A Rough Sketch of my Theory of Aesthetic Affect--from circa 1990

Back in the spring of my 26th year I was mulling over an article I'd read somewhere on memory when I became convinced that to understand memory would be to understand the entire workings of the human mind.. I have therefore over the years spent much time thinking about memory. In fact, in an earlier draft of this chapter, I used several pages to describe the theory that resulted, certain it would provide, in the most dazzling manner possible, the Final Neurophysiological Basis for all I've so far said about poetry, and all I would later say. Well, those pages so baffled the readers of my first final draft--and me as weIl, if the truth be known--that I decided to withdraw them. They will, however, reappear (revised) in a later book. I can't remain entirely silent about my theory, though. Otherwise my forthcoming discussion of aaaesthetic affect, and how it relates to poetry, will make little sense.

So, on to the "master-cells" (or "m-cells") in the brain which my theory places at the center of the human memory-process (or "retroception"). These cells are connected to sensors, such as the light-sensitive rods and cones in the retina, which are turned on by exposure to certain environmental stimuli. Once on, a sensor activates its m-cell, whereupon the m-cell distributes energy to other m-cells (and, occasionally, back to itself). This m-cell-to-m-cell energy can also activate an m-cell. Whenever an m-cell is active, the mind will experience its state as a sensation. The sensation arising from a given m-cell's activation will always be the same sensation, a sensation unique to it. That will be the case whether the cell has been activated by sensor or m-cell energy. I call sensor-mediated sensations "percepts," however, to distinguish them from m-cell- mediated sensations, which I call, "retrocepts." (A given sensation's context is generally what the mind uses to tell whether the sensation is one or the other, I might add.)

Many, perhaps most, percepts and retrocepts occur randomly and transiently. But some tend to occur together often enough for the m-cells responsible for them to form what I call a "knowlecule" (or "NAH luh kyool"). A knowlecule is a cerebral representation of some significant piece of knowledge such as the image of a car, a cat, an apple or one's Aunt Jenny--or, for that matter, the idea of knowledge itself.

For my purposes here, the main thing to understand, is that a knowlecule is a unified group of m-cells that can be perceptually or retroceptually activated (or both), whereupon it will transmit stimulation to other knowlecules in an attempt to activate them as memories. It does this via a chain of storage-cells (s-cells) called the "mnemoduct" which is responsible for routing m-cell stimulation. How it does this is, of course, the crux of my theory--but too complicated to get into here. It isn't necessary to know anything about it to follow my theory of aesthetic affect, anyway.

According to that theory, each knowlecule, in effect, tries to predict what will follow it in the awareness. Its success in doing so determines how pleasurably, or painfully, one experiences what actually follows. If a given knowlecule predicts what ensues too strongly, the result will be boredom; if it fails to predict it strongly enough, the result will be pain. If it predicts it neither too strongly nor too weakly, however, pleasure will result.

For example, if I heard the name "Laura," those of my m-cells active as a result might try to rouse a memory of my niece Laura's appearance; they would do this by sending energy to cells involved in imaging blue eyes and the other main particulars of my niece's visual appearance. Three outcomes would then be possible: (l) the energy the auditory knowlecule, "Laura," caused to be dispersed could succeed in activating a memory; (2) that energy could fail to do this but the environment present a picture of Laura, or the girl in person, and I would experience an image of Laura anyway; or (3) both the energy and the environment together could fail to provide me with an image of Laura. I would, to summarize, remember what Laura looks like, or be shown, or neither. In cases (1) and (2) the spoken word "Laura" would, in a manner of speaking, have predicted the visual image of Laura which followed. The probable result would be pleasure. In case (3), however, the word would have failed to predict what followed it and I would probably have experienced pain.

I have, of course, grossly over-simplified the matter. The word "Laura" would undoubtedly have tried for many more memories than that of Laura's visual appearance, and some of them would undoubtedly have become active. On the other hand, any image of Laura that came into my awareness would not likely have exactly matched what was "predicted" if certain cells would have gotten energy but failed to become active. And the environment would certainly have contained elements unlooked for which would have added unpredicted material to what I experienced. All that is unimportant, however: if a given knowlecule sufficiently resembles the one the knowlecule just before it "predicted," the person involved will experience pleasure; if not, the person will experience pain, or some state in between pleasure and pain. If a knowlecule is too like what the previous knowlecule predicted, though, the result will be boredom.

To account for this in more detail I hypothesize the existence of value-points of which there are two kinds: "realization-points" and "frustration-points," or r-points and f-points. Each m-cell that receives retroceptual energy (or m-cell energy) during a given event (or instant of awareness) will release r-points or f-points depending on whether or not it is activated during the next event. (Whether it then becomes active retroceptually or perceptually, or both, incidentally, is irrelevant.) The number of value-points produced will be proportional to the amount of retroceptual energy involved.

The key to my theory is that the aaaesthetic affect produced by a given knowlecule depends simply on the value-points it causes to be produced. First a brain-center determines the number of the two kinds of value-points caused by the knowlecule's activation, then what percentage of this number consists of realization-points. If this percentage is high, the knowlecule under consideration must be boring--because a very high score indicates that it was expected--or predictable. On the other hand, if the score is low, the moment is painful, because such a score indicates the knowlecule was unexpected--or disruptive. It is only a score neither too high nor too low which causes pleasure--a score (and this is only a guess) between 50 and 60 percent, perhaps. There is one further way a person can feel about an knowlecule: indifferent. This will occur when a score is either higher or lower than optimum but neither so high nor so low as to cause boredom or pain.

To sum up, my theory of aaesthetic affect is that we automatically consider that which is too familiar to be boring, that which is familiar but not too familiar to be pleasurable, and that which is unfamiliar to be painful, and that there are levels of familiarity between the boring and the pleasurable, and between the pleasurable and the painful, which are emotionally neutral--and, I might add, probably occur far more often than any other kind.

All this, it seems to me, fits in with the fact that human beings tend to withdraw from that which is painful, shun the boring, and advance toward that which is pleasurable. If, as my theory has it, it is the under-familiar which is painful, it would make sense to withdraw from it: better to retreat from something until one has come to understand it--i.e., become familiar with it--then chance its being dangerous. It is equally sensible to embrace the familiar since, if something were not good for us, it could not generally become familiar-- it would injure or kill us first. But if we stuck with the familiar too slavishly, we would never work out cultural improvements or zestfully explore our habitat; hence the value of the over-familiar's causing boredom.

As for my theory's fit with everyday experience, surely it is a rare person who has never heard some song he considered ear-damagingly bad which, when he'd heard it a few more times, turned into a favorite of his. . . only to become, after he'd heard it too many hundreds of additional times, boring beyond endurance. Tschaikowski affects most reasonably intelligent admirers of classical music this way, but there are sundry other examples. My theory similarly accounts for the importance of simple repetition in all the arts--such as the use of symmetry in architecture, repeated phrases throughout music, from popular songs to Mozart, and the recurrence of steps in dance routines. As it also provides a plausible explanation of the pleasurable effects of simple melodation like rhyme's repetition--and the avoidance of repetition by equaphors (metaphors and the like).