The Hard Problem of Gravity

[Belz...]

AFAIK nothing magical is being proposed by GWT. As quoted above..."An information becomes conscious, however, if the neural population that represents it is mobilized by top-down attentional amplification into a brain-scale state of coherent activity that involves many neurons distributed throughout the brain." As I understand it, and I stress that I'm no expert on GWT, this means that an attentional process has the power to make certain neural representations "globally available" to a great range of unconscious neuronal modules.

How does it do this?

How does it do what ? I can't make heads or tails of what you've written.

"Inner dialogue" is the term I usually see being used to convey the process we more commonly know as "thinking" - the apparent passage of thoughts in and out of awareness, making for example these discussions possible.

So it's a synonym for consciousness ? Why use it, then ?

 

[Ichneumonwasp]

But a simulation of running on a computer is in no way like running. To show this is true, compare the problem of programming a robot to run, and getting Lara to run. Getting a robot to run is really, really difficult. AFAIAA, it hasn't actually been done yet for bipedal motion. That's because the robot has to really implement the relationships between components that make up "running". A digital simulation, OTOH, does not have to cope with these relationships. The supposed interaction between the parts does not actually happen.

Depends on the simulation. You can create the appearance of running easily in a simulation, but that doesn't fully simulate running because it is just a gloss. We can do the same with consciousness -- gloss it with simple simulations. If you want to closely simulate running in all its particulars, then that is more difficult. Same with consciousness.

Regardless, it is still running. If it is not, then could you please provide a clear operational definition of running that is not met by a simulation? Running simply is translational motion via the interaction of various parts of a whole.

 

[Belz...]

Do you think consciousness is some sort of neurologicomusical chord?

That would be AWESOME! Musicians all around the world could try and find the right combination of sounds to create temporary conscious entities and they'd become real-world superheroes, using their powers to save or enslave the world !

...I gotta write this down.

 

[Belz...]

Er... no. YOU're the one who mentioned it. What does algorithms have to do with it ?

Westprog ?

 

[Ichneumonwasp]

I would certainly deny that "real" calculation was going on. If real calculation goes on on a purely functional basis, then calculation takes place every time stones are washed up on a beach.

I think you are confusing the process that takes place that we call calculation when we recognize it as such with our ability to view many things as calculation.

A computer carrying out arithmetic is calculating however we want to define it. It carries out those processes that are already defined as calculation because we set it up that way. It does so through a directed process -- and that is one of the prime reasons why we define it as calculation.

A second rock washing up on the shore -- a stochastic process -- can be viewed as simple addition; but there was no intention (as far as we can tell) for that occurrence to represent addition.

A key aspect of this thing we call consciousness is intentionality; it is an orientation toward the world that is supplied by our brains in the way that our brains are structured (and much of this arises through emotions, motivations, and feelings). We can provide computers with the same orientation.

 

[westprog]

My mistake.

Sorry for being brusque - I've had all kinds of views attributed to me on this thread.

As far as I can tell, there's a broad consensus that consciousness is probably only manifested biologically, either directly or through man-made devices. The only exception would seem to be a view not expressed here that consciousness permeates everything in the universe. The main difference is in degree of certainty.

 

[lupus_in_fabula]

Besides, I imagine that it is not the reverberation itself per se that is needed to create consciousness, rather specific frequencies or frequency groups.

I would imagine it is a necessary aspect. How else can the condition be sustained long enough for there to even be coherent and synchronized activity? I am however at the limits of my technical knowledge here, so I might be totally off target. I've always thought that frequency is the measured value of the the oscillation. In this case it seems to me that a particular frequency band correlates to more synchronized activity over longer distances (when taking into account longer conduction delays).

 

[westprog]

I think you are confusing the process that takes place that we call calculation when we recognize it as such with our ability to view many things as calculation.

A computer carrying out arithmetic is calculating however we want to define it. It carries out those processes that are already defined as calculation because we set it up that way. It does so through a directed process -- and that is one of the prime reasons why we define it as calculation.

I think that's quite close to what I was saying. If we leave the calculator switched on and we sit on it, and the batteries run out, then has it been doing calculations?

It is the human intention and usage that count.

A second rock washing up on the shore -- a stochastic process -- can be viewed as simple addition; but there was no intention (as far as we can tell) for that occurrence to represent addition.

But if I stride by and observe the stones, and count them, then they are doing calculation every bit as much as the machine.

A key aspect of this thing we call consciousness is intentionality; it is an orientation toward the world that is supplied by our brains in the way that our brains are structured (and much of this arises through emotions, motivations, and feelings).

Indeed. Calculation depends on consciousness.

We can provide computers with the same orientation.

Perhaps we can - but until I can see a precise distinction between the washed up stones and the electronic device, I will class them together.

 

[Nick227]

Nick, it's the information processing system itself that's oscillating. Any information processing system that involves loops can be viewed as an oscillator.

