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

[Paul C. Anagnostopoulos]

So you would have to assert that your experience right at this moment could be the result of billions of numbers written on paper with pencils.

Pixy says yes, without question. Do you say yes, without question?

I would be foolish to say yes given the miniscule amount of thought I've given the question. That said, yes.

~~ Paul

 

[Paul C. Anagnostopoulos]

The Algorithmic theory of consciousness asserts that the entire experience is due to the execution of the algorithm. Interaction with the environment, time, the physical means of executing the algorithm - all are totally irrelevant.

Interaction with the environment isn't totally irrelevant. It has to be simulated.

I'm not sure how many theorists actually believe this. It's a very immaterial sort of materialism. There's also a complete absence of explanation in the sense of cause and effect. Consciousness just happens when you run the algorithm. It's an "emergent property". And if you have doubts about this, it means you believe in magic, apparently.

Consider consciousness to be an internal behavior of the organism, similar to its external behaviors. The cause and effect arises when certain brain processes lead to certain internal behaviors. Consciousness doesn't "just happen."

~~ Paul

 

[Darat]

The utility of the idea is not in the likelihood that it's actually happening - it's in the concept that if an algorithm is equivalent to the algorithm that is executed in a human brain,

...snip...

I think you are misunderstanding something here - we are not talking about viewing the brain as a computer that runs an algorithm in the way that we "load" software into a computer but an algorithm that models the entirety of the human being.

 

[Paul C. Anagnostopoulos]

No. A Turing machine is deterministic.

In practice, real computers are able to produce effectively random numbers (something that is often a requirement) from arbitrary data, such as the current time. But the fact remains - it's part of the basic definition of a Turing machine that the same data ("tape") produces the same result, without fail, every time.

Why is this stopping us from simulating a coin toss to an arbitrary degree of randomness?

Since we can't rerun a human being, we can't determine how close the results would be to the first run of the human. But if it would be the same modulo randomness, then we can simulate those two runs on a Turing machine as long as we're allowed to preserve the random seed from one run to the next. If we aren't, then yes, the results would be identical. But what would be the point of placing that restriction on the machine?

If it turns out that we need a machine and a true RNG, so be it. I don't think that enhancement to the Turing machine model is particularly worrisome.

~~ Paul

 

[PixyMisa]

No.

Well, you're wrong.

A Turing machine is deterministic.

This is true. You're still wrong.

In practice, real computers are able to produce effectively random numbers (something that is often a requirement) from arbitrary data, such as the current time. But the fact remains - it's part of the basic definition of a Turing machine that the same data ("tape") produces the same result, without fail, every time.

Yes. You're still wrong.

This has the effect that when we choose a Turing machine, we are also choosing all the decisions that it can possibly make.

We are not choosing those decisions in any meaningful sense. Cf. the halting problem.

Now, back to your original error. If you have a Turing machine simulate a coin toss, and then rewind the tape and play it again, the result will always be the same.

If you toss a coin in the Universe, and then rewind the Universe and... Oh, wait, you're talking complete nonsense, aren't you?

If you have a Turing machine simulate a toin coss, and then continue on to simulate another toin coss, and another and another, you won't be able to predict those coin tosses without running the program - Cf. the halting problem again.

Or, to put it rather more succinctly: Wrong.

 

[westprog]

Interaction with the environment isn't totally irrelevant. It has to be simulated.

But the "program" will have no interaction with its environment. It will simply have data. If the experience of consciousness is identical whether or not there is interaction with the environment, then the environment is irrelevant. There's just data, presented in one format or another.

In order for this to be a valid explanation, we have to assume sharp boundaries between the organism and the environment, so that we can extract the data processing component of the brain, snip off all the input channels, and replace them with suitable data inputs. I don't see this as reflecting how human beings actually work.

Consider consciousness to be an internal behavior of the organism, similar to its external behaviors. The cause and effect arises when certain brain processes lead to certain internal behaviors. Consciousness doesn't "just happen."

~~ Paul

Certain processes lead to certain internal behaviours? That sounds very close to "just happens" to me.

