Vr-engage: a virtual reality educational game that incorporates intelligence

Cover page

Submission title:
VR-ENGAGE: A Virtual Reality Educational Game that Incorporates Intelligence.

Submission type:

Authors:
Assistant Professor at the Department of Informatics,
University of Piraeus
Address:
University of Piraeus 80, Karaoli and Dimitriou St., Piraeus 185 34, Greece Ph.D. Student at the Department of Informatics,
University of Piraeus
Address:
Ph.D. Student at the Department of Informatics,
University of Piraeus
Address:
Ph.D. Student at the Department of Informatics,
University of Piraeus
Address:

Contact author: Maria
Abstract
Electronic games have become part of children’s culture. Children and adolescents are quite happy to spend many hours playing a game just for the fun of it. On the other hand educational syllabus may look quite dry to children and adolescents. In this paper, we describe VR-ENGAGE, a virtual reality game that has been constructed for teaching geography to students in a motivating way. Thus the captivating effects of electronic games may come into service for the purposes of education. The game is enriched with learner modeling mechanisms that ensure the individualization of the interaction. The game has been evaluated and the results were very encouraging. VR-ENGAGE: A Virtual Reality Educational Game
that Incorporates Intelligence
Abstract
children inside the classroom walls [6]. Indeed, a major issue is how to design an educational system that is Computer games have become part of children’s beneficial to students. Towards this end, there is a need culture. Children and adolescents are quite happy to for the integration of successful methods, ideas and spend many hours playing a game just for the fun of it. approaches of educational software technology, such as On the other hand educational syllabus may look quite the adaptivity abilities of Intelligent Tutoring Systems dry to children and adolescents. In this paper, we describe VR-ENGAGE, a virtual reality game that has ITSs have been quite good at providing dynamic been constructed for teaching geography to students in a aspects to the reasoning ability of educational motivating way. Thus the captivating effects of electronic applications. This is mainly due to their student modeling games may come into service to the purposes of component that aims at gaining an understanding of how a education. The game is enriched with student modeling student learns and what the student’s misconceptions may mechanisms that ensure the individualization of the be. Indeed, as Self [7] points out, ITSs are computer- interaction. The game has been evaluated and the results based learning systems, which attempt to adapt to the needs of learners and are therefore the only such systems, which “care” about learners in that sense. In this paper, we present a virtual reality educational 1. Introduction
game for geography. The game is called VR-ENGAGE which stands for Virtual Reality - Educational Negotiation Game on Geography. The environment of the game aims Children and adolescents are often fascinated by at increasing students’ motivation and engagement. electronic games. Indeed, it has been widely However, the game also incorporates intelligence. It has acknowledged that electronic games are part of the the main components of an ITS, namely the domain popular culture of many children [1]. Children’s fixation knowledge, the student modeling component and the with these games initially alarmed parents and educators, tutoring component. In particular, the student modeling but educational researchers soon questioned whether the component models the student’s knowledge and his/her motivation to play could be tapped and harnessed for ability to reason plausibly about domain knowledge acquired. In this way, while playing, students may practice Hence a lot of researchers have recently highlighted both their factual knowledge on geography and their the advantages of computer games relating to education. reasoning ability and thus they are led to “enjoyable” For example, Papert [3] points out that video games teach children that some forms of learning are fast-paced, The main body of this paper is organized as follows: immensely compelling and rewarding. Boyle [4] notes that In Section 2 the environment of the game is described and games can produce engagement and delight in learning. In discussed. In Section 3 we describe the story of the game, addition it has been argued [5] that games represent one which incorporates pedagogic features from educational way in which learners can be introduced into software technology. In Section 4 we describe the student constructivist micro worlds, since users do not just study a modeling component. In Section 5 we describe an particular domain but become part of the scenario. evaluation that we conducted and finally in Section 6 we However, if educators are to include electronic games give the conclusions drawn from this research. as part of the curriculum then there is a need to do much more than invite the popular electronic games culture of and stimulating them to think, a player may turn off the 2. The Virtual Reality Environment of the