Can be, yes. But sustained reverberation implies to me that one is more dealing with something akin to a carrier frequency here. Presumably specific frequencies allow neurons to communicate more rapidly across a wider network. Whether these frequencies are set up by the information represented in the neurons or by some other feature of neuronal activity I'm not clear, but I very much doubt its the former. So, for me, self-reference as a marker for consciousness is failing again.

They are talking about self-referential information processing here.

I doubt that very much, but will read the paper.

What the...

What on Earth are you talking about now? How in Heaven are specific oscillation frequencies (or "frequency groups", whatever those are supposed to be) supposed to create consciousness, when other frequencies (or "frequency groups") do not?

Presumably because they facilitate rapid information transfer neuronally. Remember, global accessibility is consciousness. This is the basis of GWT.

Nick

 

[westprog]

Westprog ?

I'm asking you. If it isn't the execution of the algorithm, please tell me.

I'm quite willing to say what I think, but I don't think I should go into detail about what I think you think if you say you don't think that. If you don't think that consciousness is the result of execution of an algorithm, then say so. I'm not trying to trick anyone.

 

[Nick227]

I would imagine it is a necessary aspect. How else can the condition be sustained long enough for there to even be coherent and synchronized activity? I am however at the limits of my technical knowledge here, so I might be totally off target. I've always thought that frequency is the measured value of the the oscillation. In this case it seems to me that a particular frequency band correlates to more synchronized activity over longer distances (when taking into account longer conduction delays).

I also can't imagine that it is the information represented neuronally which itself sets up or maintains these sustained reverberations. It seems more likely to me that this is undertaken by another aspect of the neuron's activity.

Nick

 

[Ichneumonwasp]

Well, I think our emotions and certain sensations are a subclass of "behavioral tendencies"; they influence our conscious actions. In a context outside of a conscious mind inputs into a system could act as "behavioral tendencies" that influence the overall behavior of the system.

Basically, I agree that in a general sense your original statement is accurate [i.e. 'feelings' act as behavior tendencies] but that alone does not help distinguish why 'feelings' are felt as such.

I think we are using words in different ways.

Emotions are not properly "behavioral tendencies"; rather, they are themselves a type of behavior. When an emotion "happens" behaviors occur within our bodies. With severe fear, we "fight or flight"; and emotions are accompanied by internal behaviors (increased heart rate, etc.).

Of course emotions impact conscious actions -- they do so through the complex cortical processing that we call 'feelings'. But the emotions themselves are not the behavioral tendency. They are a behavior that can be further processed. The emotion happens, but the feeling is a further elaboration on the emotion that provides only a tendency toward further action rather than a decision already made. This elaboration -- tendencies toward action that only push us in a particular direction -- allow significant behavioral flexibility.

I suppose we could define "feelings" as a type of behavior, but it is very different type than emotion or barebones perception. Emotion and perception are things that just happen. We cannot control them in the way that we can control our feelings -- because feelings are not, strictly speaking, a behavior in action, but rather only the tendency toward a behavior.

Does that distinction make sense?

From a neurological perspective, feelings only become possible through inhibition. Real life and real brains are more complicated than this, but think along the lines of kids and adults. Kids just do stuff without further mediation. Adults sift through our behavioral options because we have a means of inhibiting bare emotion; we feel different emotions rather than act on them immediately.

That's what our prefrontal cortex is for; knock out the prefrontal cortex and people become very different because emotion is not processed properly. Knock out the cingulate gyrus bilaterally and you see folks who are awake but not really aware -- it's a state known as akinetic mutism -- where emotions don't even get to the higher brain.

I don't think for a second that this explains consciousness, but I think it goes a way toward explaining the "feeling" and "experience" bits.

 

[Ichneumonwasp]

Perhaps we can - but until I can see a precise distinction between the washed up stones and the electronic device, I will class them together.

Intentionality is structured into any program that adds numbers together. The silicon chips do nothing without instruction. Instruction is intentionality.

Rocks washing up on shore are just rocks washing up on shore unless viewed another way (as addition). We needn't view what a computer does when it adds numbers in any new way because the intentionality is already supplied in the program itself.


ETA:

Just to be clear. I do not think that this means that a computer understands that it is calculating. That type of intentionality requires more complex processing. I think we could build that into a computer, but that requires a better understanding and an actual definition of 'understanding'.

 

[lupus_in_fabula]

I also can't imagine that it is the information represented neuronally which itself sets up or maintains these sustained reverberations. It seems more likely to me that this is undertaken by another aspect of the neuron's activity.

Well, I think I wasn't clear, I was disagreeing with you.

I'm saying that the particular frequency range happens to be the measured value found. It is not the cause in and of itself; something has to rock the boat in the first place. Neurons communicate with each other back and forth. Why can't such activity can be described as information processing when looking at the whole picture?

