Perception of Color: Separating the Physical from the Supernatural

[Q_Goest]

Hi Pythagorean. My apologies in advance for the length. Note that in the following, I’ll be using a fundamental assumption that both you and Dennett appear to agree on, that being computationalism. That is, that the interactions of neurons is what produces the emergent properties of conscious experience. If we accept this, then I believe the rest of Dennett’s paper, Quinning Qualia, can logically follow from that. If we were to base our science of consciousness on something other than computationalism, then Dennett’s paper may or may not apply. I mention this because I think it’s important to understand the basic assumptions that go into an argument such as Dennett provides because different assumptions often lead to different conclusions, and the conclusions Dennett comes to seem to indicate there’s a fundamental flaw in our notions of what conscious experience is. I agree with his conclusion given his unwritten assumptions, but would disagree with his unwritten assumptions.

But I see something different in the rest of your post. You've made physical changes to the system that have brought about changes in qualia. So far, this supports my point … Now all we need is the technology that can see the connections. Then with a statistical empirical method, we go in and look at the connections and all the propositions that you made in your post (since, as you've argued, their is a physical basis for the changes in qualia. Our device is designed to "see" whether the cables are flipped 180 degrees).

Now we go a step further with our advanced alien laser cutter/manipulator and "flip the cable 180 degrees" ourselves. We can then measure the outcome (through statistical empirical methods of course, we can't trust one sample) and begin to build a model of the brain.

I understand how you came to your conclusion and I would agree that it’s logical. I’ve not done a very good job representing Dennett I’m afraid. I’ll try once more.

What you’ve pointed out by suggesting we can correlate the connections to the experience is exactly what Dennett is arguing against. Here’s how he introduces this paper:

My claim--which can only come into focus as we proceed--is that conscious experience has no properties that are special in any of the ways qualia have been supposed to be special.

He’s going to claim that qualia and experience of things such as red or pain do not have properties that are as we perceive them. That’s what he means by experiences not being special. Specifically, he means there are certain assumptions we make about our own experiences that do not coincide with a logical analysis of those experiences. Those assumptions he outlines as follows:

So, to summarize the tradition, qualia are supposed to be properties of a subject's mental states that are

(1) ineffable

(2) intrinsic

(3) private

(4) directly or immediately apprehensible in consciousness

The key property I’d like to focus on is the second one, “intrinsic”. This is the property you are also taking at face value. Your assumptions seem to indicate that you feel your experiences are intrinsic. There is an intrinsic correlation between the computation and what is experienced. Intrinsic in the sense that the correlation is intrinsic to nature. Every time that particular computation is performed, the experience will follow, and be supervenient on that physical substrate that is performing the computation. So when we flipped the plug 180 degrees, then if we believe the computation is intrinsic, then we believe that with this reversed input, the experience will also be flipped.

Now let’s go back to the thought experiment. This:
INPUT --- COMPUTATION --- OUTPUT

Produces the same output for any given input as this:
INPUT -x- COMPUTATION -x- OUTPUT

Similarly, we must also assume that every person has a slightly different computation going on inside their brain. So let’s say person 1 has this:
INPUT --- COMPUTATION(1) --- OUTPUT

And person 2 has this:
INPUT --- COMPUTATION(2) --- OUTPUT

Where COMPUTATION(1) is not equal to COMPUTATION(2). We can safely assume this because of multiple realizability. None of our brains are identical, but if we want to believe that qualia are intrinsic, then we must accept that my red is the same experience as your red. Now how many different people are possible? There are roughly 7 billion people on the Earth today and I’m guessing billions more who have already passed away. I don’t know how many different people there could possibly be, but I’m guessing it is many orders of magnitude larger than 7 billion, resulting in there being hundreds of billions or perhaps millions of billions of computations that produce the same intrinsic experience of a given color. And if we believe that a small, evolutionary step backwards or forwards does not change the experience, then we have to assume that there are also an array of different brains such as Cro-Magnon, Neanderthal and perhaps Orangutan brains must also experience the same intrinsic qualia that we homo sapiens experience. We could continue to extend this to other animals, so the number of different brains that potentially produce the same experience of the color red for example, must begin to approach an insanely huge number. If qualia are intrinsic, then we have to accept that the number of different computations that will produce the same experience is huge. I won’t say infinite, because I can’t be sure of that.