sound if s/he feels that s/he is disrupted. 3. Rewards, Prizes, Threats, Negotiation and
The environment of a game plays a very important role Virtual Companions
for its popularity. Griffiths [8] after conducting a questionnaire and interview study, found that the The story of VR-ENGAGE incorporates a lot of machine’s “aura” typified by characteristics such as elements from adventure games. However, each of these music, lights, colors and noise was perceived as one of the elements is connected to ideas and pedagogic approaches machine’s most exciting features for a large part of the The ultimate goal of a player is to navigate through a The environment of VR-ENGAGE is similar to that of virtual world and find the book of wisdom, which is the popular game called “DOOM” [9] which has many hidden. To achieve the ultimate goal, the player has to be virtual theme worlds with castles and dragons that the able to go through all the passages of the virtual worlds player has to navigate through and achieve the goal of that are guarded by dragons and to obtain a score of reaching the exit. VR-ENGAGE has also many virtual points, which is higher than a predefined threshold. The worlds where the student has to navigate through. There total score is the sum of the points that the player has are mediaeval castles in foreign lands, castles under the water, corridors and passages through the fire, temples In particular, while the player is navigating through the hiding secrets, dungeons and dragons. The main virtual world, s/he finds closed doors, which are guarded similarity of VR-ENGAGE with computer games like by dragons as illustrated in the example of Figure 1. A DOOM lies in their use of a 3D-engine. However, VR- guard dragon poses a question to the player from the ENGAGE unlike DOOM and other computer games of domain of geography. If players give a correct answer this kind is not violent at all and is connected to an then they receive full points for this question and the dragon allows them to continue their way through the VR-ENGAGE communicates its messages to students door, which leads them closer to the “book of wisdom”. through animated agents or through windows that display text. When a student is asked a question s/he may type the answer in a dialog box. The user interface employs two types of animated agent, the dragon which is the virtual enemy of the player and the virtual companion of the player. Both types of animated agent use synthesized voice as well as written messages. However, their voices are different so that the player may distinguish between them. The reason why the animated agents use voice is that there are studies that show that voice messages may be more effective than written ones in the way that students react to the educational applications (e.g. [10]). In addition, it was considered important for the “aura” of the game. Players are also allowed to select whether they want background music or not. If they do, they are allowed to select the background music that they prefer from a menu. The reason why in VR-ENGAGE there is a high degree of choice for the status of the background music is that there Figure 1. A door guarded by a dragon which asks a is controversy as to what effects background sounds may have on performance. For example, a study conducted by Smith [11] has shown that background sounds may be However, if a player is not certain about the correct stimulating but they may also have negative effects on answer, s/he is allowed to ask the dragon for a performance. On the other hand, another experiment, “negotiation”. In this case the student is allowed to make a which involved five computer games [12] has shown guess for which s/he has to provide a justification. The among other things that sound level had little influence on amount of points that the student is going to receive in the performance scores and errors. Therefore, in VR- negotiation mode, depends on how close the student’s ENGAGE, which is primarily aiming at educating players answer is to the correct answer and/or how plausible the reasoning that s/he has used is. If the answer that the In terms of the game score, which represents the student gives is absolutely correct then the dragon allows student’s mark, the “negotiation’’ option may be better him/her to proceed through the door. However, if the than an immediate incorrect answer of the student. This is answer is not completely correct then the system performs so because the student may gain points for a plausible error diagnosis. The results of the diagnosis are guess (although they do not gain full marks) whereas if communicated to the student through the virtual s/he gives an immediate answer to the dragon, which is companion agent that appears to help the student. incorrect s/he does not receive any points at all. The idea In the negotiation mode, the student modeling behind this design decision is to encourage students to tell component performs error diagnosis based on a cognitive the truth about what they know and what they do not know theory of Human Plausible Reasoning [13]. At the end of and to practice their reasoning skills when they do not this interaction, possible errors of the student and/or evidence of the student’s lack of knowledge on a topic are recorded in the long term student model. For example, the 4. Student Modeling in VR-ENGAGE
