Has consciousness been fully explained?

[PixyMisa]

To quote your reference paper "We argue that all of those measures provide distinct windows into the same distributed state of conscious processing."

So they say that they think it demonstrates a distributed state of processing. Which fits nicely with many neural net models.

Right. They're still talking about neurons signalling each other the exact same way they always do - just now the neurons are in a large-scale feedback loop and so produce a regular, detectable signature as the signal propagates around the loop.

There's no sudden discovery of a magical other. That's just Piggy's misinterpretation of the study.

 

[Piggy]

But if I write a program that simulates the behaviour of a car I am modelling the car, no matter how detailed the simulation.

But I am not really concerned with what you call it - if you want to call it an actual brain rather than a simulation of a brain I am fine with that.

I am more interested in whether you think the virtual person it was attached to would exhibit the external behaviour we normally associate with conscious people.

Let me see if I understand you....

What you seem to be asking is whether or not the combination of a mechanical machine and a computer -- if the combo were designed to do all the things humans do -- would do all the things humans do.

Sounds like a tautology to me.

In any case, I don't see any relevance of this thought experiment to the questions at hand.

The better question is this: If we were to design and build a conscious machine (which must be possible somehow since we know conscious organic machines already exist) how would it have to be rigged up in order for the function of consciousness to operate successfully?

The answer is "We don't know" because we don't know how the brain and body are rigged up to do it, and that's our only known exemplar. So as of now, we have no template for accomplishing this feat.

But by glossing over the question of the setup, and simply asking about a computer of unknown design and a mechanical device of unknown design interacting in unspecified ways, you've proposed a meaningless thought experiment.

 

[Piggy]

Strangely there are studies that show we are dreaming all night and various levels of arousal are involved. Dropping an object that makes a loud noise will also show that most sleeping people are 'aware' at some level.

I'm not aware of any such studies, although I am aware of recent studies on REM and non-REM dreaming.

Of course, if you define "dreaming" as "any brain activity" then you've merely launched into a Humpty-Dumpty-ism.

If we were actually dreaming all the time, then we would never wake up without a dream in our heads, and of course we do this quite regularly.

As for the intrusive sound example, this is yet another demonstration of the distinction between conscious awareness, on the one hand, and other types of "awareness" on the other. And there are slews of these.

While you're dreaming, your mind is still "alert" in non-conscious ways (as it always is) to what's going on around it, and within it, and in the body. That's non-conscious processing. And it is those non-conscious processes which "decide" whether or not an input -- from the body or from the senses -- is of sufficient importance to send it off to the apparatus that will make you consciously aware of it.

Ditto the cocktail party effect: You're in a noisy room when suddenly you become aware that you just hear someone say your name a split second ago. The name "jumps out at you". That's because the non-conscious processes that work as gatekeepers have a strong enough match between that sound and an extremely active and highly connected pattern (your name) to warrant routing it into those systems which perform the feat of incorporating it into your conscious experience.

These gatekeeper processes don't actually "care" that it's your name. They just get a strong enough threshhold of connections to trigger the routing decision.

 

[Piggy]

Pixy, I have to run, so I'll catch up with responses later, but I can't let this go....

And the exciting thing about this discovery is that, at last, we have an activity of the organ which is correlated with consciousing, but not with other activities of the brain.

No. The article says nothing of the sort, and nothing from which you could rationally infer that.

How, then, do you interpret this passage from the article?

A new paper suggests that four specific, separate processes combine as a "signature" of conscious activity. By studying the neural activity of people who are presented with two different types of stimuli – one which could be perceived consciously, and one which could not – Dr. Gaillard of INSERM and colleagues, show that these four processes occur only in the former, conscious perception task.

Bolding and italics mine.

Is that clear enough? Only during conscious processing, not during non-conscious processing.

 

[Dancing David]

Right. They're still talking about neurons signalling each other the exact same way they always do - just now the neurons are in a large-scale feedback loop and so produce a regular, detectable signature as the signal propagates around the loop.

There's no sudden discovery of a magical other. That's just Piggy's misinterpretation of the study.

I disagree, the signal is a signal, the question is how does it arrive there, we should have a better understanding of how in twenty years.

It could be related phase effects across the areas, it could be separate but synchronous events. Hard to say.

My guess is that it is the associative network in use and therefore the scattered processes do have to coordinate. You have visual possessing, you have the recognition of letters, the associated and conditioned learning of words, the associated and conditioned memories and emotions.

Are they separate or coherent? We don't know. What level of coherence across systems?

I think that I have to ponder Piggy's use of 'execution', I do not tend to see the 'programs' as 'running' but more of 'flowing' the neural network is comparable to a stream bed, where each passage of the water makes the channel deeper. With the proviso that sometimes flow can deepen or make shallow the channels in other areas.