To make matters worse, each brain is also plastic, rewiring itself as time proceeds. Yet even with this rewiring, if qualia are intrinsic, then the these new computations that are taking place as time goes on, must also produce the same experience inside the same, rewired brain. This includes rewiring of inputs and outputs.

Next, I’d like to suggest that there could be identical computations that produce different qualia. Let’s take a simple example. Experimenters have taken rat neurons and grown them inside Petrie dishes on top of electrical arrays that allow the neurons to interact with electrical impulses from this two dimensional array. In one example, the neurons were made to ‘fly’ an aircraft (flight simulator inside a computer) straight and level. The way this was done isn’t that hard to understand. They were able to influence how the neurons reacted by inputting signals at different points on this array over which this sheet of neurons were grown. Then they would monitor the outputs/interactions and took the output they needed to get the flight simulator to fly straight and level. They could have equally made those same outputs make the simulator fly perfect barrel rolls or loops. They could equally have used those same outputs to control stop lights in a large city, or controlled the processes in an air separation plant, or the flights between various major cities. They were in no way forced to take these outputs and only fly this particular flight simulator straight and level with them. The outputs could have produced any infinite number of different phenomena. In other words, for any given computation, we can take that computation and use it for any output whatsoever. The output is not restricted to any special output that only that particular computation can produce. There is no 1 to 1 correlation between a computation and the use of the output for that computation. And the same can be said for the input. That input didn’t need to come from the experimenter’s computer. It could have come from the pressure and temperature transducers located around an air separation plant. Or they could have come from the change machine and timers in a laundrymat. The input isn’t intrinsic to the computation in any meaningful way.

Now we put these two together. There are all these different computations that produce the same qualia. And any given computation can be made to do any number of different things given a specific input and output. This seems to indicate that there is nothing intrinsic about a computation such that there exists a 1 to 1 relationship between a given experience and the computation on which that experience supervenes.

Dennett argues that there is nothing intrinsic to qualia and comes to the same conclusion, though he uses ‘intuition pumps’ to try and convince you that he’s right. I guess I don’t accept his ‘intuition pumps’. I agree with his conclusions however where he states:

So when we look one last time at our original characterization of qualia, as ineffable, intrinsic, private, directly apprehensible properties of experience, we find that there is nothing to fill the bill. In their place are relatively or practically ineffable public properties we can refer to indirectly via reference to our private property-detectors-- private only in the sense of idiosyncratic. And insofar as we wish to cling to our subjective authority about the occurrence within us of states of certain types or with certain properties, we can have some authority--not infallibility or incorrigibility, but something better than sheer guessing--but only if we restrict ourselves to relational, extrinsic properties like the power of certain internal states of ours to provoke acts of apparent re- identification. So contrary to what seems obvious at first blush, there simply are no qualia at all.

I should clarify my interpretation of that last sentence. Dennett doesn’t seem to mean that qualia don’t exist as we might literally take that sentence. He means they can’t be intrinsic, among other things. He’s saying that qualia can’t have the properties we perceive them as having. He’s saying one person’s experience of red for example, can’t be the same, seemingly stable experience every day forever and that those qualia in general must vary depending on what computation is being performed. There is nothing special (intrinsic to nature) about a given qualia.

 

[Pythagorean]

Hi Pythagorean. My apologies in advance for the length.

Good evening Q_Goest, no worries about the length. I welcome your comments.

Note that in the following, I’ll be using a fundamental assumption that both you and Dennett appear to agree on, that being computationalism. That is, that the interactions of neurons is what produces the emergent properties of conscious experience.