student may have been asked the following question: “What is the capital town/city of the geographical The student modeling component of VR-ENGAGE compartment called Achaia (in Greece)?” While being in examines the correctness of the students’ answers in terms the negotiation mode, the student may give an answer of the students’ factual knowledge and reasoning that they such as: “My guess is that Rio is the capital of Achaia. I have used. Information about each student concerning know that Rio belongs to Achaia; Rio is an important his/her knowledge and reasoning ability, is recorded in town in Achaia. Therefore, it is likely that Rio is the his/her long term student model. The long term student capital of Achaia.” The student’s guess may be correct or model [18] keeps a history record of the student and is incorrect; in the case of the example, it is incorrect updated every time the student answers a question. The because Patras is the correct answer. However, the long term student model is used to adapt the presentation reasoning that s/he has used may reveal whether the of lessons to the particular student’s knowledge and student has a good knowledge of geography and whether The student modeling capabilities needed for the In this sense the game provides an environment where negotiation mode of the game, are based on Human there is opportunity for a negotiating teaching-learning Plausible Reasoning theory, henceforth referred to as dialogue between the ITS and the students. Collaborative HPR. This theory formalizes the plausible inferences discourse is an issue that has attracted a lot of research based on similarities, dissimilarities, generalisations and energy in the recent years (e.g. [14],[15]). The process of specialisations that people often use to make plausible becoming an expert in a certain domain should no longer guesses about matters that they know partially. Important be solely viewed as the acquisition of a representation of inference patterns in the theory are the statement correct knowledge; the knowledge to be acquired should transforms. These inferences may lead to either correct or incorrect guesses; in any case these guesses are plausible. If a player does not know the answer at all or has given HPR has been adapted and used previously in an incorrect answer without having asked for negotiation, intelligent environments for novice users of UNIX [19] then s/he does not receive any points and may only and for novice users of a Graphical User Interface [20]. continue his/her way if s/he asks for help. In such cases Moreover, it has been applied in an ITS authoring tool the virtual companion appears and lets the student know [21]. The previous adaptations of HPR in a variety of what the correct answer is, so that the door may be domains, which were very different from one another and opened. In addition the virtual companion suggests to the from the present one, shows that HPR could be very student to read a particular section of the lesson, which is promising as an underlying reasoning mechanism in mostly relevant to the question that s/he did not know how educational applications. Therefore, it has been adapted to answer correctly. The appropriate section is selected for the particular circumstances of an educational based on the error diagnosis performed by the student computer game that aims at teaching students both the domain of geography and the way to reason about facts The existence of the virtual companion in the game has been considered quite important for the promotion of the In the context of the game, HPR has been used to add student’s sense of collaboration similarly with a lot of human-like reasoning abilities to the animated agents that other recent educational systems. For example, Kay [17] interact with the students. In particular, when a student is notes that there is a growing acknowledgment of the asked a question from the domain of geography, HPR is importance of the learner’s social context therefore used to perform error diagnosis in case of an error and to systems are increasingly being designed for learners find out how close the erroneous answer has been to the working in groups of real or simulated peers. correct one. The outcome of the negotiation process is experiment. The class were divided into two groups of 8 recorded to the long term student model and is used to children, group A and group