0
   

The mind cannot be a computer. A proof via turing machines.

 
 
Reply Fri 12 Mar, 2010 12:24 am
The best way to understand a computer is to understand a turing machine.
The question really ought to be "Is the mind a turing machine?".

Strong AI is the claim that "the mind is a turing machine". There are problems if this is indeed the case. In the following, i will suppose strong AI is true so that i will derive a contradiction. This is to prove the oppose of what is claim.

fact 1: If the mind is a turing machine, then it has the same "power"/"comprehension" as your desktop PC.

fact 2: A turing machine ( TM) can simulate another turing machine. e.g:
For TM A, and TM B, and that A is turing equivalent to B, then A can simulate B, vise versa. This means, A can do what B can do, vise versa.

Let us derive a contradiction by supposing strong AI is true.

Let TM P be the mind/"conscious person".

Let TM C be a a desktop computer that can decide what P will do with with censory input w. Denote this by <P,w>. So, if C accepts <P,w>, then P accepts sensory input w. Else, P does not accept sensory input w.

We define the TM P with the following specification:

P=" on input w.
1. Via recursion theorm( Kleene's recursion theorem - Wikipedia, the free encyclopedia), obtain a description of P, or <P>.

2. Simulate C with input < P, w>

3. if C accepts <P,w>, then reject.
if C reject <P,w>, then accept.

"

Contradiction occur at 3, because if P accepts w, then P reject w.
If P reject w, then P accepts w.


So our initial assumption must be wrong. " Strong AI is true" is false, or there must not be a TM C that decides <P,w>.


If the mind is a computer, and there is another computer that could figure out what this mind is going to do with by simulating it with input w, then this leads to a contradiction. Thus The following premises must be falses:

1. The mind is a computer.
2. there is a computer that could simulate this mind with input w that wields some definite output in finite number of steps.


Reference:
Sipser, Michael, "introduction to the theory of computation".
Chapter 3, 4, 5, and section 6.1( recursion theorem)
  • Topic Stats
  • Top Replies
  • Link to this Topic
Type: Discussion • Score: 0 • Views: 1,024 • Replies: 7
No top replies

 
Jebediah
 
  1  
Reply Fri 12 Mar, 2010 12:36 am
@TuringEquivalent,
I don't understand the part where you say "contradiction occur at 3, because if P accepts w, then P reject w". You might be interested in the blue brain project though:

Blue Brain Project - Wikipedia, the free encyclopedia

It's an attempt to build a more brain like computer than a desktop pc.
TuringEquivalent
 
  1  
Reply Fri 12 Mar, 2010 12:51 am
@Jebediah,
Jebediah;138873 wrote:
I don't understand the part where you say "contradiction occur at 3, because if P accepts w, then P reject w". You might be interested in the blue brain project though:


I am quite sorry i wrote this thread. I thought it would be readable...
The actual thing i wrote in the op post is the actual proof of how you would do it in sipser `s book.

For example: Imagine a computer P with input w that has access to it` s own program, called this <P>. Within in P, P can simulate program <P> with input w. Now, imagine a computer C, that could sort of decide what P will do by input <P, w>. Now, Let P run C with input <P,w>. If C accepts, then we set P to reject the input w. If C reject, then we set p to accept the input. Thus, we have a contradiction.

Quote:
Blue Brain Project - Wikipedia, the free encyclopedia

It's an attempt to build a more brain like computer than a desktop pc.

Don` t matter, because Turing machines are the most powerful.

---------- Post added 03-12-2010 at 02:01 AM ----------

Let me explain this is more human terms:

Imagine you are about to figure out given some input information how "you" will react to "a car coming toward you", before you react to a car coming toward you 3 minutes later. Then, with this information, you decide to do the oppose. If you think hard, you know that this lead to a contradiction.

