Robot as Tutee
This paper explores the possible advantages of substituting teachable agents in a learning environment, with a humanoid robot as the non-human tutee. Teachable agents are used as an extension to educational games in order to leverage engagement, reflectio
- PDF / 93,699 Bytes
- 7 Pages / 439.37 x 666.142 pts Page_size
- 31 Downloads / 175 Views
Abstract This paper explores the possible advantages of substituting teachable agents in a learning environment, with a humanoid robot as the non-human tutee. Teachable agents are used as an extension to educational games in order to leverage engagement, reflection and learning. The learning environment is engaging and shown to be effective for learning and promote self-efficacy in experimental studies in authentic classroom settings. Features beneficial for learning which are further enhanced by a robot compared to an agent are identified. These include embodiment of the robot; a social, empathic behaviour, better conversational abilities which together provide a better role model of an ideal learner for the student to identify with. Keywords Robotics Role-model learner
⋅
Tutoring
⋅
Teachable agent
⋅
Robot tutee
⋅
1 Introduction Robots are starting to enter the classrooms, not as substitutes for human actors but as teaching tools or teaching assistants [1, 2]. According to [2], robotics is mostly used in computer science education, in domain-specific subjects such as geometry or physics where movement and spatial cognition are involved, in foreign language learning or as assistive tools for cognitive or social support, for example as storytelling assistant in pre-school [3]. Most applications are devoted to the engineering field not to the wider scope as a general educational tool [1]. However, in these classes students also learn general 21st century skills such as communication, collaboration, creative thinking, and problem-solving skills [4, 5]. Robots can play various roles in education; from design material in engineering to virtual companions in other learning situations. Our concept belongs to a learning L. Pareto (✉) University West, Media and Design, Trollhättan, Sweden e-mail: [email protected] © Springer International Publishing Switzerland 2017 M. Merdan et al. (eds.), Robotics in Education, Advances in Intelligent Systems and Computing 457, DOI 10.1007/978-3-319-42975-5_24
271
272
L. Pareto
situation where the human acts as a teacher/tutor to the virtual companion but where the purpose of the virtual companion is to act as a productive counterpart for the human learning [6]. There is a distinction between virtual companions that can be taught and those that can learn: to be teachable it is enough to appear to learn from the tutor, they do not need to ‘learn’ [7]. To simulate a tutee there are two objectives: to model a behavior that appears to progress due to teaching activities and that is understandable to the human tutor in order to monitor the teaching activity. In contrast to [8] where a robot learns social behavior from a child by machine learning, our robot tutee merely appears to learn and the focus is on features that support the human learner in becoming more competent. The aim of this paper is to propose the Student Tutor And Robot Tutee (START) learning concept, based on a virtual tutee enhanced mathematical game, which is shown effective for learning conceptual understanding and r
Data Loading...
