Reply
 
Thread Tools Display Modes
  #51  
Old 04-13-2020, 06:19 PM
TriPolar is online now
Guest
 
Join Date: Oct 2007
Location: rhode island
Posts: 42,217
Quote:
Originally Posted by Crane View Post
Tripolar #42,

Allow me to define an aspect of consciousness as a point of discussion:

Consciousness includes the instantaneous act of summing immediate sensory input to evaluate the availability of resources
and the level of threats. The result of this conscious act is a gradient of feelings from urgency -> complacency -> satisfaction.


To some degree a computer can simulate these acts. And the computer can invoke an annunciator to indicate the position it has calculated on the result gradient. But, at what point does the computer adder ever see (know) more than the two values at its inputs and the value at its output. Does it 'know' the difference between a value and an address? Does it know whether the number will be go to memory location or a timer or a peripheral? They are all addresses in the memory map and the adder never 'sees' the map. The adder has no means to be aware of is context.
Your concept of the working of computers is too far from reality. The computer is a machine running a process which does much more than the basic operations of a computer. Just as an automobile is much more than a single piston in it's engine.

Quote:
An analog circuit comes closer. An operational amplifier with a bank of input resistors, each connected to a sensor, and a single feedback resistor will sum the inputs and provide an output for the result gradient. The operation is continuous and the inputs are concurrent. This is a small part of the operation of a neuron. An argument could be made that the operational amplifier is conscious of its input and operation, because the op amp is a single component that 'sees' the entire operation concurrently. The same argument cannot be made for the adder. It sees nothing but arbitrary bits.
Nonsense. There is nothing an analog circuit can do that a digital one cannot with a high degree of resolution to the point where the outputs are indistinguishable insofar as consciousness is concerned.

Last edited by TriPolar; 04-13-2020 at 06:23 PM.
  #52  
Old 04-13-2020, 06:27 PM
TriPolar is online now
Guest
 
Join Date: Oct 2007
Location: rhode island
Posts: 42,217
Missed the edit window:

The continuous nature of a changing analog signal does not magically produce consciousness.
  #53  
Old 04-13-2020, 10:57 PM
Crane is offline
Guest
 
Join Date: Jan 2011
Location: New Mexico
Posts: 1,324
Tripolar,

You are correct. I am using the term computer to mean only the Central Processing Unit (CPU).

"The computer is a machine running a process which does much more than the basic operations of a computer. Just as an automobile is much more than a single piston in it's engine."-Tripolar#51

Computer based systems accomplish complex tasks using the basic operations of a CPU. The complex tasks to be accomplished are not defined at the time of manufacture of the CPU. The CPU is only capable of manipulating binary numbers in the memory map. Some of the memory addresses are used for input devices and some for output. The CPU doesn't make any distinction between sensors, memory and output devices. It's all memory and it's all numbers. The CPU may be attached to a complex system that drives a car. The system generates numbers describing the status of the car and requires numbers in response that result in driving the car. The CPU reads from and writes to the memory locations under control of the program. Any significance assigned to the values in memory is provided only by the program and the attached system The CPU executes the program in memory. The complex system drives the car. The same CPU does exactly the same thing in a toy or appliance. The attached complex systems are different - the CPUs are not.

You are correct. The CPU can serially simulate the operational amplifier. But the CPU never sees all of the data simultaneously - in context. The continuous nature of the analogue signal is a component of consciousness. One could make an argument for consciousness in the op amp. The same is not true of the CPU.
  #54  
Old 04-14-2020, 12:17 AM
TriPolar is online now
Guest
 
Join Date: Oct 2007
Location: rhode island
Posts: 42,217
Quote:
Originally Posted by Crane View Post
You are correct. The CPU can serially simulate the operational amplifier. But the CPU never sees all of the data simultaneously - in context. The continuous nature of the analogue signal is a component of consciousness. One could make an argument for consciousness in the op amp. The same is not true of the CPU.
You have stated this but not made any argument for it. Why would an analog be any more conscious than a digital circuit that produces the same result?
  #55  
Old 04-14-2020, 10:10 AM
Crane is offline
Guest
 
Join Date: Jan 2011
Location: New Mexico
Posts: 1,324
1. The operational amplifier has some structural similarity to a neuron.

2. The operational amplifier is a single component that possesses neuronal characteristics.

3. The operational amplifier handles all instantaneous values simultaneously.

4. The operational amplifier operates on the data stream continuously

a. The CPU has no structural similarity to a neuron.

b. The CPU does not possess any neuronal characteristics

c. The CPU never 'sees' all of the data and does not distinguish between input and output

d. The CPU operates on the data in a serial pieces. It may be interrupted causing the operation to be completely discontinuous.

