Gaming@School: Combining Virtual Learning Environments and Intelligent Pedagogical Agents for a new Approach to Classroom Lessons Annalisa Terracina and Massimo Mecella Dipartimento di Ingegneria Informatica, Automatica e Gestionale Antonio Ruberti, Sapienza Università di Roma, Italy [email protected] [email protected] Abstract: The term “digital native” coined by Prensky in 2001‐perfectly describes the nowadays students that continuously deal with technology. Thus, technology should be a prominent part of the learning process and should be intended as a support for teachers and learners. In this work, we propose a constructivist approach in which youngsters are called to be the main actors of the learning process and in which a personal construction of their knowledge is a must, starting from their needs and their motivations. Role playing games fit very well these requirements. The approach that we propose in this work combines two fundamental aspects (that commonly are not combined): the integration of Intelligent Pedagogical Agents ‐ IPAs (that provide personalized instruction, increase learner motivation, and act pedagogically on behalf of the learner) in Virtual Learning Environments – VLEs (that add value to the educational process by giving new possibilities and computational‐richness support). In this work, we propose that each student has her own IPA (on her own personal device) that guides her throughout the role playing game. Keywords: role playing game, intelligent pedagogical agents, virtual learning environment 1. Introduction Prensky coined the term digital native more than ten years ago: “Our students today are all native speakers of the digital language of computers, video games and the Internet". We are persuaded that this term still describes a generation of students that grown up tight to technology and that somehow suffer from the lack of technology in classroom lessons. In this paper we propose an approach and a conceptual architecture of a supporting teaching tool that take into account two main objectives in new teaching trends (cf. Soliman et al (2010)): Virtual Learning Environments (VLEs) Intelligent Pedagogical Agents (IPAs) The main idea is a VLE that in turn is a role playing game. The role playing game is a social game in which each student becomes a player with her abilities and her tasks. In order to succeed, all the players should work to achieve a common objective/goal. The storyboard is designed in a way that there is an evolution in the role playing game and a progress in the level of learning as well. The idea of helping students in the process of learning in a different way with respect to the classical approach finds support in many psychological studies and previous work, in particular we refer to Howard Gardner theory (1983): “We might think of the topic as a room with at least five doors or entry points into it. Students vary as to which entry point is most appropriate for them and which routes are most comfortable to follow once they have gained initial access to the room. Awareness of these entry points can help the teacher introduce new materials in ways in which they can be easily grasped by a range of students; then, as students explore other entry points, they have the chance to develop those multiple perspectives that are the best antidote to stereotypical thinking”. Table 1 reports the five entry points theorized by Gardner: for each one of them we give a link with the adoption of a role playing game (second column). In the third column, a concrete example of a possible actuation is given. Notably, in 1999 Gardner added a sixth entry point: Social ‐ Use group settings, role‐play and collaborative arrangements, which perfectly complies with our proposed approach. In the above scenario, students face with numerous learning opportunities and therefore require intelligent support and guidance. The use of IPAs is proposed as support during the game evolution and each student has 843 Annalisa Terracina and Massimo Mecella its own IPA: they act as learning facilitators and guide the learners in the virtual environment, by explaining topics, answering questions, giving feedbacks, helping the learners to collaborate with others, providing personalized learning support. IPA combines different abilities including intelligence and pedagogical orientation; they are autonomous and not directly guided by users. As suggested by Woolf at al. (2013), one of the Artificial Intelligence (AI) grand challenges in education is “Mentors for every learner”. The rest of this paper is as follows: Section 2 introduces the platform from the user point of view (external), whereas Section 3 describes the conceptual architecture (internal). Section 4 compares our proposal with the relevant state of the art. Some concluding remarks will be finally presented in Section 5. Table 1: Gardner’s theory of five entry points and role‐playing game Howard Gardner Five Entry Points Narration entry point (read or tell a story) Logical‐quantitative entry point (provide data, use deductive reasoning, examine numbers, narrative plot structure, cause and effect relationship) Philosophical entry point (big questions about reasoning and the way of reasoning) Why role playing game The VLE in the game coincides with a plot (tell a story) that evolves during the game During the game, students should solve problems and specific assigned task. Aesthetic entry point (emphasize sensory, activate aesthetic sensitivities) Students should consider pros and cons of every possible solution. They should discuss all together and understand the implications of their choices. The information/social space is explicitly represented as a 3D immersive world Experimental entry point (hands‐on‐ approach, dealing directly with materials, simulation, personal explanations) The game requires that students take actively part in the story by solving problems and finding solutions. The team discussion is also a must. Examples (based on the same storyboard) Conquer an exoplanet and colonize it Form a team and solve the following single task: (a) how long the journey will last; (b) how much fuel is needed; (c) gravity on the new planet (i.e., the exoplanet) Evaluate the consequences of colonizing a planet: (a) what to do if the planet is already inhabited; (b) how to protect the local environment The 3D immersive world can be partially customized and students can choose the preferred configuration Simulate a different gravity on the new planet and: (a) discover which force we need to lift a stone; (b) how heavy we are 2. Using the approach and the platform The envisioned VLE partly runs on a central server (e.g., an Interactive Whiteboard) and partly on mobile devices directly provided by the school or owned by the students themselves. As shown in Figure 1, the teacher introduces the scenario to the students and explains the problems that they have to solve during the game. After that, the teacher designates a master that behaves at the same level as the