I'm reluctant to label myself that specifically yet. I don't completely understand computationalism yet. There are definitely aspects of computation I agree with (especially what you mentioned above). But the definition you gave above is only one defining aspect of computationalism. Alone it seems to be a narrow view of physicalism.

I say narrow because it's not necessarily neurons doing all the work. Neuroscience is more recently beginning to recognize the role glial cells play (says my neurobiology instructor; I always thought they were merely a support system myself) though it's new enough not to be in our textbook.

We do know that we can represent a neuron with an equivalent circuit. But I don't have any reason to think that this is the only important aspect of what's going on (though, we must admit that it plays a significant role).

What I can say about computationalism is that the brain seems to be very well capable of it. But does that mean it's limited to it? I don't think so. At least, not in the traditional input/output sense.

I mention this because I think it’s important to understand the basic assumptions that go into an argument such as Dennett provides because different assumptions often lead to different conclusions, and the conclusions Dennett comes to seem to indicate there’s a fundamental flaw in our notions of what conscious experience is.

We agree on this. As Dennett said in one of his presentations "Everybody thinks they're an expert on consciousness". Everyone has their definition.

Kristof Koch agrees. One of the way he is confronting that problem is by proposing several different kinds of consciousness that exist simultaneously in a single brain.

Specifically, he means there are certain assumptions we make about our own experiences that do not coincide with a logical analysis of those experiences. Those assumptions he outlines as follows:

Just an FYI. I can only comment on the ones you comment on as I haven't heard of Quinning Qualia and I'm not completely sure of the way those terms are going to be applied to consciousness or your interpretation of it until you open discussion about it. Obviously, "intrinsic" alone doesn't tell me much, since I've used that term in a number of different ways in my undergrad degree, but since you've extrapolated here:

Intrinsic in the sense that the correlation is intrinsic to nature. Every time that particular computation is performed, the experience will follow, and be supervenient on that physical substrate that is performing the computation. So when we flipped the plug 180 degrees, then if we believe the computation is intrinsic, then we believe that with this reversed input, the experience will also be flipped.

If you read my links, you might understand why I view this is a harmful (in terms of productive discussion) oversimplification. It assumes a reversible connection... I don't know what this correlates to neurally, or if it's even a valid correlation of anything in the neural sense.

Now let’s go back to the thought experiment. This:
INPUT --- COMPUTATION --- OUTPUT

Produces the same output for any given input as this:
INPUT -x- COMPUTATION -x- OUTPUT

Similarly, we must also assume that every person has a slightly different computation going on inside their brain. So let’s say person 1 has this:
INPUT --- COMPUTATION(1) --- OUTPUT

And person 2 has this:
INPUT --- COMPUTATION(2) --- OUTPUT

Where COMPUTATION(1) is not equal to COMPUTATION(2). We can safely assume this because of multiple realizability. None of our brains are identical, but if we want to believe that qualia are intrinsic, then we must accept that my red is the same experience as your red. Now how many different people are possible? There are roughly 7 billion people on the Earth today and I’m guessing billions more who have already passed away. I don’t know how many different people there could possibly be, but I’m guessing it is many orders of magnitude larger than 7 billion, resulting in there being hundreds of billions or perhaps millions of billions of computations that produce the same intrinsic experience of a given color. And if we believe that a small, evolutionary step backwards or forwards does not change the experience, then we have to assume that there are also an array of different brains such as Cro-Magnon, Neanderthal and perhaps Orangutan brains must also experience the same intrinsic qualia that we homo sapiens experience. We could continue to extend this to other animals, so the number of different brains that potentially produce the same experience of the color red for example, must begin to approach an insanely huge number. If qualia are intrinsic, then we have to accept that the number of different computations that will produce the same experience is huge. I won’t say infinite, because I can’t be sure of that.

I think we can be sure that it's infinite, since we're capable of recursive operations. However, this doesn't concern me since it's not the individual signals themselves that matter, it's the classification of signals.

"You never cross the same river twice.".

In the same way, you never listen to the same set of audio signals twice. You may play a song twice in a row, and recognize it as the same song, but because of the inherent non-linearity of the real world, you'll never receive the signal in the exact same way.