The only explanation as to why we need to assume that there is something called "consciousness" involved with algorithms is that there are human beings who claim to possess it. It explains nothing about the behaviour of algorithms.

 

[westprog]

I think you are misunderstanding something here - we are not talking about viewing the brain as a computer that runs an algorithm in the way that we "load" software into a computer but an algorithm that models the entirety of the human being.

Then we are describing a model not just of the supposed data processing of the human brain, but the actual physical state of the human body. How can this be equivalent?

 

[westprog]

Why is this stopping us from simulating a coin toss to an arbitrary degree of randomness?

Since we can't rerun a human being, we can't determine how close the results would be to the first run of the human. But if it would be the same modulo randomness, then we can simulate those two runs on a Turing machine as long as we're allowed to preserve the random seed from one run to the next. If we aren't, then yes, the results would be identical. But what would be the point of placing that restriction on the machine?

If it turns out that we need a machine and a true RNG, so be it. I don't think that enhancement to the Turing machine model is particularly worrisome.

~~ Paul

If you change the Turing model, then all the theory relating to how Turing machines work has to be dumped. Turing machines are deterministic. It's a fundamental characteristic.

If you think that a better model for the human mind would be a Turing machine with a built in RNG built in, then that's up to you.

 

[westprog]

Well, you're wrong.


This is true. You're still wrong.


Yes. You're still wrong.


We are not choosing those decisions in any meaningful sense. Cf. the halting problem.

Now, back to your original error. If you have a Turing machine simulate a coin toss, and then rewind the tape and play it again, the result will always be the same.

If you toss a coin in the Universe, and then rewind the Universe and... Oh, wait, you're talking complete nonsense, aren't you?

Why do you need to rewind the universe? You just throw the coin again.

This is in the context of making decisions. I said that for any given decision, assuming the same data, the Turing machine will always give the same result. The coin will not.

For some applications, this lack of determinacy is necessary. That's why we often use coins as random generators.

If you have a Turing machine simulate a toin coss, and then continue on to simulate another toin coss, and another and another, you won't be able to predict those coin tosses without running the program - Cf. the halting problem again.

What relevance the halting problem has to people making decisions is left to the class, I suppose. Perhaps a link to a big irrelevant Wikipaedia article at this stage?

Or, to put it rather more succinctly: Wrong.

That's good too.

 

[drkitten]

No. A Turing machine is deterministic.

So is a coin flip unless you want to get into quantum effects.

In practice, real computers are able to produce effectively random numbers (something that is often a requirement) from arbitrary data, such as the current time. But the fact remains - it's part of the basic definition of a Turing machine that the same data ("tape") produces the same result, without fail, every time.

This has the effect that when we choose a Turing machine, we are also choosing all the decisions that it can possibly make.

Not at all. If you simply define the "data" as the problem data plus the date-and-time at which it was run (which is easy enough to put in), you get effective randomness. If you define "data" as the problem data plus the output of a scintillation counter, you get quantum randomness.

 

[drkitten]

If you change the Turing model, then all the theory relating to how Turing machines work has to be dumped. Turing machines are deterministic. It's a fundamental characteristic.

The machines are deterministic. The data is not.

And, in fact, the question of whether the human mind is deterministic given fixed data is an interesting open question in philosophy. You may be familiar with it.

 

[rocketdodger]

I was Robin's 'one confident "yes"'.

Your statement was, as usual, emphatic, far-reaching, and factually false. (Hint: Algorithms do nothing without data.)

Actually, I was the first yes, and by far the most confident.

But of course this is only because I am a gamer and I relish the idea

 

[rocketdodger]

Since you don't have a mathematical definition of decision I think that's going to be a problem for a start. A mathematical definition for decidability is something different.

Wait .. are you seriously claiming that there might be a mathematical definition for decidability without there being a mathematical definition for decision?

Here is a clue -- you know the process they speak of when they define decidability? That is a decision.

 

[rocketdodger]

Then we are describing a model not just of the supposed data processing of the human brain, but the actual physical state of the human body. How can this be equivalent?

Because you have neurons in every portion of your body?

And in any case, why does it make a difference? If you can simulate the entire body and have consciousness arise, then I would strongly suspect that you can simulate just the head and have consciousness arise, since in the real world there are many people who are very conscious yet have entire chunks of their body missing.