 

[Nick227]

I'm saying that the particular frequency range happens to be the measured value found. It is not the cause in and of itself; something has to rock the boat in the first place. Neurons communicate with each other back and forth. Why can't such activity can be described as information processing when looking at the whole picture?

I'm sure it can. However, if we consider a group of neurons resonating at, say, 50Hz, it strikes me that whilst this frequency may favour inter-neuronal communication, rather as a carrier wave favours the transmission of information in radio waves, it does not itself contain much information.

Thus, that the information represented neuronally either does or does not self-reference IMO makes no difference. I very much doubt that you get neurons to resonate by getting them to reference their neighbour's information. I could be wrong but it seems to me unlikely. So, as far as I can tell, and the linked paper doesn't seem to go into these neuronal dynamics, self-reference is not a factor here in creating the conditions for conscious access.

Nick

 

[Ichneumonwasp]

I'm sure it can. However, if we consider a group of neurons resonating at, say, 50Hz, it strikes me that whilst this frequency may favour inter-neuronal communication, rather as a carrier wave favours the transmission of information in radio waves, it does not itself contain much information.

Thus, that the information represented neuronally either does or does not self-reference IMO makes no difference. I very much doubt that you get neurons to resonate by getting them to reference their neighbour's information. I could be wrong but it seems to me unlikely. So, as far as I can tell, and the linked paper doesn't seem to go into these neuronal dynamics, self-reference is not a factor here in creating the conditions for conscious access.

Nick

I'm a bit confused by what you are saying, I'm afraid. Do you have the idea that neurons contain information in some way?

The frequency of neuron firing, the pattern of firing, the types of connections created between them (strong, weak) and the resonant frequencies (in the cortex this generally occurs at 40 Hz) are the information. Neurons are just cells. They don't contain any information in and of themselves.


ETA:

It is very likely that I am just misreading you.

But, if anyone else is interested and does not know how the nervous system works, it is the slight variations in firing that code information.

If the whole brain does that same thing over and over again (one big reverberating loop), then no cognitive work can be performed. That "whole brain reverberating loop" is check-out time. It's what happens (essentially) during NREM sleep (but sleep is a bit more complicated than that), and what a seizure basically is. Absence seizures -- when kids stare off and are unresponsive for several seconds -- consist in a reverberating thalamocortical loop at a frquency of 3 Hz, for instance.

 

[Belz...]

As far as I can tell, there's a broad consensus that consciousness is probably only manifested biologically, either directly or through man-made devices. The only exception would seem to be a view not expressed here that consciousness permeates everything in the universe. The main difference is in degree of certainty.

How did you come to that conclusion ? Are we reading the same thread ?

I'm asking you.

Asking me what ? I want to know how you think it's algorithmic because I don't understand what you mean. Until I do I find it difficult to answer you.

I'm quite willing to say what I think, but I don't think I should go into detail about what I think you think if you say you don't think that.

Sounds reasonable

 

[westprog]

Depends on the simulation. You can create the appearance of running easily in a simulation, but that doesn't fully simulate running because it is just a gloss. We can do the same with consciousness -- gloss it with simple simulations. If you want to closely simulate running in all its particulars, then that is more difficult. Same with consciousness.

Regardless, it is still running. If it is not, then could you please provide a clear operational definition of running that is not met by a simulation? Running simply is translational motion via the interaction of various parts of a whole.

Right. Translational motion. If translational motion is involved, then some emulation of the behaviour is involved. If there's no motion - actually something going from one place to the next - then it's just a simulation.

Running is a fairly vague thing. We might argue whether a man is running or just walking fast. But we know that a picture isn't running, and a story in a book isn't running, and a film of someone running isn't running. There has to be that translational motion. Indeed, we could claim that someone on a treadmill is runnning, even if he's not getting anywhere, but at least real legs are moving up and down.

 

[westprog]

Intentionality is structured into any program that adds numbers together. The silicon chips do nothing without instruction. Instruction is intentionality.

Rocks washing up on shore are just rocks washing up on shore unless viewed another way (as addition). We needn't view what a computer does when it adds numbers in any new way because the intentionality is already supplied in the program itself.


ETA:

Just to be clear. I do not think that this means that a computer understands that it is calculating. That type of intentionality requires more complex processing. I think we could build that into a computer, but that requires a better understanding and an actual definition of 'understanding'.

The issue of intentionality is important - though I'm not sure that it still means that the calculator is necessarily calculating if sat on.

I want to go into this in more detail because of the thought experiment of a perfect simulation of the universe done in stones laid out on sand. I'm convinced that this is not equivalent to the actual universe, even if the actual universe is entirely quantised and discrete. But I think it's necessary to say why.

 

[rocketdodger]

As far as I can tell, there's a broad consensus that consciousness is probably only manifested biologically, either directly or through man-made devices.

I explained why that is, and why it has nothing to do with the human-centric anthropomorphic doctrine you seem to be advocating.

It is a result of the mechanism of natural selection. Period.