Creative understanding

  The number of tasks needed is one indication of the complexity of the reading process. It is this complexity which has made reading a highly researched area, especially since the start of the Twentieth Century. Researchers in psychology and education have attempted to codify exactly what combination of specific cognitive skills are needed to produce the behavior we call reading. Since the 1950s, researchers in the field of artificial intelligence have also attempted to model the reading process. The researchers hoped to demonstrate the validity of their reading theories by producing computer programs which could read and comprehend texts. Unfortunately, even with the large research effort devoted to reading issues, a cohesive theory still does not exist. Part of the reason for this is a seeming inability to provide a theory which covers the range of the human reading experience--in the final analysis, reading seems to be a super-cognitive activity, making heavy use of other cognitive systems, such as ``general'' reasoning and memory. Largely due to this expanse of abilities, it is often the case that a single research effort can neither provide an adequate explanation of the process nor provide any predictive theories. However, a researcher must be aware of the entirety of the reading process if any complete theory is to be produced. Since reading is such a connected web of tasks, to focus on only one aspect can potentially lead to an oversimplification of the problem.  

While it is important to be aware of the complete reading process, it would be difficult to produce a complete theory of reading from a single line of research. A particular area of focus is needed for the research, hopefully one that can take advantage of the awareness of the entirety of reading behavior. The area of   creative understanding which I have been pursuing exists at a level beyond the task breakdown; indeed, the process can be said to emerge somewhat from the interaction of the tasks. Consider the next part of Zoo:

In the past they had sometimes been treated to three-legged creatures from Venus, or tall, thin men from Mars, or even snake-like horrors from somewhere more distant. This year, as the great round ship settled slowly to earth in the huge tri-city parking area just outside of Chicago, they watched with awe as the sides slowly slid up to reveal the familiar barred cages. In them were some wild breed of nightmare--small, horse-like animals that moved with quick, jerking motions and constantly chattered in a high-pitched tongue. The citizens of Earth clustered around as Professor Hugo's crew quickly collected the waiting dollars, and soon the good Professor himself made an appearance, wearing his many-colored rainbow cape and top hat. ``Peoples of Earth,'' he called into his microphone.

The crowd's noise died down and he continued. ``Peoples of Earth, this year you see a real treat for your single dollar--the little-known horse-spider people of Kaan--brought to you across a million miles of space at great expense. Gather around, see them, study them, listen to them, tell your friends about them. But hurry! My ship can remain here only six hours!

And the crowds slowly filed by, at once horrified and fascinated by these strange creatures that looked like horses but ran up the walls of their cages like spiders. ``This is certainly worth a dollar,'' one man remarked, hurrying away. ``I'm going home to get the wife.''

All day long it went like that, until ten thousand people had filed by the barred cages set into the side of the spaceship. Then, as the six-hour limit ran out, Professor Hugo once more took microphone in hand. ``We must go now, but we will return next year on this date. And if you enjoyed our zoo this year, phone your friends in other cities about it. We will land in New York tomorrow, and next week on to London, Paris, Rome, Hong Kong, and Tokyo. Then on to other worlds!

He waved farewell to them, and as the ship rose from the ground the Earth peoples agreed that this had been the very best Zoo yet. . . .

In addition to the sheer complexity of the reading process, there is an additional problem with respect to describing it: by its nature, reading is a creative process. The average reader will often be faced with texts which contain concepts which are unknown to them. For reading comprehension to be successful, these unknown concepts must be understood. What, exactly, does a three-legged creature from Venus look like? Do they really exist or is Hoch lying to the reader? How does the reasoner reconcile the novel elements of the story with their existing background knowledge?

  In the theory described in this dissertation, creative understanding is the process by which a reasoner handles any novel concept, including those that arise during the reading process. Due to the ambiguity of natural language, the term creative understanding could have one of two meanings. It could refer to the type of understanding which takes place on novel concepts; i.e., it is utilized when normal or ``mundane'' understanding fails. While this would be one approach to handling novel concepts, it is not the one undertaken here. The philosophy followed in this work is that understanding novel concepts is done by the same process as understanding mundane concepts; rather than being on or off, understanding exists on a continuum, ranging from the mundane to the creative. Under this approach, creative understanding refers to the fact that understanding is inherently creative. This point is important to keep in mind for the rest of this work to avoid possible confusion and to avoid searching for a mundane understanding process.

To further see the need for creative understanding, finish reading the story:

Some two months and three planets later, the silver ship of Professor Hugo settled at last onto the familiar jagged rocks of Kaan, and the queer horse-spider creatures filed quickly out of their cages. Professor Hugo was there to say a few parting words, and then they scurried away in a hundred different directions, seeking their homes among the rocks.

In one, the she-creature was happy to see the return of her mate and offspring. She babbled a greeting in the strange tongue and hurried to embrace them. ``It was a long time you were gone. Was it good?''

And the he-creature nodded. ``The little one enjoyed it especially. We visited eight worlds and saw many things.''

The little one ran up the wall of the cave. ``On the place called Earth it was the best. The creatures there wear garments over skins, and they walk on two legs.''

``But isn't it dangerous?'' asked the she-creature.

``No,'' her mate answered. ``There are bars to protect us from them. We remain right in the ship. Next time you must come with us. It is well worth the nineteen commocs it costs.''

And the little one nodded. ``It was the very best Zoo ever. . . .''

    The reader of this story will need to develop a new concept of what a ``zoo'' is. Existing concepts are not sufficient to produce the proper understanding. One possibility is that the reader will use the existing idea of a ``zoo'' and draw an analogy to the kind of zoo used in the story. Another possibility is that the reader will be reminded of a safari and use that in an analogical fashion to understand the novel usage of the concept. In both of these cases, the reasoner needs to expand existing concepts in order to arrive at an understanding of the novel zoo.

The claim supported by this work is that   THE ACT OF UNDERSTANDING INVOLVES CREATIVE BEHAVIOR WHICH REQUIRES A SET OF TASKS AND A KNOWLEDGE THEORY TO EXPLAIN. This dissertation presents evidence that the knowledge elements are as necessary to the success of the research as the process side is, a fact which will be supported by my computational model. An effective knowledge system enables the processing side to act in a more efficient   manner--CONCEPTS NEED TO BE REPRESENTED AND ORGANIZED FOR EFFICIENT UNDERSTANDING. The claim, at this level, for the processing side is simply organizational--A SET OF TASKS ACT TOGETHER TO PRODUCE UNDERSTANDING. Before elaborating on these claims, however, we first explore the knowledge claims which this research makes.