But this is fine! We have an excellent classification and generalization system in terms of computation. It also leads to problems, of course (stereotyping and prejudicing, faulty assumptions when comparing things that are have similarities... such as the gravitational and electromagnetic fields for example).

I have some suggestive evidence that colors are, in fact, intrinsic. The obvious example is black and white. If you do not agree, I will extrapolate, but I assume most philosopher who speak of qualia and color have an argument in their pocket for this since they never bring up black vs. white and the point seems to be self-evident to me. By not explaining it, I hope to prove it's self-evidence as well.

Here's my suggestive evidence:
Babies prefer the colors black, white, and red. They are drawn to them because of the contrast (in an anecdote way, I can confirm this as a father of two months now) between them. This is important to me. If everything was all one color, would we see color? I don't think we would even have a word for it. There would be no evolutionary purpose for sure. Perhaps the significance then, is in the comparison of one color to another.

Humans also tend to associated common emotional terms with colors:
red - anger and sex... intensity
black -depression
yellow - happiness (I've actually heard annoying here too
(more in the link)
http://en.wikipedia.org/wiki/Color_symbolism_and_psychology

Also, evolutionarily, the significance of color in nature allows animals to be afraid of poisonous things. It even allows animals that aren't poisonous to pretend they're poisonous (cuttle fish are one example).

The taste of sweetness signifies that something is good for you (remember this is evolutionarily... our taste buds haven't evolved to our modern day where sugar is mass produced enough to be bad for you. Sugar was much more rare in our evolutionary history)

To make matters worse, each brain is also plastic, rewiring itself as time proceeds. Yet even with this rewiring, if qualia are intrinsic, then the these new computations that are taking place as time goes on, must also produce the same experience inside the same, rewired brain. This includes rewiring of inputs and outputs.

Of course, this is fine in the sense of a neural network learning, classifying, and generalizing to create a world model based on its inputs.

I wouldn't say that they must produce the same experience, I would say the same class of experience. Also, are you sure that the plasticity occurs in the part of the brain where experience is concerned? Or is the plasticity a response to the experience? I.e., we remember to avoid negative experiences and do so by classifying and predicting them in future experiences. The way we react to the experience itself may change, (and so we have a new after-experience). For instance, we may feel disappointment that we managed to have the same crappy experience a second time.

But you can't conveniently choose where plasticity occurs for the purpose of your argument, especially when neuroscientists themselves don't know where consciousness "is" in the brain yet. Plasticity may not affect the base experience of a sensation, only the emotional reactions to that base experience. But even if it does, all that remains necessary for me is that the experience is of the same class.

As an aside, plasticity is reduced in a lot of the brain for adults. Strangely enough, our olfactory system is one of the regions of the brain where plasticity remains in adults. Of course we used to believe that all plasticity stopped in the adult brain, which we now know is not true. However, it's not near as plastic as the youth so I just want to make sure no one has gotten that impression.

It's late here now and I'm tired, so I should probably respond to the rest of your post at a later time, unless I've changed the direction of our debate in which case you'll want to reply to what I've already said so far.

Cheers,
Pythagorean

 

[Simon76]

Little random, and no particular reason, and this is probably a silly question...

If for example you're flooded by ultraviolet light - lots of it, as per typical nightclub, I'm told white glows brightly. Is ultraviolet a misleading name however? Would a person usually be left with a purple/blue blur / afterimage / tinge to everything, or only see white clothing glowing brightly?

I'm assuming the latter, as ultraviolet is invisible to the human eye?

Also, colour banding is gradual - might there be a wavelength used for some ultraviolet lights that happens to sit closer to the blue band and therefore may be weakly picked up by the eye?

(Same for infrared)

Thanks

 

[binbots]

Everyone may see there own versions of colour. When we all agree something is green, all we are agreeing on is the name of the colour green. But my green may look different than your green, but we both agree it is green. A persons truth to them is absolute truth.