Has consciousness been fully explained?

[Dymanic]

Slog? Slog?

It's a work of art! One does not slog through a work of art!

A valid objection. My intent was to convey that GEB demands an exceptionally high level of effort on the part of the reader, and while I would argue that some works of art do the same (Picasso's "Guernica" being one possible example), I will admit to having made an unfortunate choice of words. "Slog" implies drudgery, and GEB is anything but that.

Sorry, Piggy

 

[Meed]

Fans of fMRI should take a look at this.

http://www.wired.com/wiredscience/2009/09/fmrisalmon/#ixzz12LbIGKOr

Neuroscientist Craig Bennett purchased a whole Atlantic salmon, took it to a lab at Dartmouth, and put it into an fMRI machine used to study the brain. The beautiful fish was to be the lab’s test object as they worked out some new methods.

So, as the fish sat in the scanner, they showed it “a series of photographs depicting human individuals in social situations.” To maintain the rigor of the protocol (and perhaps because it was hilarious), the salmon, just like a human test subject, “was asked to determine what emotion the individual in the photo must have been experiencing.”

The salmon, as Bennett’s poster on the test dryly notes, “was not alive at the time of scanning.”

If that were all that had occurred, the salmon scanning would simply live on in Dartmouth lore as a “crowning achievement in terms of ridiculous objects to scan.” But the fish had a surprise in store. When they got around to analyzing the voxel (think: 3-D or “volumetric” pixel) data, the voxels representing the area where the salmon’s tiny brain sat showed evidence of activity. In the fMRI scan, it looked like the dead salmon was actually thinking about the pictures it had been shown.

“By complete, random chance, we found some voxels that were significant that just happened to be in the fish’s brain,” Bennett said. “And if I were a ridiculous researcher, I’d say, ‘A dead salmon perceiving humans can tell their emotional state.’”

The result is completely nuts — but that’s actually exactly the point. Bennett, who is now a post-doc at the University of California, Santa Barbara, and his adviser, George Wolford, wrote up the work as a warning about the dangers of false positives in fMRI data. They wanted to call attention to ways the field could improve its statistical methods.

You have to be careful about trying to learn about psychology and brain sciences from popular news articles and TED talks. They're very often misleading, as their purpose is to entertain and be sensational rather than to teach you the actual science. The only way to properly assess empirical results without the risk (or at least a decreased risk) of being mislead is to actually read the studies and to know something about statistics while doing so. For example, a study that attempts to predict what choices you'll make before you'll make them based on some hypothesized brain oxidization correlate may successfully predict the subjects' choices 55% of the time. This is not much higher than the success rate predicted by chance, however it may be a statistically significant degree above the level predicted by chance. Statistically significant != actually significant. Note also that when chance alone is operating a statistically significant result will still be found 1 out of every 20 times.

 

[westprog]

Let's consider the idea that a thermostat with a recording device (or some other such device) has, in some sense, a form of running. Evaluating this as a theory, I'd ask the following questions:

• How does the hypothesis of running help explain the behaviour of the system?
• How would we, even conceptually, measure this property?
• How would we quantify running in such a system?
• How would we test the hypothesis that such a device is running?
• How would this hypothetically running device differ from one that was hypothetically not running?
• Given the answers to the above questions, of what scientific application is this theory?

Yes, you've done this trick before. But it doesn't actually work in this case, because we can objectively look at something and tell whether running is present - we can measure it, we can quantify it, we can tell the difference between something that is or is not running - and while running is not a particularly scientific concept, it is certainly useful.

And while you can do the other trick of sticking in a verb in place of a noun to make something sound nonsensical, in this case it isn't fooling anyone.

 

[Myriad]

Let's consider the idea that a thermostat with a recording device (or some other such device) has, in some sense, a form of consciousness. Evaluating this as a theory, I'd ask the following questions:

I'm not willing to consider the idea that "a thermostat with a recording device" has any form of consciousness, since it does not meet all the characteristics I described:

Then, by my current working definition of conscious experience (the existence of self actions in a narrative constructed from memory and under evaluation), the thermostat would be conscious.

