Thursday, March 13, 2008

The integration problem

In this post, I had hinted at the "integration problem" towards the end. I wish to address it in more detail here. Although I myself am sympathetic to subjective idealism, I cannot sometimes help wondering if the hard problem is for real, and whether the terra firma assumption is true.

Even if we assume that the hard problem might itself not lend itself to an answer, there is no mistaking the fact that there are neural correlates to subjective experiences. This is the basis for the reason mind-altering drugs and even alcohol work. I foresee that in the future we can develop technology that can extend our sensory experiences to heights beyond even the reach of currently available mind-altering drugs, and at the same time, having a hundred percent safety factor. Also, such altered experiences can be brought back to "normal" almost instantaneously through the flick of a switch. In short, a hedonist's holy grail come true.

I predict that within the next half century or so, there would be a new field of science/philosophy called "Qualia Engineering" (QE) and this would explore how neural processes relate to actual qualia perceived, and would spawn a vast plethora of "virtual drug" machines. One of the obstacles to be overcome is how to interface our organic brains to these machines in a non-invasive way.

Once the ball of QE gets rolling, it would spawn exciting theoretical advances in the same manner that quantum mechanics did in the 1920's and 30's. Who knows, even my "no tinted glasses conjecture" might get to be proved right (or wrong).

Please note that QE as a field might not come anywhere closer to solving the "hard problem". And the presence of the "hard problem" as a philosophical conundrum should in no way deter the development of this field. An analogy is the philosophical question of why the immensely successful Standard model of physics is correct, which might probably never get answered. I however do get the feeling that QE will have a lot to do with quantum mechanics.

Having made the above general prediction, I wish to now address some of the concepts that I think are relevant.

The first is that, QE would be able to create consciousness! After all, qualia, by definition, are stuff perceived by conscious entities.

If QE can indeed produce self-contained conscious entities, how is it going to help us extend our sensory experiences? This brings us to the integration problem. As I had hypothesized in previous posts, each conscious entity is associated with a 3-dimensional binding space called the common binding space (CBS). One can unify two different conscious entities by unifying their CBSs into one. In our own brains, the corpus callosum does this. The integration problem specifically deals with how the CBS of our brain gets unified with the CBS of an external device, so that we ourselves get to experience the out-of-world experiences that are theoretically impossible with our brains as such. It is not known as of present whether the corpus callosum has a classical or quantum mechanical role to play. Even if it is the latter, it may not be a great practical hindrance in integrating external CBS's with our own. Heck, a lot of the electronics these days borders on the quantum mechanical anyway.

I also wish to bring up the concept of an eigenhue recognition unit (ERU), a fundamental unit of some sort that needs to be "excited" for us to perceive a quale of unit intensity at some point in the CBS. I believe that there is an associated density function which depends on the spatial location of the CBS and sets the limit for the maximum intensity of a particular eigenhue at a particular spot in the CBS. For example, in the case of vision, the density function is zero towards the back of your head, but not so for the case of sound, where it is possible to localize sound coming from there. In addition, there is also a degree of fuzziness, given by the standard deviation for each point of the CBS that sets the limit on how precisely we can localize any particular eigenhue in the CBS. (This standard deviation is not isotropic, and can have different values in different directions.) For example, a light source can be highly localized, but not sound.

I also conjecture that in addition to the "normal" excited state of an ERU, there also exists at least an additional state called the "imagination state". This allows us to imagine sensory experiences where there is none, or even something contradictory. For example, when we see a blue lamp, we can imagine it glowing red and even draw comparisons between our imagined red lamp and another real one as to whether they have the same hue, or which is brighter. Since we are seeing blue during the whole process, it seems that the imagination state can co-exist with the "normal" excited state, which probably implies that they are orthogonal and can exist as a superposition.

I had also suggested in this post that the density function applies to imagination states as much as real ones. Thus, one cannot just see things to the back of one's head, one cannot imagine them as well. I have called the non-zero values of the density function the access for a particular quale in the CBS. We are unable to experience or even imagine any sensory experience which exceeds the density function value at any point in the CBS. And we would be unable to experience or even imagine a very bright light - there are just not enough ERUs available to do the job. And in a similar manner, the associated standard deviation function sets the lower limit for the spatial resolution for a particular quale. We can say we don't have any access to resolving an A0 piano note to within an inch.

There are evolutionary and physical reasons for both intensity and spatial resolution access limitations. The upper limit for intensity is set by the intended behavioral response that a particular quale elicits. Too much can be counter-productive. For example, consider pain. A certain amount of pain enables an animal to keep away from harmful stimuli. But too much can cause it to get traumatized to such an extent that it behaves in an irrational fashion where it doesn't learn to keep away from them. If it is too much, it can even die due to the mental trauma.

In case of sound, there is no need for the quale localization to be within a fraction of an inch since the wavelengths of the sounds that cause those qualia might have wavelengths several times that amount. Thus, the spatial resolutions are in sync with the physical phenomena they represent. But this does not mean that the quale corresponding to a A0 note cannot have a resolution of a millimeter, in theory. In fact, I predict that QE would enable us to hear sounds with the same resolution as a visual image.

Besides intensity and spatial resolution access limits, I had also hinted in this post about access of operators on classes of qualia. For example, if we take notes of music theory, we have relational operators like "an octave above" or "a fifth below", and so on. It is possible that in case of tone deafness, the patients lack access to the pitch relation operator. And it is entirely possible that we lack relational operators between the eigenhues of color (red, green, and blue) when in theory they might exist. If so, QE would allow us to enjoy chromatic melody just like music.

I had also discussed the idea of platonic qualia spaces. Because of the architecture of the human brain, the range of qualia we can experience is quite limited in terms of the types, the intensities, the spatial range and resolution, and lastly, any relational operators within a class of qualia eigenhues.

So it is possible that there are entirely new colors in the platonic qualia space that we cannot experience or even imagine. QE will actually let us enjoy these new colors.

I had also hypothesized in the same post that the platonic pitch range is infinite. It might be possible, through QE, to enjoy piano solos played on a hyper grand piano having 176 keys!

And we can have orgasms like none before. Apart from 1000x the intensity of "normal" orgasms, we can also have them in any part of the CBS, like the back of your head, or a mile to the front, or both, if you wish!

And we can get to experience entirely new senses. It might actually be possible to feel what it is like to be a bat. Thomas Nagel would certainly be pleased!

The interface to the human brain would be through a device that is non-invasive or maybe minimally invasive. When one wants to return to the "normal self", one can just remove the interface and proceed as usual. Once the QE device is off, not only would we lose the capacity to experience the new expanded qualia, we would lose the ability to imagine them as well. Thus we can see and imagine a fourth color for as long as the QE device remains interfaced to our brains. The moment it gets disconnected, we wouldn't know what the fourth color is like, and we would only remember that we experienced a fourth color, but cannot really even reminisce it any longer, at least till we turn the QE device back on. This is because the ERUs as well as the associated window functions reside in the device, not in our brains.

I also feel that there might be conservation theorems regarding qualia, and we cannot develop runaway situations of hedonistic pleasures, since there is a cost associated with them, and would be governed by ethics committees. But my own conviction in this aspect is probably a lot lesser than the other points I have made.

We will have to wait and see if all this is just fantasy or something that would actually come true!