B. The division of children adapt the presentation of the teaching material to the into two groups was based on the human teachers’ individual student. Moreover, in case a student asks for selection of children in such a way that the two groups had negotiation when s/he is expected to give an answer to a the same distribution of students having good, average question in geography, the system employs the inference mechanism of HPR to evaluate the plausibility of the Group A was given VR-ENGAGE to work with for two student’s answer in terms of the model of human hours. Group B was given an ITS which consisted of the reasoning that HPR represents. In the negotiation mode, underlying reasoning mechanisms of VR-ENGAGE but the student is asked to give explicitly the reasoning for the had a simple user interface with no game. Both groups answer that s/he gives and is not certain about. were told by their human teachers that they had to For example, the question: “What is the capital complete a test using the software. In the environment of town/city of Achaia?” that was mentioned previously, VR-ENGAGE this meant that they had to open all doors corresponds to the statement: capital(Achaia)=Patras, in a virtual world and reach the exit. In the environment of where “capital” is a descriptor, “Achaia” is an argument the ITS they had to answer a set of questions, which were and “Patras” is a referent. Based on HPR, the erroneous displayed to them in plain text and context. The rules for answer that the student has given in the example: the students’ receiving their marks through the software capital(Achaia)=Rio corresponds to a similarity referent were the same for both groups. In both applications, the transform because the two towns belong to Achaia and mode of negotiation existed. However, in the ITS there they are similar in terms of the importance of their was no adventurous context associated with it. Finally, harbours. Therefore the student’s answer is considered both groups were supervised by two computer assistants close to the correct one and the student receives some who helped them with their interaction with the computer. marks for his/her answer (although not full marks of After the children had used the programs, the marks course). However if the student gives a totally irrelevant they had obtained and the errors they had made were answer then s/he does not receive any marks at all. collected in their user protocols since all their actions had been recorded. Then, the designers of VR-ENGAGE 5. Evaluation
On average, the students who had used VR-ENGAGE had spent more time with the system than the students who One important reason for the incorporation of an ITS had interacted with the ITS. This was partly due to the fact into a virtual reality game was the objective of making that there was more to explore in the game therefore educational software more engaging and motivating than students needed more time to complete the game. other forms of software while retaining and even However, most importantly, it was also partly to the fact improving the underlying reasoning mechanisms. At a first that the players of VR-ENGAGE had spent more time glance, this might look as an obvious achievement of the reading the lessons that were shown to them than the other game. However, there may be students who are not used group of students. This showed that VR-ENGAGE was to games and thus might not like them in classroom. On the other hand, the educational game may fall short of the After the interaction with either of the two systems, expectations of frequent game players since they may students were asked again to answer the same test have high demands for the game environment. Finally questions where they had originally made a mistake. The students may be distracted by the game and may not learn players of VR-ENGAGE remembered the correct answers from the educational content of the application. to a higher extent than the other group of students. This In view of the above, an evaluation of VR-ENGAGE showed that VR-ENGAGE had achieved its aim of being has been conducted so that the design assets and at least as effective as an ITS in the learning outcomes and deficiencies could be highlighted. The evaluation mainly was in fact better than the ITS in this respect. consisted of a comparison between VR-ENGAGE and an Finally, the interview showed that the players of VR- ITS with a conventional user interface but with the same ENGAGE were fascinated by the idea of a game in the underlying reasoning mechanisms as VR-ENGAGE. This classroom and they were certainly happier than the other kind of evaluation was conducted as an experiment, which group of students. However, most of them also involved school children and took place in classrooms spontaneously commented on the game elements before while human tutors were present but were not actively they were even asked about them. In general, they pointed out that the game would be better as a game