Abstract
This paper describes a design and practice of a student experiment on “artificial intelligence (AI).” AI has a wide range of topics that are independent of each other so that it is difficult for students to understand relationships between the topics. To improve this problem we designed our AI experiment based on Minsky's Mental Activity Model, which has six layers each of which expresses a level of human mental activity. By associating each week's topic of the experiment to the layer, students can easily grasp the relationship between the topics. In the student experiment, students program a mobile agent both with an actual equipment (e-puck) and in the simulator environment. The student experiment conducted for three years and we did a questionnaire each year. The result shows most of the students understood an association between the weeks' topics and Minsky's model layers, which also indicates they grasped the relationship between the AI topics.