 

[Dancing David]

I'm not aware of any such studies, although I am aware of recent studies on REM and non-REM dreaming.

Of course, if you define "dreaming" as "any brain activity" then you've merely launched into a Humpty-Dumpty-ism.

If we were actually dreaming all the time, then we would never wake up without a dream in our heads, and of course we do this quite regularly.

As for the intrusive sound example, this is yet another demonstration of the distinction between conscious awareness, on the one hand, and other types of "awareness" on the other. And there are slews of these.

While you're dreaming, your mind is still "alert" in non-conscious ways (as it always is) to what's going on around it, and within it, and in the body. That's non-conscious processing. And it is those non-conscious processes which "decide" whether or not an input -- from the body or from the senses -- is of sufficient importance to send it off to the apparatus that will make you consciously aware of it.

Ditto the cocktail party effect: You're in a noisy room when suddenly you become aware that you just hear someone say your name a split second ago. The name "jumps out at you". That's because the non-conscious processes that work as gatekeepers have a strong enough match between that sound and an extremely active and highly connected pattern (your name) to warrant routing it into those systems which perform the feat of incorporating it into your conscious experience.

These gatekeeper processes don't actually "care" that it's your name. They just get a strong enough threshhold of connections to trigger the routing decision.

I will look it up, the report I read (a long time ago) said that subjects reported dreaming when woken at all phases of sleep, there is a substantial difference in the nature of the dreams reported.

And so the consciousness may be related to arousal and awareness which is a possible analog of REM vs non-REM and levels of arousal and awareness.

More later especially on false positives in random noise events.

 

[westprog]

For the same reason Westprog says that digital computers are not digital but analog machines.

By establishing semantic barriers the discussion does not occur about the concepts behind them.

I'm trying to discuss the concepts. Pre-emptive multitasking systems are different conceptually from Turing machines.

So the fact that Turing did not conceive of modern computing techniques means Westprog can just categorically make statements about the concept of a Turing machine.

I really don't understand this. We have these two models of computing. One is a better fit than the other. When I point this out, it's like I spat in the soup. Why is it so important that we stick to the Turing model, when it cannot explain functionality that's present in the brain, when we have an alternative model which will explain such functionality?

And something that was understood over fifty years ago - the concept of the non-deterministic interrupt - is hardly a "modern computing technique". It's part of just about every computing system.

 

[rocketdodger]

The better question is this: If we were to design and build a conscious machine (which must be possible somehow since we know conscious organic machines already exist) how would it have to be rigged up in order for the function of consciousness to operate successfully?

The answer is "We don't know" because we don't know how the brain and body are rigged up to do it, and that's our only known exemplar. So as of now, we have no template for accomplishing this feat.

But this is categorically untrue, Piggy.

Thousands of individuals have a very good idea of what they would try in such a scenario -- including myself. If you want a high level view I can tell you I would try using a bunch of perceptron like networks to feed into interconnected boltzmann machines in various ways.

In every case the limiting factor is just manufacturing capability and in turn computing power. Right now we just don't have the tools or power to create and model the kind of systems that would be necessary. But that doesn't mean nobody has a clue regarding what they would try if we did have those resources.

And keep in mind that such resources are coming on line very rapidly, in fact. So it s only a matter of time.

 

[AlBell]

But this is categorically untrue, Piggy.

Thousands of individuals have a very good idea of what they would try in such a scenario -- including myself. If you want a high level view I can tell you I would try using a bunch of perceptron like networks to feed into interconnected boltzmann machines in various ways.

In every case the limiting factor is just manufacturing capability and in turn computing power. Right now we just don't have the tools or power to create and model the kind of systems that would be necessary. But that doesn't mean nobody has a clue regarding what they would try if we did have those resources.

And keep in mind that such resources are coming on line very rapidly, in fact. So it s only a matter of time.

As soon as you begin to understand how many orders of magnitude you've underestimated the problem, you can probably start with technology available at that time. You or your great-great-great-etc-grandchild.

The current hand-waving remains just that. I suppose it's at least as productive as discussing angels dancing on pins.

 

[westprog]

In case anyone is interested, I spent a good amount of hours formulating a response to this challenge of westprog's regarding "physical definition of computation."

The whole thing requires its own thread, and I am too lazy to write it all out, but the following brief should be enough to convince anyone that respectably has business being here discussing it in the first place.

Basically, it comes down three ideas: to stable systems of particles versus unstable ones, how systems interact with each other, and what computation actually does.

The concept of "stable" here is tricky, because -- per westprog's insistance -- we can't use a definition that is human dependent and saying "stability" means "it exists into the future for longer" begs the question of "what exists? Who defines what constitutes the 'system?'" So by "stable" I mean only that other systems can physically interact with said system in a similar way for a longer duration. For instance, a chunk of metal floating in space is more stable than a gas cloud in space because any system is able to interact with a chunk of metal in the same way for much further into the future, compared to a gas cloud whose molecules dissapate across space very quickly, etc.

