言語による概念の学習に関するCAI実験 : 学習順序の自己選択

Purpose: The experiment was conducted to find the followings: (1) Some capability of giving a learner a freedom to choose learning sequence by himself when studying concepts presented literally. (2) Relation between hierarchical structure of learning material and learner's behavior. Method: Forty-six students from two universities were randomly assigned to the control and the experimental groups. Definition of matricies and laws of physics were selected as learning materials. Both of them are generally skimmed in high school mathematics II and high school physics, respectively. Computer control programs for these materials were written in the extended FORTRAN language for TOSBAC-40 TSS system. The learning materials were taken in 35mm film slides and presented by a computer controlled random slide projector. Except an introduction section of the slide series, each of these CAI course was devided into four sections; pre-experimental section, experimental learning section (learning part and practics part), application test section, and ending section. Each control group subject studied, at a CAI terminal, definition of ten terms of metrices in a set order, practiced what he had learned, and took application test. After a short break, he also studied and practiced definition of ten laws of physics in the same fashion, and took application test. The subject in the experimental group studied and practiced the same materials, but were allowed to select the order in which he saw the terms and laws defined. Time spent at each course section was recorded for individual subject as his learning activities proceed. Order of learning sequence was kept for the experimental group subjects, and responses to the test items were also stored in a computer file. Analysis was made mainly based on the stored data. Findings: (1) Behavior for selecting learning sequence was observed among two-thirds of the subjects studying material with hierarchical structure-metrices. Among those studying the material without such structure-laws of physics-, about half of the subjects demonstrated behavior of self-selection. (2) It was observed that in case of the material with hierarchical structure, the learner begins to select their sequence from basic knowledge, and that the learner seemed to have selected somewhat in random fashion if the material has no such structure. However, general pattern of selection behavior is not clear because only a small number of subjects were available for further analysis. (3) An average learning time of the given sequence (control) group was found to be almost equal to the average learning time of the selection (experimental) group. (4) Regardless of the contents of material, an average practice time of those who selected sequence was significantly shorter than that of those studied in the given order. (5) Regardless of the program contents, the average score of application test for those who made a selection of sequence seemed to be a bit higher than that of those studied according to the given order. However, the difference of these average scores was not significant. Concluding remarks: According to a traditional approach, the instructor organizes contents of his instruction in some way that seems logical to him. He may present items to be learned one by one, or he may order his presentation according to degrees of difficulty or subject matter. In either case, the instructor has a control over the sequence by which students learn the information. It may be conceivable that learners would want to deceide the order of learning differently from that taken by their instructor. The present experiment at least disclosed that selection of sequence by learners themselves (learner control) significantly shorten the necessary practice time, keeping learning time and score of application comprehension test almost equal to those of non-selection (teacher control) case.