teacher following the approach in which a student can “learn how to learn by teaching” (Clements 1996); in each session of the game, the master should be a different student, so everyone can experience a role of greater responsibility. Then, the master with the help of the teacher, can form teams and assign a specific role to each student. The student, from now on, becomes a player with her specific role and her own task as well (depending on the level of the student). In this phase, an IPA is assigned to each student/player that will drive her all along the game. The relation between the student and her IPA should progress all along two paths: the learning aspect (giving tips and advices related to the topics and to the tasks assigned) and the emotional/pedagogy one (the interaction depends on the feelings of the student). Each student is assigned a task and each single contribution allows her to solve the more general complex problem. The student should solve the task possibly on her own at home (homework) or during classroom lessons, depending on how the teacher would like to organize the work. The player is invited to share her solution and discuss it with the others. When the team achieves a solution, the master can verify it, and if the provided solution is correct, the game proceeds to the next level. 844 Annalisa Terracina and Massimo Mecella Figure 1: Actors involved, their roles and the actions to initiate the game 3. The architecture Figure 2 shows the three main components that run on the central platform, namely the Game Engine (GE), the Intelligent Pedagogical Engine (IPE), and the Teacher Workplace (TW). For each student/player, there are the corresponding components on her personal device, namely the Game Individual Task Handler (GITH), the IPA and the Student Workplace (SW). VLE and devices are connected through the Internet. Figure 2: Role playing game architecture 845 Annalisa Terracina and Massimo Mecella 4. Related work A major concern of e‐learning systems over traditional methods is the lack of face‐to‐face interactions that can be filled up by the use of IPA. Recent use of 3D immersive virtual environments has shown effectiveness in improving the motivation towards learning, and therefore, combining both approaches can provide significant contributions to solve the mentioned deficiency, as stated by Soliman et al (2010). He also argues that research has not yet exploited the potential of the use of intelligent agents in VLEs; indeed most of the work on pedagogical agents has focused only on the visual appearance and interface with the learner as a character. Conversely, in our approach we aim at exploiting the potential of intelligent agents (intelligence through AI reasoning techniques, social abilities, intelligent resource location, negotiation, etc.). The term Virtual Learning Environment has been sometimes abused. According to Dillenbourg (2000), a VLE must comply with the following rules: 1. “The information space has been designed. 2. Educational interactions occur in the environment, turning spaces into places. 3. The information/social space is explicitly represented. The representation varies from text to 3D immersive worlds. 4. Students are not only active, but also actors. They co‐construct the virtual space. 5. Virtual learning environments are not restricted to distance education. They also enrich classroom activities. 6. Virtual learning environments integrate heterogeneous technologies and multiple pedagogical approaches. 7. Most virtual environments overlap with physical environments.” In our proposal, we surely comply with the points 1 to 6. Moreover, in traditional tutoring systems, the character is assigned to a tutor only, whereas in our proposal a character is associated with each learner, in an immersive 3D space. 3D virtual worlds such as Second Life, Sun Wonderland, and EDUSIM have been used for educational purposes as VLEs (cf. Kluge et al (2008)). Therefore, in line with our proposal, they provide examples of the possible benefits of collaborative learning scenarios (see also Chang et al (2009) and Gutl et al (2009)). 5. Conclusion Nowadays, we deal with digital native students, and therefore we need to follow them on their digital requirements; on the other side, learning remains a slow process that requires curiosity, devotion and time to absorb and elaborate information. From a pedagogical point of view, a role playing game, even if using a virtual learning environment, provides enough time to elaborate and think to problems and solutions. In addition it gives a key role to collaboration and cooperation that sometimes in school are not sufficiently considered, limiting solely to individual results. In this paper, we have proposed an approach and the supporting platform integrating Virtual Learning Environments and Intelligent Pedagogical Agents. The platform is currently under development, and the approach will be evaluated by experimentally using it in classes, especially for teaching STEM disciplines (physics in particular). References Chang V., Gütl C., Kopeinik S., Williams R. (2009). Evaluation of Collaborative Learning Settings in 3D Virtual Worlds. International. Journal of Emerging Technologies in Learning (iJET), 4. DOI: 10.3991/ijet.v4s3.1112 Clements P.C. (1996). A Survey of Architecture Description Languages. In Proc. 8th International Workshop on Software Specification and Design. Dillenbourg P. (2000). Virtual Learning Environments. EUN Conference 2000. Gardner, H. (1999). Multiple approaches to understanding. Instructional design theories and models: A new paradigm of instructional theory, volume 2, 69‐89. edited by Charles M. Reigeluth Johnson W.L. (2003). Interaction Tactics for Socially Intelligent Pedagogical Agents. In Proc. 8th International Conference on Intelligent User Interfaces. Kluge S., Riley E. (2008). Teaching in Virtual Worlds: Opportunities and Challenges. Issues in Informing Science and Information Technology, vol. 5 Lester J.C., Converse S.A., Kahler S.E., Barlow S.T., Stone B.A., Bhogal R.S. (1997). The persona effect: Affective impact of animated pedagogical agents. In Proc. CHI’97, pp. 359‐366 Prensky M. (2014). Digital Natives, Digital Immigrants. Available at: www.marcprensky.com (accessed February 20, 2014). 846 Annalisa Terracina and Massimo Mecella Soliman M., Guetl C. (2010). Intelligent Pedagogical Agents in Immersive Virtual Learning Environments: A Review. In MIPRO 2010 Proceedings of the 33rd International Convention, pp. 827‐832. Opatija, Croatia. IEEE. Strickland C. The five entry point of Howard Gardner. Available at: http://dilangley.wikispaces.com/file/view/bENTRY%20POINTS%202009%20sec.pdf/131349357/bENTRY%20POINTS% 202009%20sec.pdf (accessed May 7, 2014) Wanga, Johnson, Mayerc, et al(2008), “The Politeness Effect: Pedagogical Agents And Learning Outcomes, ”International Journal on Human‐Computer Studies,Vol. 66. Woolf P.B., Lane H.C., Chaudhri V.K., Kolodner J.L. (2013). AI Grand Challenges for Education. AI Magazine, 34(4) 847