According to Millward (1981), a relatively complex learning problem may be seen to consist of two kinds of knowledge: descriptive, and procedural. According to Rohwer (1980), effective learners often construct relationship among information items by elaborating events in which the items are integral components. Related to both of the above, a new teaching method based on a set-theoretical model has been introduced. This model is besed upon the concept of a binary notation:
a R b
where
Applying this to Chinese characters, we have
With reference to our model, the attributes a and b are the descriptive knowledge, while R represents the procedural knowledge. For example, take the character DAWN (Fig. 4.6b), for which:
It is the relation, R = above, that connects a and b to form an "event" (Rohwer, 1980), namely:
The Sun (is) above the horizon.
Therefore, the beginning of the day, DAWN.
In Rohwer's model of elabotative conception, the basic units involved are "events". But, what is the process by which associative information is encoded into events? And, what is the process of retrieval? (i.e., how are previously stored events reactivated?) The retrieval of associative information depends on the generation of cues that are germane to previously encoded events. Such event-related cues may be either sel-generated or generated in response to external reminders. Two predictions follow:
(i) "The effect of providing elaborative prompts at encoding should be magnified when study-list items are presented as reminders during recall tests."
(ii) "In the absence of external reminders, as in free recall, prior elaborative encoding should not enhance initial access to study-list groups, but should increase the number of items recalled when at least one member of such a group is retrieved."
Rohwer further pointed out that two factors contribute to learner differences: the elaborative propensity, and event repertoire. Both of these vary greatly among individuals.
Effective learning hinges upon a perfect match between the organization of (i) the external knowledge structure, and (ii) the internal representation. While the former is built upon the attributes and relationships of the formal elements of a character, the latter relies on the encoding of information in the LTM.
Our purpose is to provide an instructional design externally that will fit well with the human system for knowledge acquisition. The best design is that system which is tailored to human needs and functioning.
From studying human knowledge acquisition systems, we come to understand that (1) internally we organize knowledge hierarchically, (2) stimuli from the external world are percieved via the process of "feature analysis", (3) both the bottom-up and top-down processes are used by us, (4) different means for encoding knowledge are used, namely: (a) for smaller units of knowledge in the LTM, we use the PNR, while (b) for larger and more complex units, we use "schemas" via the "family-resemblance" structure, and (5) to facilitate recall from the LTM, we employ "elaboration" and "reconstruction" techniques.
In the instructional design for Chinese characters, we intend to make use of the above knowledge and incorporate it into our instructional design features. Since a character is a 2-dimensional visual image, composed of distinctive meaningful parts, called the "formal elements", a character may be learned via both the bottom-up and the top-down processes. For smaller units of a character (i.e., the formal elements), we shall use teh PNR as an aid for learning; for larger and more complex units, we shall use the "family-resemblance" structure to integrate learning. To facilitate deeper learning, we shall use "elaboration" in our instructional material, in which are embedded the cues to facilitate recall.
For example, in Fig. 7.1 is such an instructional design, in which a character is first presented in the 2-dimensional abstract form in modern script. Its counterpart in Small Seal is then shown next to it, by its right side (Fig. 7.1a). The Small Seal representation is added because it is more revealing, both in form and in meaning. The proper order of strokes is then shown. In Fig. 7.1b is shown the unfolding process. Going through this also serves as a means of elaboration. Finally, the hierarchical tree is presented to consolidate the learning according to the set-theoretic model (Fig. 7.1c).
However, for a relatively complex character it is advantagous to apply the hierarchical analysis first (Fig. 4.10). By the hierarchical analysis, the immediate constituents (I.C.'s) are revealed, as well as the original meanings of the I.C.'s. This process is carried out down to the next level if the additional information is deemed to enhance learning. The expert's discretion is exercised here regarding the depth of the hierarchy to be revealed because too much extranous information could burden the learner. Fig. 4.9 shows the appropriate levels of the hierarchy (compare this with the PNR of Fig. 4.8). The succinctness of information presentation is obvious.
At the end of the instructional process a summary is presented to achieve an integrating effect (Fig. 4.10). The whole process, in fact, is an act of "depth processing", which reinforces later recall.
Fig. 7.2 shows the instructional design for the set of related characters headed by the derived signific TWO: {THREE, KING, PRINCE}. Fig. 7.3 is an instructional design for the set starting with the derived signific RAIN, through other necessary components, to the culmination of the compound character SPIRIT. Fig. 7.4 shows yet another series that started with the most basic components to the final character FROST. In short, these are examples to show the practically of the UNFOLDMENT method. It also takes advantage of the iconic nature of some of the atomic characters by illustrating their stages of evolution whenever possible. These pictorial attributes can be implemented with the video disk technology to give dramatic effects.
In Electrical Engineering, when working with transmission lines, one is often concerned with the proper values of (1) the "source impendance", (2) the "load Impendance", and (3) the "impendance of the line". It is well known that if these impendances are not matched at either and end of the line, then a "discontinuity" exists. A discontinuity causes distortion of the signals transmitted, as well as energy losses. The best condition can be realized by a process called "impedance-matching".
This "impedance-matching" concept may be borrowed and applied to the design of the external knowledge system such that it matches the internal system of the human being. In other words, we desire to have the external "data structure" to match the internal "knowledge structure" of the learner.
The internal and the external system may be viewed as two information processing systems, of which each has its own "data format". It is well known that if the data formats are different in a real computing environment, data conversion must take place first, before any communication is possible between the two systems. However, such a data conversion process is most time- and energy-consuming, as well as error-phone. In the case of the human system, there are incurred the additional expenditures of frustration and anxiety.
While redundant paths increase the probability of recall, competing paths also tend to reduce the speed of retrieval. Therefore, in our instructional design, we shall make use of the PNR diagramm as a tool for reducing the number of the competing paths to a minimum.
We envision that when the instructional material is properly designed (i.e., it matches how the human functions internally in knowledge acquisition), then the learner should be able to learn with great ease and efficiency.
A new teaching method was created as the result of this research. Pertaining to this new teaching method, we have made an instructional design using a set of cards on related characters to facilitate learning. In this design, we have incorporated numerous design features that take advantage of the cognitive, psychological research findings.
This new learning method possesses a number of advantages. It incorporates Ausubel's Concept Learning Theory, and Millward's Computational Theory. In addition, the new method was built upon Rohwer's model of elaborative conception for effective teaching.
Furthermore, by combining the knowledge gained from the external system for Chinese orthography and the internal system for knowledge acquisition, we have designed a new teaching method which was tested in a cognitive psychological experiment, and found to be superior.