do heart bypass is because

  1. We fully understand the process that is going on
  2. We've actually done it

neither of which apply to the brain.
Click to expand...
Wrong. A Cardiopulmonary bypass machine or an artificial heart does not even closely replicated how the heart functions. But it achieves the same end effect.

Do you see the big hint there?
 
drkitten said:
which is why computer architects tend to prefer systems that can be decomposed into independently designed and tested -- isolated -- modules.
Click to expand...

I would like to add that from a software point of view, it isn't even natural to make things like that. We professional software engineers spend most of our time trying to find modular solutions to problems for which a tightly coupled solution is straightforward because of exactly what drkitten says -- it has engineering advantages, not functional ones.

There is no reason at all that such structure would be somehow inherent in computing and in fact if you don't constantly fight it such structure constantly disappears so if anything it is the opposite. Almost every day I have to change swathes of my own code because it is so nasty that nobody else (and even myself, on rare occaisions) would be able to debug it.
 
Last edited:
westprog said:
There's nothing about the human brain which necessarily can't be replicated. What I'm saying is that you can't abstract certain functionality and be sure that what you are leaving out doesn't matter.
Click to expand...
What is it that is being left out?

Any functioning system cannot be arbitrarily changed without knowing what your doing.
Click to expand...
How is it being arbitrarily changed?

There's no point in assuming that the functionality of the brain wouldn't be changed by exchanging computers for neurons. We'd need to actually do it. The reason that we are able to do heart bypass is because

  1. We fully understand the process that is going on
  2. We've actually done it
neither of which apply to the brain.
Click to expand...
You are making a lot of unwaranted assumptions. You seem to suggest that we will always be completly in the dark unless we fully understand everything. I think that is nonsese. Like the German engineers who reverse engineered rader we can learn about the brain while we are reverse engineering it. And in fact we are doing just that.

 
rocketdodger said:
We professional software engineers spend most of our time trying to find modular solutions to problems for which a tightly coupled solution is straightforward because of exactly what drkitten says -- it has engineering advantages, not functional ones.
Click to expand...
Yeah... modularity makes it easier for (ahem) brains to understand, modify, and extend software.

Not that this proves nothing in itself, but I think it's quite interesting to note.
 
Last edited:
rocketdodger said:
I would like to add that from a software point of view, it isn't even natural to make things like that. We professional software engineers spend most of our time trying to find modular solutions to problems for which a tightly coupled solution is straightforward because of exactly what drkitten says -- it has engineering advantages, not functional ones.
Click to expand...

Depends on the scale on which you work, I suppose. I suspect that most "professional software engineers" are solving problems for which a tightly coupled solution is not only not straightforward, but not even possible, because there's no way that a single human engineer could keep track of the necessary sagans of lines of code. Somewhere in a corner office in Microsoft is a person who is saying "we really need to increase disk latency" and is hiring a person to work just on disk latency issues. Or, more accurately, is hiring a person who is going to hire and head up a team to work just on disk latency issues. He's probably not going to know or care much about the planned improvements to DirectX 75.0, because some other person is being hired to head up that team.

Modular design is a gift to top-down management; the idea that systems could spontaneously self-assemble and sort themselves into a working ensemble was revolutionary when it was proposed in biology and still hasn't made it into management thinking.

.... but of course, that's more or less exactly what the brain does. Given the muddy solutions that evolution usually comes up with, it's a wonder that the brain is as modular as it is.
 
drkitten said:
I suspect that most "professional software engineers" are solving problems for which a tightly coupled solution is not only not straightforward, but not even possible, because there's no way that a single human engineer could keep track of the necessary sagans of lines of code.
Click to expand...

Err, that is exactly what I mean.

I don't think "straightforward" implies that it would actually be easier, or even simpler. I think it just implies "easy for a human to understand the process immediately."

Conceptually, all software problems have a trivial straightforward solution -- a gigantic switch construct (also known as "20 questions code" in the industry).
 
Dancing David said:
!Kaggen said:
Exactly.

When we define consciousness as the ideal result of a material process then we end up with an material world defined ideally.

When we define consciousness as the material result of an ideal we end up with a ideal world defined materially.

We need to outgrow the limitations of our language.
Click to expand...

Or just label the behaviors and not worry about the impossible to decide. (IE both materialism and idealism can not be distinguished.)
Click to expand...

I assume you mean tell the difference empirically?


P.S. I am still thinking about your last two posts, have not forgotten :)
 
drkitten said:
Somewhere in a corner office in Microsoft is a person who is saying "we really need to increase disk latency" and is hiring a person to work just on disk latency issues.
Click to expand...
Yeah, that sounds like Microsoft allright.
 
paximperium said:
Wrong. A Cardiopulmonary bypass machine or an artificial heart does not even closely replicated how the heart functions. But it achieves the same end effect.

Do you see the big hint there?
Click to expand...

I don't think you realise what a functional specification is. The point of a functional specification is not to describe what happens, but to describe what function is needed to be fullfilled. A functional description of the heart tells us what the heart is required to do - hence it's possible to replace it with a pump.

In order to replace the brain with an entirely different system, we need to be able to accurately define the function that needs to be fullfilled.
 
RandFan said:
What is it that is being left out?
Click to expand...

Whatever is different between a neuron and a computer. As it is at present entirely impossible to replace a neuron with a computer, I think that's still something of an issue.

How is it being arbitrarily changed?

You are making a lot of unwaranted assumptions. You seem to suggest that we will always be completly in the dark unless we fully understand everything. I think that is nonsese. Like the German engineers who reverse engineered rader we can learn about the brain while we are reverse engineering it. And in fact we are doing just that.

Click to expand...

I've already said that what is needed is a functional specification. That's not the same thing as understanding everything. We can view the brain as a black box, but if we comprehend what it does, we can in theory duplicate it. That seems to be the reverse of what is being attempted with the neural net approach, where we don't fully understand what the brain is doing, but we try to duplicate the general layout.
 
You must log in or register to reply here.

Back
Top Bottom