I do not believe, and have not argued, that an operational amplifier is conscious. It does however have some characteristics of a conscious system (as defined above). The CPU has none.

The illusion of consciousness created by computers is impressive. My printer constantly communicates with it's manufacturer. When it is low on ink it orders more and bills my credit card. I never see the transaction. I just get ink in the mail. Is my printer conscious.
  #56  
Old 04-14-2020, 10:26 AM
TriPolar is online now
Guest
 
Join Date: Oct 2007
Location: rhode island
Posts: 42,217
Quote:
Originally Posted by Crane View Post
1. The operational amplifier has some structural similarity to a neuron.
I'm stopping at this point in my response because that's another claim without explanation. I can simplify this.

Is a single neuron conscious? If you think it is you have a lot of explaining to do, please be precise and detailed in what you mean.

If a single neuron is not conscious then explain how multiple neurons can form a conscious entity. When you do that you will also have the explanation for how a digital computer can form a conscious entity.
  #57  
Old 04-14-2020, 10:38 AM
RaftPeople is offline
Guest
 
Join Date: Jan 2003
Location: 7-Eleven
Posts: 6,908
Quote:
Originally Posted by TriPolar View Post
Nonsense. There is nothing an analog circuit can do that a digital one cannot with a high degree of resolution to the point where the outputs are indistinguishable insofar as consciousness is concerned.
I wonder if this really is true.

I remember reading that some mathematician proved for the n body problem with 5 bodies, it's possible to accelerate one of the bodies to infinity in finite amount of time without a collision.

If this was your analog circuit, it seems like it has the ability to get into a state that can't be represented by a digital computer.
  #58  
Old 04-14-2020, 10:47 AM
RaftPeople is offline
Guest
 
Join Date: Jan 2003
Location: 7-Eleven
Posts: 6,908
Here's another angle on the digital simulation of analog:
What if the analog circuit can solve some problems in polynomial time and the digital simulation is in exponential time. It's possible that the digital version can't be effectively used for the size of some of the problems our brain solves, and maybe some of those are the ones that result in consciousness.
  #59  
Old 04-14-2020, 10:48 AM
TriPolar is online now
Guest
 
Join Date: Oct 2007
Location: rhode island
Posts: 42,217
Quote:
Originally Posted by RaftPeople View Post
I wonder if this really is true.

I remember reading that some mathematician proved for the n body problem with 5 bodies, it's possible to accelerate one of the bodies to infinity in finite amount of time without a collision.

If this was your analog circuit, it seems like it has the ability to get into a state that can't be represented by a digital computer.
I added that part about "insofar as consciousness is concerned". I can't rule it out, but I'm not convinced it's a factor in consciousness. You've commented on neural processing before, do you think there's something special about analog processing in regard to consciousness?

Quote:
Originally Posted by RaftPeople View Post
Here's another angle on the digital simulation of analog:
What if the analog circuit can solve some problems in polynomial time and the digital simulation is in exponential time. It's possible that the digital version can't be effectively used for the size of some of the problems our brain solves, and maybe some of those are the ones that result in consciousness.
As a practical matter that may turn out to be true without some advances in technology. I don't think it affects the definition of consciousness though.

Last edited by TriPolar; 04-14-2020 at 10:50 AM.
  #60  
Old 04-14-2020, 11:43 AM
Crane is offline
Guest
 
Join Date: Jan 2011
Location: New Mexico
Posts: 1,324
Ed,

Thanks for the thoughts.

"The operational amplifier has some structural similarity to a neuron."

The statement is true and I explained it above. The input from the sensors are similar to dendrites, the input resistor values act like the synapsis and the transfer function acts like the body of the neuron. I have built simple neural nets from op amps. Operational amplifiers are not intelligent and they are not neurons, but they are structurally similar. The CPU is not and that is the topic under discussion.

If you are considering a PC in a box rather than just the CPU then you have the problem of the program structure. The PC is driven by an interrupt stack. It's 'thoughts' are brief and chaotic.

So, my printer engages in conscious acts. Is my printer conscious?
  #61  
Old 04-14-2020, 12:28 PM
RaftPeople is offline
Guest
 
Join Date: Jan 2003
Location: 7-Eleven
Posts: 6,908
Quote:
Originally Posted by TriPolar View Post
I added that part about "insofar as consciousness is concerned". I can't rule it out, but I'm not convinced it's a factor in consciousness. You've commented on neural processing before, do you think there's something special about analog processing in regard to consciousness?
Not really. Logical transformation of information, whether by analog processes or digital processes, doesn't "feel" like it leads to consciousness.

