DESIGNING AND PROGRAMMING GAME-BASED LEARNING WITH HUMANOID ROBOTS: A CASE STUDY OF THE MULTIMODAL “MAKE OR DO” ENGLISH GRAMMAR GAME WITH THE PEPPER ROBOT
Berliner Hochschule für Technik (GERMANY)
About this paper:
Conference name: 15th annual International Conference of Education, Research and Innovation
Dates: 7-9 November, 2022
Location: Seville, Spain
Abstract:
Game based learning is a popular approach to designing engaging educational experiences enhancing motivation and enjoyment while at the same time supporting the attainment of learning outcomes. Game-based learning with humanoid robots as a relatively new field of research and practice has the potential to enrich educational curricula, including higher education. Current designs for game-based learning with humanoid robots have been developed for different domains including adaptive spelling games for language learning, team-based educational escape rooms for history and political science classes, or serious games for enhancing waste recycling skills in the field of ecological sustainability.
This paper presents the initial design and programming of a game-based learning scenario for English grammar in university settings, in which the humanoid robot Pepper is used as a game host and companion to students on the campus. The game-based learning scenario called “Make or Do” is intended as an educational, interactive quiz in two versions, i.e. a single player version and a team-based version. The first version of the game for one-on-one play between the human and the robot, was tested with a small group of five students. The second version of the game, in which teams of students will compete against each other, is currently being designed. The paper describes the instructional design, programming and evaluation results of the first version of the “Make or Do” game. In the first version the game consists of two main parts: the interaction with the Pepper robot and the tablet application which is fixed to the chest of the robot. The Pepper robot responds to the input of the learner with vocal statements, facial expressions and head movement. The paper also introduces the planned second version of the game for teams of students and outlines the game play and the interaction flow as an extension of the first version. Both versions of the game are designed as multimodal game-based learning scenarios. While the first version is focused on the interaction through the tablet of the robot combined with speech output, gestures and body movements of the robot, the second version extends these interaction elements by the recognition of speech and non-verbal sounds as well as movement inside the classroom (forward, right, left). Both proposed game-based learning scenarios with the humanoid robot Pepper are intended to be integrated into the English language curriculum at university level. The evaluation results from the testing of the first version of the game show that students enjoyed playing the game with the Pepper robot, and perceived Pepper as a friendly, kind, pleasant, funny and relaxed interaction partner. Moreover, the results indicate that after playing the game the overall mood of the participating students improved and students felt more relaxed compared to the mood before the game. Finally, the paper discusses a number of possible improvements of the human-robot interaction in the English grammar game based on the recommendations made by students and on the observations of the development team. It is anticipated that the deployment of the second version with extended multimodal interaction capabilities of the Pepper robot and the team-based game design combined with verbal and non-verbal sound recognition and the movement of the robot in the classroom will advance student engagement with and enjoyment of English grammar learning.Keywords:
Game-based learning, human-robot interaction, Pepper robot, English language, English grammar.