The second idea is that similar interactions between stable systems (really, the interaction is what is stable, but you get the idea ) by definition can occur more frequently than between non-stable ones. Mathematically one can say stable systems can "recognize" each other with much higher reliability because they exist in a form that is "recognizable" for a longer time. For example, if two compounds are in a solution and tend to react with each other, then the longer each of those species exists on its own the more frequently they will be able to react. If each species was instead only transiently existed the reaction rate would be much slower.

Now both of those ideas are just a mathematical and scientific basis for the intuitively clear notion that things that are stable exist longer. Is it better to exist longer? No -- there is no "better" in this discussion. Things that exist longer just exist longer and in turn that means other things that also exist longer can interact with them more reliably.

Here is the point -- the physical definition of computation. I have already established that there is a reason systems would exist longer than other systems. I have already established that if there was a way a system could keep itself existing longer it would exist longer. By definition, there is a selective pressure favoring things that exist longer, so any systems that do it -- by any mechanism -- will statistically exist with higher frequencies as time goes on, all else being equal. The only way a system of particles can be said to increase its own stability is via computation

Let's consider an example of a system of particles becoming more stable. How about, say, a lava flow congealing into a rock? Is that a system of particles increasing its own stability? Yes. It does so by releasing heat into the environment. It is then much more stable - in fact, it's capable of retaining its configuration for millions of years.

Does it do this via a process of computation? I can't see how that's the case.

What is computation? It is the property of a stable system that can be mathematically described as mapping the set of all external states to a SMALLER set of internal states, where "state" is defined according to the idea of "stability" described in the first idea. A system can use a series of computations to exhibit different behaviors such that it's stability is increased. It is that simple.

All life does this (in fact, one can say that life and the machines life creates are the only things that exhibit series of computations -- or "computes" -- although I am sure a few other examples might be found). That is what makes life life -- it computes and by doing so it increases the stability of the life system and in turn it exists longer into the future.

So long story short, the next time westprog tries to wiggle out of questions with this "physical definition of computation" stuff, know that it isn't even an issue anymore. I defined it, it exists, it is a process that only certain systems undergo, not all. And most importantly, It is not human-centric in any way. The idea is completely independent of humans.

It would have been pretty impressive if RD had managed to concoct an entire new physical theory after several hours intensive thinking. In fact, it would probably be unique in the history of science. Newton and Einstein needed many years of reflection.

So it's not surprising that this concept of "stability" isn't actually quantifiable, or measurable, or that there's a string of non sequitors holding the shaky structure together. Still, at least there's some thinking going on.

The problem is of course that the most stable physical system is an entirely inert one. What is preserved by life is not stability, but patterns of information - and defining that is a big job.

There is a property which covers this kind of thing - it's entropy. For an example of how to misuse entropy to give life a claim to a special physical nature, check out a range of creationist sites.

 

[Piggy]

But this is categorically untrue, Piggy.

Thousands of individuals have a very good idea of what they would try in such a scenario -- including myself. If you want a high level view I can tell you I would try using a bunch of perceptron like networks to feed into interconnected boltzmann machines in various ways.

In every case the limiting factor is just manufacturing capability and in turn computing power. Right now we just don't have the tools or power to create and model the kind of systems that would be necessary. But that doesn't mean nobody has a clue regarding what they would try if we did have those resources.

And keep in mind that such resources are coming on line very rapidly, in fact. So it s only a matter of time.

I agree that it's just a matter of time.

And no doubt, many folks have an idea of what they'd like to try.

But as it stands, we don't yet know which, if any, of those ideas would actually work in practice because we haven't yet documented exactly what the brain is doing when it generates consciousness.

 

[Piggy]

Does it do this via a process of computation? I can't see how that's the case.

Well, Stephen Wolfram, for example, has an interesting way of viewing weather systems as performing computations.

 

[westprog]

As soon as you begin to understand how many orders of magnitude you've underestimated the problem, you can probably start with technology available at that time. You or your great-great-great-etc-grandchild.

The current hand-waving remains just that. I suppose it's at least as productive as discussing angels dancing on pins.

The AI singularity is always just around the corner. Everything is pretty well understood, and then we'll have it. C'mon now! Just one more push!

 

[Piggy]

I will wait on the defintion of your use of executive apparently, thats cool.

By "executive" I mean "carrying out". That is, without some executive (in this sense) mechanism, the signal associated with constricting the iris won't do anything. The overt behavior won't happen. You'll be left with just the neural chain reactions in the brain.

Something has to translate (so to speak) between the IP-supported-by-just-enough-hardware-to-accomplish-it and the bodily function which is not merely IP.