I can see how it can lead to intelligence, which just seems like effective modeling of the environment allowing for good predictions and problem solving in support of some specific goals. I can see how flipping pieces of paper on the floor can produce output that I would consider "intelligent".

But I don't see how flipping pieces of paper can result in consciousness.
  #62  
Old 04-14-2020, 12:32 PM
TriPolar is online now
Guest
 
Join Date: Oct 2007
Location: rhode island
Posts: 42,217
Quote:
Originally Posted by Crane View Post
"The operational amplifier has some structural similarity to a neuron."
Sure, you can find them analogous in certain ways.

Quote:

If you are considering a PC in a box rather than just the CPU then you have the problem of the program structure. The PC is driven by an interrupt stack. It's 'thoughts' are brief and chaotic.
You should stop trying to describe how computers work, you don't have the details right at all, nor does it matter any more than describing how atoms work has anything to do with the capabilities of an op amp in a larger circuit, or in the neurons in our brains.

Quote:
So, my printer engages in conscious acts. Is my printer conscious?
In my minority opinion, your printer, within it's own environment, is close to conscious. How close? I'm not sure, it might even be conscious by my own definition and the given constraint.

Not knowing the specifics of your printer I'll use my computer as an example. It can tell whether it is running on battery or is plugged in. It can tell how much battery it has left to use, what the CPU temperature is, what the ambient light level in the room is, and some general concepts of how much power other peripheral components use. And it is also keeping track of how much I use it. When it goes on battery power it adjusts power consumption to maximize battery life according to established parameters and all of the various information it has, including the history of usage, so there is a subjective result based on the environment and experience.

I don't yet have a good definition for how much more it takes for my computer to be conscious within it's own simplistic environment, but I see that as a good starting point for examining the nature of consciousness. Clearly though, that extra something that goes beyond the basic processing of sensory input to make a conscious entity will be far more complex in a human level consciousness than in an artificially constrained one.
  #63  
Old 04-14-2020, 12:39 PM
Mijin's Avatar
Mijin is offline
Guest
 
Join Date: Feb 2006
Location: Shanghai
Posts: 9,718
Crane I think we're missing the woods for the trees here.

Even if we concede that a computer could never simulate the human brain due to some fact of it being digital or serial or whatever, we still don't have grounds for saying a computer can't be conscious. Because we have no reason to assume necessary requirements include things like being analogue.

Computers can perform lots of functions that previously could only be performed by human brains.
So it is insufficient to say "Human brains are conscious, computers are unlike human brains, therefore computers can never be conscious", the logic doesn't follow. Because we could have used the same logic to say, for example, that computers will never perform facial recognition.
  #64  
Old 04-14-2020, 04:16 PM
Crane is offline
Guest
 
Join Date: Jan 2011
Location: New Mexico
Posts: 1,324
Mijin,

Thanks for the comments. Sorry, perhaps I was more thinking out loud than constructing an argument.

We do need to identify what we mean by both 'computer' and 'consciousness'. Perhaps a good computer example is Watson. IBM intentionally took on an impossible task to demonstrate the ability of their system. It's ability to play Jeopardy was amazing. It not only played the game it was good at it. And, the game was played only by doing what humans do. Voice input/output, and Watson had to push the button, just like any other contestant to answer a question (statement). I know that Watson is made up of many individual computing modules. I have no idea how it is organized or programmed, but I do know that it uses the same CPU adder technology as all other numerical computers.

So, what is the definition of 'conscious' and does Watson meet it? It does not meet mine above.
Reply

Bookmarks

Thread Tools
Display Modes

Posting Rules
You may not post new threads
You may not post replies
You may not post attachments
You may not edit your posts

BB code is On
Smilies are On
[IMG] code is Off
HTML code is Off

Forum Jump


All times are GMT -5. The time now is 09:54 PM.

Powered by vBulletin® Version 3.8.7
Copyright ©2000 - 2020, vBulletin Solutions, Inc.

Send questions for Cecil Adams to: cecil@straightdope.com

Send comments about this website to: webmaster@straightdope.com

Terms of Use / Privacy Policy

Advertise on the Straight Dope!
(Your direct line to thousands of the smartest, hippest people on the planet, plus a few total dipsticks.)

Copyright 2019 STM Reader, LLC.

 
Copyright © 2017