if it had more In particular, the experiment involved a class of 16 virtual objects, more background sounds and more school children of 11-12 years old and two human adventure. This was due to the fact that most of them were teachers of this class that were present during the familiar with commercial virtual reality games therefore they compared VR-ENGAGE with them and had higher Garden: Girl’s Responses to Electronic Games”. Journal of Computers in Math and Science Teaching, 13(4), 1994, The fact that the students who had used VR-ENGAGE commented on aspects concerning the game itself showed [7] Self J. “The Defining Characteristics of Intelligent a potential of this game to be used by children at their Tutoring Systems Research: ITSs Care, Precisely”, leisure time. This would mean that VR-ENGAGE could International Journal of Artificial Intelligence in replace other computer games, which did not have any Education ,10 , 1999, pp. 350-364. educational value, in the children’s preferences for their [8] Griffiths, M.D. “Technological addictions”. Clinical entertainment. Indeed, this is an issue that is going to be Psychology Forum , 76, 1995, pp. 14-19. addressed in future versions of VR-ENGAGE. [10] Walker, J.H., Sproull, L. & Subramani, R. “Using a 6. Conclusions
Human Face in an Interface”. In B. Adelson, S. Dumais & J. Olson (eds.) Proceedings of ACM Conference on Human Factors in Computing Systems CHI , 1994 , pp. Educational applications may benefit from the technology of virtual reality games, which can increase the [11] Smith, A., Whitney, H., Thomas, M., Perry, K., students’ engagement and motivation. However, one Brockman, P. “Effect of caffeine and noise on mood, major problem of this kind of educational application is performance and cardiovascular functioning.” Human the construction of the game itself and the connection of Psychopharmacology-Clinical and Experimental, 12, pedagogy and student adaptivity with the story of the game. The approach taken in VR-ENGAGE that we [12] Wolfson, S. & Case, G. “The effects of sound and described in this paper offers a solution to this problem. colour on responses to a computer game”. Interacting VR-ENGAGE employs animated agents who take part in with computers, 13, 2000, pp. 183-192. the story of the game by asking questions, and by [13] Collins A. and Michalski R. “The Logic of providing adaptive advice and collaboration to the Plausible Reasoning: A core Theory”, Cognitive Science student. The tutoring adaptivity to the student’s needs is provided by a domain-independent reasoning mechanism [14] Moore D. “A framework for using multimedia within that performs error diagnosis and records the student’s argumentation systems”. Journal of Educational progress in the student model. The system has been Multimedia and Hypermedia 9(2), 2000, pp. 83-98. evaluated and the results of the evaluation show that it has [15] Baker, M. “A Model for Negotiation in Teaching- greater acceptability and effect on students than an Learning Dialogues”. International Journal of Artificial educational application with a conventional user interface. Intelligence in Education, 5(2), 1994, pp. 199-254. [16] Andriessen J. and Sanberg. “Where is Education 7. References
Heading and How about AI?”. International Journal of Artificial Intelligence in Education, 10, 1999, pp. 130- [1] Provenzo, E.F “Video Kids: Making sense of Nintendo”. MA: Harvard University Press, Cambridge, [17] Kay, J. “Learner Control”, User Modeling and User Adapted Interaction, 11, 2001, pp. 111-127. [2] Mc Grenere, J.L. “Design: Educational Electronic [18] Rich E. “Users are Individuals: Individualizing User Multi-Player Games. A Literature Review”. Technical Models”, International Journal of Man-Machine Studies, Report 96-12, the University of British Columbia, [19] Virvou M. and Du Boulay B. “Human Plausible [3] Papert, S. “The Children’s Machine: Rethinking Reasoning for Intelligent Help”. User Modeling and User School in the Age of the Computers”. Basic Books, New Adapted Interaction, 9, 1999, pp. 321-375. [20] Virvou M. and Kabassi K. “Evaluation of the [4] Boyle T. “Design for Multimedia Learning”. Prentice Advice Generator of an Intelligent Learning Environment”. Proceedings of the IEEE International [5] Amory, A., Naicker, K., Vincent, J. & Claudia, A. Conference on Advanced Learning Technologies (ICALT “Computer Games as a Learning Resource”. Proceedings of ED-MEDIA, ED-TELECOM 1998. World Conference [21] Virvou M. “A Domain-Independent Reasoning on Educational Multimedia and Educational Mechanism for an ITS Authoring Tool”. Proceedings of Telecommunications, Vol.1, 1998, pp. 50-55. ED-MEDIA 2000 World Conferences on Educational [6] Inkpen, K., Upitis, R., Klawe, M., Lawry, J., Multimedia, Hypermedia and Telecommunications, 2000, Anderson, A., Mutindi, N., Sedighian, K., Leroux, S. & Hsu, D. “We Have Never-Forgetful Flowers In Our

Source: http://www.gkatsion.kman.gr/2002/2002-ICALT-Kazan/ICALT_2002_VirvouManosKatsionis.pdf