Design and User Preferences of Pedagogical Agents for an Intelligent Tutoring System for EFL
摘要
Intelligent language tutoring systems (ILTS) are powerful tools in improving learning performances by supporting adaptive and individualized feedback. In addition, gamification elements embedded in ILTS, such as pedagogical agents providing affective feedback, can foster learners’ motivation, a driving factor for the learning progress. In this article, we describe the design and user preferences of 17 different animal- and UK-themed pedagogical agents designed for implementation in an ILTS for seventh-grade English as a foreign language (EFL) learners. In two user studies, we investigated 91 sixth- to eighth-graders’ ratings and rankings of the agents, and their reasoning for choosing their personal favorite. Results revealed that most of our designed agents received positive ratings above the middle value of a 5-point Likert scale, and we finally selected three agents for implementation, i.e., Big Ben as the overall favorite, Mouse as the females’ favorite, and Yeti as the males’ favorite. Students argued most often for their favorite that it looks likable, cute, best drawn, funny, least childlike, and that they liked the character. In line with the expert hypothesis, UK-themed characters were most often attributed to speak English well and know a lot.