To meet that definition, the thermostat needs to be able to evaluate patterns of recent input by comparison (some form of statistical analysis, perhaps) against narratives of past events constructed from memory, said past events including the analysis that took place at that time and the thermostat's own actions. This also requires (though I didn't state explicitly before) that the results of past analysis affected the thermostat's past actions and so the results of its present analysis must affect its present actions as well, otherwise there would be no correlations to evaluate.

Since this processing must ultimately turn evaluation into decision, the system also needs goals relating the two. A reasonable set of goals for a thermostat would be (1) to maintain the temperature as close to a desired value as possible; (2) to use as little energy as possible; (3) to run the heating and/or cooling system on as long a duty cycle as possible (avoiding rapid cycling that increases wear and tear on the machinery). Since these three goals are contradictory (appropriate weighting functions would be built in), and the success or failure of any given decision in achieving goal #1 is based in part on factors outside the thermostat's direct knowledge or control (the outside temperature, wind, sunshine, people opening doors, heat-producing occupants and machinery, etc.) and yet those factors do have patterns that could in principle be discovered and adapted to (day/night, seasons, weekends, repeatable faults e.g. someone propping an exterior door open), there is plenty of potential evaluation to be done.

Now, we can address the questions:

How does the hypothesis of consciousness help explain the behaviour of the system?

It explains how the system is able to learn to act in anticipation of events. For example, if the temperature begins rapidly dropping it might turn on the heat "early" before the low limit of the best temperature range is reached, where an ordinary thermostat would not switch until the low limit actually occurs. But if you caused rapid cooling on purpose, then removed the cooling when the heat came on early, so that the anticipation resulted in overshooting the desirable temperature range, the system would soon stop responding that way.

How would we, even conceptually, measure this property?

The same limited ways we measure it in humans. How well is the system performing? How quickly and effectively does it react to deliberate anomalous input (aka "stimulation")?

How would we quantify consciousness in such a system?

Beats me. How do we quantify it in humans? (Or are humans not conscious? Sorry I have trouble keeping track of who holds the various points of view in this thread.)

How would we test the hypothesis that such a device is conscious?

Besides observing its performance and responses as above, we can put a test button on the thing. We push the button, and if the system is entering the necessary records into memory, and properly performing evaluations of past patterns versus present ones, a green LED turns on.

How would this hypothetically conscious device differ from one that was hypothetically not conscious?

We could test them side by side, for performance and reaction to stimuli. How about a three-way test: an ordinary thermostat, a system like the one above except that it is unable to include records of its own past analysis or actions in the memory (thus ruling out consciousness by my definition as memories that don't exist can't be included in the evaluation), and the complete system.

Given the answers to the above questions, of what scientific application is this theory?

Modeling the evolution of conscious experience and its related cognitive phenomena, identifying the benefits in terms of adapting to real environments and events.

More efficient thermostats (perhaps at the cost of making them, literally, temperamental).

Respectfully,
Myriad

 

[westprog]

I've followed it and I have to agree with westprog. It seems that you're misinterpreting his objections.

He's done this so often that I no longer attribute it to misunderstanding - because in each case I've explained at tedious length exactly what I am saying. At no stage have I claimed that a rock has exactly the same properties as a silicon chip, and I'm fairly confident that RD knows this. He'd rather argue with the strawman because when it comes to the specific claims that he's making, he's short on evidence.

In case there's any ambiguity - I'm saying that the properties that RD is attributing solely to the computer and the cell are, if we examine the processes carefully, also present in the rock. RD interprets this as saying that it is impossible to distinguish between the rock and the computer chip.

Thanks for the feedback, btw. It's more welcome because it's supportive of my POV, but it would be equally valuable if you disagreed. It helps to get external perspective on squabbles.