There's no reason to expect consciousing to be any different.

Just as with regulating body temperature or heartbeat or anything else you can think of that the body does, something has to "carry out" the task based upon the output, or there is no behavior beyond the background logic, the entire purpose of which (of course) is to drive the behavior.

 

[Piggy]

Well it is a bit more that toe to heel, there are the direct dendrite connections as well, which would allow for greater synchronization of cells that the neurotransmission at the synaptic cleft.

The 'brain wave' thing is always a conundrum at this point because it is as in so many things a possible byproduct of the other mechanisms. When it is shown how it has an influence on neural behavior at various levels, then we will know.

So it could be like the heart muscle, where the synchronized pulse wave is essential to the function of the heart or it could just be a side effect from taking the EEG.

Just as with a super large old fashioned computer like ENIAC, we could probably find cycles in the EM pulses (which brains don’t use they use a biochemical phase shift in the cell wall), but the question is how does that relate.

It might be significant , it might not.

To quote your reference paper "We argue that all of those measures provide distinct windows into the same distributed state of conscious processing."

So they say that they think it demonstrates a distributed state of processing. Which fits nicely with many neural net models.

The changes in the gamma and beta band spectrums may be causal or they may be side effects of that processing.

Yes, that's why I was careful to say that it is correlated with conscious processes but not with non-conscious processes. That's certainly a breakthrough, but definitely not an explanation of how the brain does consciousness.

Progress is being made, certainly, but we haven't yet cracked the nut.

 

[Piggy]

Show me an example of self-referential information processing that has nothing to do with conscious awareness. Just one. That would, after all, quite demolish my position.

Seriously?

Well, ok.

SRIP, of course, means IP which uses its own outputs as inputs. Or we can say the system queries its own state. [ETA: If you're using the word to mean something else, we can go from there.]

The simplest example of SRIP involved in actions other than Sofia events would be recursion, which is of course extremely common to the point of near ubiquity.

Also, if we take a concert's sound equipment as a single system, a feedback howl is the result of the system using its own output as an input.

In the body, we can consider the cocktail party effect, wherein certain daemons -- whose activity we are never consciously aware of, and who work "upstream" from the conscious processes -- are constantly checking input to see if it's important enough to route over to conscious awareness. If a particular pattern bounces back a very high level of association with other important patterns, it gets shuttled over.

Which is why the word "Piggy" jumps out at me, but not at my friend Joe who I'm having a conversation with. Without using outputs as inputs or querying its own state, my brain couldn't accomplish this.

 

[Piggy]

Right. They're still talking about neurons signalling each other the exact same way they always do - just now the neurons are in a large-scale feedback loop and so produce a regular, detectable signature as the signal propagates around the loop.

There's no sudden discovery of a magical other. That's just Piggy's misinterpretation of the study.

But I'm not proposing any "magical other" and never have been.

We all agree that the brain does consciousness, and that the brain alone does consciousness. (ETA: By "alone" here I mean that it does it without the aid of other organs.)

What I disagree with is the "IP only" interpretation, which (among other problems) leads to absurd conclusions, such as the contention that working out the brain's activities symbolically on pen and paper would generate a Sofia event.

ETA: A schematic of this loop, describing the structures involved and their order, would be quite helpful here.

 

[westprog]

Well, Stephen Wolfram, for example, has an interesting way of viewing weather systems as performing computations.

There's potential for an objective theory dealing with what computation is and when it is and isn't going on.

 

[PixyMisa]

I disagree, the signal is a signal, the question is how does it arrive there, we should have a better understanding of how in twenty years.

It could be related phase effects across the areas, it could be separate but synchronous events. Hard to say.

All of which is just neurons, switching and signalling. No new process involved. I'm not sure where we really disagree here.

I think that I have to ponder Piggy's use of 'execution', I do not tend to see the 'programs' as 'running' but more of 'flowing' the neural network is comparable to a stream bed, where each passage of the water makes the channel deeper. With the proviso that sometimes flow can deepen or make shallow the channels in other areas.

Sure - the signal flows from one neuron to it's neighbours and out across the brain. We can track that in real time. That's something that's discussed in the MIT Introduction to Psychology lectures.

 

[PixyMisa]

In every case the limiting factor is just manufacturing capability and in turn computing power. Right now we just don't have the tools or power to create and model the kind of systems that would be necessary. But that doesn't mean nobody has a clue regarding what they would try if we did have those resources.

And keep in mind that such resources are coming on line very rapidly, in fact. So it s only a matter of time.

Indeed, we could do it now, but computers are getting cheaper so fast that by the time you got such a large system built and working, you'd be able to replace it for 10% of the cost. So it only makes sense to practice with smaller systems (e.g. Blue Brain) and wait while Intel and AMD cut the cost of the project for us by about 30% per year.