If you can predict what you will say 3 minutes from now, and you docided to do the oppose, then you must not be able to predict 3 minutes from now. contradiction
0 Replies
 
Jebediah
 
  1  
Reply Fri 12 Mar, 2010 08:26 am
@TuringEquivalent,
But I don't see why a computer couldn't predict what it would do in three minutes, and then do the opposite. If you had a separate process for each neuron and had copied the structure of the brain. That wouldn't be a Turing machine I suppose, and you could say it isn't even a computer.

But I don't know if that's important because I don't know what the significance of the argument is to you. People think in terms of analogies, and the "body as a machine/mind as a computer" is one way to do that.
Derek M
 
  1  
Reply Fri 19 Mar, 2010 03:13 pm
@TuringEquivalent,
TuringEquivalent wrote:
Imagine you are about to figure out given some input information how "you" will react to "a car coming toward you", before you react to a car coming toward you 3 minutes later. Then, with this information, you decide to do the oppose. If you think hard, you know that this lead to a contradiction.


A computer couldn't calculate what it was going to do in some odd amount of time, barring its making certain potentially incorrect assumptions. The reason being it would have to know in advance precisely what changes to the system would occur during the intervening period, including changes to the system made as a result of its own calculations. It seems to me, it would have to know what it was going to calculate to perform its calculations, and so what you describe seems impossible for computers as well as humans.
0 Replies
 
TuringEquivalent
 
  1  
Reply Sat 20 Mar, 2010 07:20 am
@Jebediah,
Jebediah;138943 wrote:
But I don't see why a computer couldn't predict what it would do in three minutes, and then do the opposite.



1. Suppose, the computer works, it does X 3 minutes from now.

2. The computer found out X, and instead does Y 3 minutes from now.

Suppose 1&2 are true, then my question to you is: does computer do X, or Y?





Quote:
"body as a machine/mind as a computer" is one way to do that.


Would that not entail the mind is a computational process? turing machine can model that.

---------- Post added 03-20-2010 at 08:32 AM ----------

Derek M;141324 wrote:
A computer couldn't calculate what it was going to do in some odd amount of time, barring its making certain potentially incorrect assumptions. The reason being it would have to know in advance precisely what changes to the system would occur during the intervening period, including changes to the system made as a result of its own calculations. It seems to me, it would have to know what it was going to calculate to perform its calculations, and so what you describe seems impossible for computers as well as humans.



That example is a concrete analogy( i use it because it is much more simple) , but it should not be taken too seriously. There is no appeal to time, processes, and "what it is going to do".

The complete mathematical proof for the undecidability of the halting problem using two turing machines, by feeding inputs to one another in such a way to produce a mathematical version of "Turing machine X halts on input w" if and only if " Turing machine X halt on input w".
0 Replies
 
Derek M
 
  1  
Reply Sat 20 Mar, 2010 02:45 pm
@TuringEquivalent,
TuringEquivalent wrote:
The complete mathematical proof for the undecidability of the halting problem using two turing machines, by feeding inputs to one another in such a way to produce a mathematical version of "Turing machine X halts on input w" if and only if " Turing machine X halt on input w".


I'm not seeing how the halting problem means human cognition isn't Turing-compatible.
TuringEquivalent
 
  1  
Reply Sat 20 Mar, 2010 03:07 pm
@Derek M,
Derek M;141665 wrote:
I'm not seeing how the halting problem means human cognition isn't Turing-compatible.



I think it is very much similar to godel theorm.
0 Replies
 
 

Related Topics

How can we be sure? - Discussion by Raishu-tensho
Proof of nonexistence of free will - Discussion by litewave
Destroy My Belief System, Please! - Discussion by Thomas
Star Wars in Philosophy. - Discussion by Logicus
Existence of Everything. - Discussion by Logicus
Is it better to be feared or loved? - Discussion by Black King
Paradigm shifts - Question by Cyracuz
 
  1. Forums
  2. » The mind cannot be a computer. A proof via turing machines.
Copyright © 2024 MadLab, LLC :: Terms of Service :: Privacy Policy :: Page generated in 0.03 seconds on 04/24/2024 at 09:45:44