 

[westprog]

Yet

1) we could replace your brain with a turing machine that could control your body in exactly the same way your brain does ( with sufficient technology )

Among other things that the brain does, it is probably also a Turing machine, or perhaps it contains a number of Turing machines. A device that might replace the brain might also be a Turing machine. Could such a device replace the brain using only the functionality of a Turing machine? No. The definition of a Turing machine is fairly clear. It's also clear that the functions of the brain in controlling the body are not present in a Turing machine. A Turing machine doesn't interact with the real world. The human brain does.

There have been some claims that a practical implementation of a Turing machine would probably have some kind of mechanism built in which would sort of kind of be able to do the real-time control, somehow. Even if this were true, it's irrelevant. We are not in a position to create any such machine. We are considering what is necessary to build such a machine, as a thought experiment. The assertion is that an implementation of a Turing machine could replace the brain. The assertion is palpable wrong.Any such machine would have to have functionality which is not part of the Turing machine definition.

2) people are conscious at times when their brain is not controlling their body in any way that a reasonable person would consider as having anything to do with the consciousness.

Try again?

I'm impressed, in a way, that RD manages to squeeze so much invalid argument into such a compact space.

Firstly, RD's contention is as given above ("we could replace your brain with a turing machine"). The claim is not that the conscious functions of the brain can be duplicated by the Turing machine. It is that the Turing machine can control the body. We don't need to refer to consciousness, with its hazy definitions, to know that the brain's functionality is not and cannot be explained by the Turing model. That is something that is immediately apparent.

Secondly, while it's certainly the case that the brain and nervous system control the body in a way that does not involve consciousness - it is also the case that voluntary control functions do involve consciousness. We can pick up an object and be conscious of what we are doing. Thus the motor and sensory functions of the brain are not separate from consciousness - they are intricately entangled.

 

[westprog]

Wrong.

He has consistently made the claim that any turing equivalent system can be conscious.

He has never, ever, ever made any post to the effect that 'if a turing machine is just sitting there it will be conscious because it is a turing machine' or any of the other absurd strawmen that people are throwing around.

Why, when what I actually said is right there in the post, do you deliberately misrepresent it? I said that PM's position is that the consciousness of the brain springs from its function as a Turing machine. To equate that statement with "if a Turing machine is just sitting there it will be conscious" is astonishingly dishonest. Did you mean to delete what I'd actually said, and forget?

 

[westprog]

More efficient thermostats (perhaps at the cost of making them, literally, temperamental).

I've a feeling that we may be at cross purposes here, so before I have a go at your post can I clarify - do you think that devices such as the one you describe actually exist, or could possibly be created - or are you enumerating the benefits if a device such as a thermostat were to be made conscious in some way?

I don't want to end up arguing against a position that you do not, in fact, support.

 

[PixyMisa]

However, to give a straightforward answer: no, cruise controls do not compare present and immediately recent input with narrative reconstructed from memory that includes their self actions. So by my present understanding of the phenomenon of conscious experience, the cruise control on your car is not conscious.

Agreed. Cruise control may be implemented as a function of a computer system that we would consider conscious for other reasons, but in itself it's a regulatory feedback mechanism like a thermostat or a steam-engine governor. These can all be - and typically are - implemented as systems too simple for consciousness.

 

[rocketdodger]

I've followed it and I have to agree with westprog. It seems that you're misinterpreting his objections.

Can you explain what you mean? Perhaps you can phrase his objections in a way that makes more sense to me.

 

[rocketdodger]

Yes, you've done this trick before. But it doesn't actually work in this case, because we can objectively look at something and tell whether running is present - we can measure it, we can quantify it, we can tell the difference between something that is or is not running - and while running is not a particularly scientific concept, it is certainly useful.

And while you can do the other trick of sticking in a verb in place of a noun to make something sound nonsensical, in this case it isn't fooling anyone.

What is happening in a running human that is not also happening in a lava flow?

