1999 Fiscal Year Final Research Report Summary
Management of Cognitive Conflict with Qualitative Reasoning Techniques in Virtual Experimental Environment
Grant-in-Aid for Scientific Research (B)
|Allocation Type||Single-year Grants |
|Research Institution||Osaka University |
TOYODA Junichi Osaka University, The Institute of Scientific and Industrial Research, Professor, 産業科学研究所, 教授 (00029456)
INABA Akiko The Institute of Scientific and Industrial Research, Osaka University, Research Associate, 産業科学研究所, 助手 (10304049)
MATSUDA Noriyuki The Institute of Scientific and Industrial Research, Osaka University, Research Associate, 産業科学研究所, 助手 (40294128)
KASHIHARA Akihiro The Institute of Scientific and Industrial Research, Osaka University, Assistant Professor, 産業科学研究所, 講師 (10243263)
HORIGUCHI Tomoya Faculty of Information Engineering, Kobe University of Mercantile Marine, Lecturer, 情報システム工学講座, 講師 (00294257)
HIRASHIMA Tsukasa Dept. of Artificial Intelligence, Kyushu Institute of Technology, Associate Professor, 情報工学部, 助教授 (10238355)
|Project Period (FY)
1997 – 1999
|Keywords||Intelligent Educational Systems / Cognitive Conflict / Qualitative Reasoning / Self-motivation / Behavior Simulation / Error Visualization / Metaphor|
In the field of intelligent tutoring systems, how to motivate learners has been one of central issues recently. Many researchers insist that creation of cognitive conflict is an effective way. It occurs when a phenomenon in real world is different from a learner's prediction. Behavior simulation is a promising method to do this. A simulation which reflects a learner's error can be different from the simulation of correct behavior which he/she predicted. The purpose of our research is to establish the method how to create conflict effectively by using behavior simulation. The followings are necessary : (1) to generate unreal simulation based on a learner's erroneous formula, (2) to identify the mechanism by which human differentiate two phenomena and to automatize it, (3) to evaluate whether the "different" phenomena really cause cognitive conflict and to implement this mechanism.
In the first year, task (1) has been performed. The difficulty is that the violation of constraint in learne
rs' formulae is too various to generate simulation by the unified method. We solved this problem by limiting the variation of errors to deal with. The simulator which generate unreal simulation has been implemented and evaluated.
In the second year, we worked at task (2). The criteria which human is supposed to use to differentiate two phenomena is assumed and verified by cognitive experiments. The mechanism which diagnoses the difference with this criteria is proposed (by using qualitative reasoning). It has been implemented and evaluated.
In the last year, we accomplished task (3). Various factors which affect the total performance of simulation (i. e. whether it cause cognitive conflict) are structurized by introducing two viewpoints. One of them concerns the visual effect of simulation and the other its method of expression. Each factor's way of influence has been verified by cognitive experiments. Based on these analyses, we have constructed the mechanism which controls behavior simulations considering their educational effectiveness. The implemented system has been evaluated, and the result was mostly supportable.
Through this research project, the method how to create cognitive conflict effectively by using behavior simulation has been established, and it gives a guideline to utilize virtual experimental environments of great potential, we believe. Less
Research Products (22 results)