 

[westprog]

Wait .. are you seriously claiming that there might be a mathematical definition for decidability without there being a mathematical definition for decision?

Here is a clue -- you know the process they speak of when they define decidability? That is a decision.

But that is not a definition.

 

[PixyMisa]

Then we are describing a model not just of the supposed data processing of the human brain, but the actual physical state of the human body. How can this be equivalent?

The brain receives data, Westprog. If you modelled a brain without modelling the data it receives, it wouldn't be much of a model, would it?

 

[PixyMisa]

Actually, I was the first yes, and by far the most confident.

But of course this is only because I am a gamer and I relish the idea

Okay, that's three yesses so far.

 

[RandFan]

Randfan, can we capitulate.

Looking at your post I guess not. Would that it were so.

You say that my argument is equivalent the the Chinese Room argument because colloquially "understanding" is more or less equivalent to "consciousness". Is that fair?

Yes, but it doesn't matter. I'm willing for the sake of argument forgo that premise.

I have said that these terms are not equivalent and in any case the CR argument has a different structure - it hinges on what the operatives do or do not understand whereas mine does not.

Which is meaningless to the original point.

Suppose I am about to go in for an operation.

Yes.

I understand my wallet exists when I put it in the valuables locker.

Yes.

Then I am unconscious for a period....

Whooaaaa... Stop, you can't keep going. Are you aware of your walet at that time? THAT'S the point.

...and when I wake up I understand that my wallet exists even before I have seen it.

This is caled memory and it is only works when you are conscious.

So colloquially I say that I still understand. I don't say that I stopped understanding and started understanding.

There's no need to keep using the word colloquialy. For sometime now I've said you don't need to accept my usage of the word. It's not central to my point. You are being boorish. Words are not laws of physics. They are simply a means to convey information. If you object to my usage of a word I'm ameniable to forgo the use of the word. I tried to explain it to you and you won't listen to me so fine. Let's move on. {sheesh}.

My intuitive understanding is that understanding persists during periods of unconsciousness.

I never said otherwise. This is a straw man you've invented.

Now the second part - are there ever moments when you don't understand what you are perceiving?

IT DEPENDS ON WHAT YOU MEAN BY UNDERSTAND? Do you understand that you are aware?

 

[RandFan]

I can only deal with what he says, not try to guess what he means.

This is being obtuse. It's not controversial at all. The point is trivially true. Have you ever heard of a thing called inference?

The objection I made was precisely that he had not clearly defined what he means by "understand".

HE IS TRYING TO REBUT AI.

He is saying that an AI system does not understand. That it has no intentionality.

To what end? Seriously. Here's a guy that you concede doesn't think that the computational model can explain consciousness. He sets out to rebut the Turing test WHICH IS SUPPOSED TO BE A TEST FOR ARTIFICIAL INTELLIGENCE and you want to argue that one doesn't have anything to do with the other.

Seriously?

He could have clarified to say "I mean consciousness" but he didn't.

In fact he later made a separate argument that consciousness was not computational, because nothing was intrinsically computational.

And you are unable to figure out what the hell was his point about the Chinese Room as it relates to consciousness? Unbelievable.

I don't know if you are just yanking my chain or if you took exception at the beginning and now just can't find a way out.

The Turing test is proposed to ostensibly test for AI.
Searle sets out to rebut the Turing Test.
Searle's purpose is to rebut AI.

This isn't controversial and Searle has not step forward to complain about the many millions of web pages and sources that link the Chinese Room to consciousness. He has never made a point that his argument is being misaplied. There's no reason to think his argument has been misaplied.

 

[Darat]

...snip...

In order for this to be a valid explanation, we have to assume sharp boundaries between the organism and the environment, so that we can extract the data processing component of the brain, snip off all the input channels, and replace them with suitable data inputs. I don't see this as reflecting how human beings actually work.

...snip...

Yet we see real-world examples of practical systems that shows this is in fact how human beings work, from cochlea implants, to embedded cameras, to system that can respond the other way around i.e. "mind controlled" devices.