 

[rocketdodger]

At no stage have I claimed that a rock has exactly the same properties as a silicon chip, and I'm fairly confident that RD knows this.

Correct -- you have not claimed such a thing.

You have, however, said that every single property anyone has brought up over the last 4+ years is present in both systems.

You have also categorically refused to provide examples of properties that the systems do not share, or even hint at what such properties might be like, or even mention that yes you do think such properties must exist.

In fact, I have specifically asked you "do you think there are properties that a silicon chip has that a rock does not have?" and you refused to respond with a concrete answer.

Is any of this not true? Prove me wrong. Please.

He'd rather argue with the strawman because when it comes to the specific claims that he's making, he's short on evidence.

I have nothing to argue with but the strawman because you have repeatedly demonstrated your unwillingness to explain any of your positions.

 

[rocketdodger]

Why, when what I actually said is right there in the post, do you deliberately misrepresent it? I said that PM's position is that the consciousness of the brain springs from its function as a Turing machine. To equate that statement with "if a Turing machine is just sitting there it will be conscious" is astonishingly dishonest. Did you mean to delete what I'd actually said, and forget?

You should check your own posts before you accuse someone else of dishonesty. You did not say

PM's position is that the consciousness of the brain springs from its function as a Turing machine

what you said was

PM has consistently made the claim that consciousness springs entirely from the function of the brain as a Turing Machine - and nothing else

bold mine

If you don't see the difference then ... well, it wouldn't surprise me

 

[rocketdodger]

Among other things that the brain does, it is probably also a Turing machine, or perhaps it contains a number of Turing machines. A device that might replace the brain might also be a Turing machine. Could such a device replace the brain using only the functionality of a Turing machine? No. The definition of a Turing machine is fairly clear. It's also clear that the functions of the brain in controlling the body are not present in a Turing machine. A Turing machine doesn't interact with the real world. The human brain does.

There have been some claims that a practical implementation of a Turing machine would probably have some kind of mechanism built in which would sort of kind of be able to do the real-time control, somehow. Even if this were true, it's irrelevant. We are not in a position to create any such machine. We are considering what is necessary to build such a machine, as a thought experiment. The assertion is that an implementation of a Turing machine could replace the brain. The assertion is palpable wrong.Any such machine would have to have functionality which is not part of the Turing machine definition.

I think it is understood that what we mean is a turing equivalent system that has the necessary hookups to interface with the rest of the body.

Kind of like ... a brain ?

I'm impressed, in a way, that RD manages to squeeze so much invalid argument into such a compact space.

Firstly, RD's contention is as given above ("we could replace your brain with a turing machine"). The claim is not that the conscious functions of the brain can be duplicated by the Turing machine. It is that the Turing machine can control the body. We don't need to refer to consciousness, with its hazy definitions, to know that the brain's functionality is not and cannot be explained by the Turing model. That is something that is immediately apparent.

What on Earth do you mean "explained by the Turing model?"

The "turing model" doesn't explain anything except how a turing machine runs a software program ( the "tape" ).

You are looking in the wrong place -- the turing machine is just the substrate. All the "explanation" is in the program it runs.

Secondly, while it's certainly the case that the brain and nervous system control the body in a way that does not involve consciousness - it is also the case that voluntary control functions do involve consciousness. We can pick up an object and be conscious of what we are doing. Thus the motor and sensory functions of the brain are not separate from consciousness - they are intricately entangled.

I guess you haven't ever had a dream.

 

[westprog]

You should check your own posts before you accuse someone else of dishonesty. You did not say

If you don't see the difference then ... well, it wouldn't surprise me

There is no difference. And the statement is accurate. The additional words are merely emphasis. The key word is function. Of course I realise that the Turning machine has to be functioning. That's why I said so. If PM were contending that there were any other necessary element required apart from the function of the Turing machine, then he's had endless opportunities to state as much.

 

[westprog]

I think it is understood that what we mean is a turing equivalent system that has the necessary hookups to interface with the rest of the body.

Right. Except that that is not a Turing equivalent system. It's something else. So why does PM insist on repeatedly claiming that a Turing Machine can do everything that a brain does when it obviously can't?

I don't know what PM means, but if he means "A Turing machine plus X" then he has to explicitly state that, not say something quite different. I already said that the brain has the functionality of a Turing Machine. I also stated that other functionality is absolutely essential. If the above is a concession to this point I welcome it.

Kind of like ... a brain ?

Oh, brilliant. We can replace a brain with something "Kind of like... a brain". There's some revolutionary thinking.

What on Earth do you mean "explained by the Turing model?"

The "turing model" doesn't explain anything except how a turing machine runs a software program ( the "tape" ).

You are looking in the wrong place -- the turing machine is just the substrate. All the "explanation" is in the program it runs.

That's why I used the phrase "Turing model". I know that the Turing machine is implementing an algorithm. I've explained this over and over. You cannot fail to realise it.

I guess you haven't ever had a dream.

It's quite possible to have thoughts which are conscious, and which don't arise from sensory inputs or lead to actions. However, the fact that it is possible to carry out conscious actions, and to perceive sensory information, means that consciousness is bound up with the senses, and cannot be considered as something disconnected from them.

 

[Piggy]

No, Piggy, I'm asking for you to quote me ever saying otherwise.

There's no point in me combing over several threads with hundreds of posts each, when you can simply say whether or not you believe you have a sense of felt individual awareness.

If I'm wrong, I'm wrong. Simple as that.

So, do you? Or don't you?

If you do, well, then, I was obviously wrong.

 

[Piggy]

If you disagree with the Church-Turing thesis, you are free to find the error in the proof.

What proof?

You haven't offered any proof with regard to your claims regarding the brain.

Like cornsail and westprog, I searched your links (and other sources I found elsewhere) and failed to find that claim.

So it's not up to any of us to disprove your claim. It's up to you to prove it.

You can do that -- if it can be done -- by explaining how Church-Turing proves that it is impossible for the brain to do anything at all a computer can't do (which is pretty extraordinary to begin with, since we don't know how the brain does all that it does).

Links are fine, as long as you go to the trouble to post excerpts from the relevant passages along with your explanation of how they are relevant.

Once you get around to doing that, then we'll have something to talk about.

Until then, we've only got assertions coupled with links which, as far as I can tell, don't actually support the assertion.

 

[Piggy]

A good place to start would be with a related but simpler proof - the Halting Problem.

Which proves what?

I find one reference to the brain in that article:

Not all theoretically possible machines are subject to the Church–Turing thesis (e.g. oracle machines are not). It is an open empirical question whether there are actual deterministic physical processes that, in the long run, elude simulation by a Turing machine. It's also an open question whether any such process could usefully be harnessed in the form of a calculating machine (a hypercomputer) that could solve the halting problem for a Turing machine amongst other things. It is also an open empirical question whether any such physical processes are involved in the working of the human brain, thus whether humans can solve the halting problem.

The term consciousness does not appear at all.

So, what do you think this proves?

 

[Piggy]

So long as the mechanism being slowed fully contains the mechanism of consciousness, it makes no difference. Consciousness is maintained.

Not necessarily.

For instance, if you slow down your computer's processing speed drastically, it will not play a music CD very slowly. The music will not play at all.

As I've mentioned before, consciousness is a bodily function, a behavior.

It cannot be caused by computation.

Events in OPR have physical causes. There is no reason to believe that conscious awareness -- as strange as it is -- will be any exception. If it is an exception, I'd love to hear an explanation of how that works (seriously, because it would be way cool!).

If it's true that the synchronization of the "signature" waves is an essential part of the physical mechanism -- because, again, there must be some sort of physical mechanism -- then how is it possible to propose an infinite range of speed, i.e. an absense of stall point?