An Educational Virtual Environment for the Teaching of Road Safety

ViSTREET: An Educational Virtual Environment for the Teaching of Road Safety Skills to School Students Kee Man Chuah, Chwen Jen Chen, and Chee Siong Teh Faculty of Cognitive Sciences and Human Development, Universiti Malaysia Sarawak, 94300 Kota Samarahan, Malaysia [email protected], {cjchen,csteh}@fcs.unimas.my

Abstract. Virtual reality (VR) has been prevalently used as a tool to help students learn and to simulate situations that are too hazardous to practice in real life. The present study aims to explore the capability of VR to achieve these two purposes and demonstrate a novel application of the result, using VR to help school students learn about road safety skills, which are impractical to be carried out in real-life situations. This paper describes the system design of the VR-based learning environment known as Virtual Simulated Traffics for Road Safety Education (ViSTREET) and its various features. An overview of the technical procedures for its development is also included. Ultimately, this paper highlights the potential use of VR in addressing the learning problem concerning road safety education programme in Malaysia. Keywords: virtual reality, educational virtual environments, road safety education, instructional technology.

1 Introduction The utilisation of computer simulations in enhancing teaching and learning has become popular in recent years largely due to technological advancements in threedimensional (3D) graphic processing and declining costs of computer peripherals. Virtual reality (VR) is a more recent technology that is used for computer simulations. VR enables users to interact with three-dimensional data, creating a potentially powerful interface to both static and dynamic information [1]. Studies conducted by various researchers [2, 3, 4, 5] further reveal that VR offers a large number of possibilities in instructions due to its capabilities, which are absent in other tools. The key capability is that VR helps learners to experience and visualise directly some physical properties of objects and events that are unavailable or unfeasible in the real world due to distance, time, cost, or safety reasons. In light of this, VR is regarded as a potential instructional tool to provide simulated training and skills teaching in dangerous or logistically impossible circumstances such as roads with heavy traffics, house on fire or coal mine. VR is thus increasingly eminent in prevention training as well as emergency or disaster management [6]. H. Badioze Zaman et al. (Eds.): IVIC 2009, LNCS 5857, pp. 392–403, 2009. © Springer-Verlag Berlin Heidelberg 2009

ViSTREET: An Educational Virtual Environment

393

One area of concern in which VR can provide a plausible solution is road safety education that is often confined to the use of verbal teaching and printed materials and impractical to be carried out on real roads [7]. The present work aims to explore the potential of VR in addressing these issues pertaining to the current implementation of road safety education in Malaysia. Based on the meticulous review of materials used for the teaching of road safety skills in the classroom and prior related studies, a VRbased learning environment known as Virtual Simulated Traffics for Road Safety Education (ViSTREET) is designed and developed. ViSTREET aims to complement the current road safety curriculum in Malaysian schools by providing authentic road safety practices to school students aged 12 to 14. In particular, ViSTREET teaches students on crucial pedestrian skills such as detecting dangerous situations, gap timing and safe place finding [8] by ‘placing’ them in life-like simulated traffic conditions within the learning environment. This paper explains the system design of the ViSTREET learning environment as well as the technical procedures for developing the learning environment.

2 Road Safety Education in Malaysia Malaysia is among the countries that has consistently recorded a large number of road traffic accidents in proportion to its population annually over the last three decades [9]. A closer look at the statistics on fatalities due to road accidents in Malaysia shows that pedestrians are among the top three high-risk groups, after motorcyclists and motorists [10]. In addition to that, studies conducted by Road Safety Department of Malaysia and Malaysian Institute of Road Safety Research (MIROS) reveal that a majority of pedestrian casualties involve children and young teenagers aged 9 to 14 [9]. Therefore, under the Malaysian Road Safety Plan (2006-2010), the Road Safety Department in collaboration with Ministry of Education has initiated a road safety education programme targeting school students. The programme is being introduced into primary and secondary schools in stages, starting with Year One in 2008. It will be fully implemented in all schools by 2011 and at all school levels [11]. The programme has its emphasis on teaching pedestrian safety skills to school students by using training materials like posters, multimedia, video and pamphlets. Teachers are also told to use roads within the school compound to provide students necessary practical training. Real-world practical training in pedestrian skills is known to be highly effective at improving the performance of children as young as 8 years of age. The ideal context for practical training would seem to be at the roadside and there is no doubt that roadside training can be highly effective [8]. When conducted at the roadside, however, this training can be time-consuming, labour intensive, dangerous and subject to disruption from poor weather and a lack of traffic situations of the types required. In addition, the teaching of pedestrian safety skills using printed materials such as pamphlets or posters (refer to Fig. 1 for an example) has its limitation in the sense that learners cannot visualise the traffic scenarios in a more concrete manner. On the other hand, the use of multimedia application in teaching road safety skills is mainly focused on factual drills without road simulations that allow active participation of the students. In general, these applications present traffic situations with the use of static

394

K.M. Chuah, C.J. Chen, and C.S. Teh

Fig. 1. Poster Showing Road Crossing Tips as provided by Road Safety Department of Malaysia

or animated graphics and ask students to memorise key safety skills by identifying the correct actions. While this can be useful to highlight salient safety facts, it does not train crucial road safety skills in an authentic manner. As such, VR-based learning environment offers a way round these difficulties. The VR capabilities of allowing three-dimensional visualisation of problem as well as guided exploration make it well-suited to be used as an instructional tool to address the aforementioned learning problems.

3 VR for Educational and Training Purposes A typical definition of the term VR is an image produced by a computer that surrounds that person looking at it and seems almost real. Here, the word reality refers to the external physical world and when it exists virtually, the reality suggests something can be explored by our senses, and yet does not physically exist [12]. There is a general acceptance that VR is about creating acceptable substitutes for real objects or environments, and is not really about constructing imaginary worlds that are indistinguishable from the real world [13]. VR-based environments have two principal variants: immersive and non-immersive VR. The present study focuses on non-immersive VR, which is also commonly known as desktop VR. Desktop VR displays 3D graphical virtual world on a standard computer screen and allows user responses via generic input devices such as keyboard and computer mouse. Desktop VR is chosen mainly due to the much lower cost that it incurs as compared to immersive VR. Despite its lower cost, desktop VR is equally powerful in creating life-like virtual environments for user to explore. According to Burdea and Coiffet [12], VR is capable of affording constructivist learning due to its ability to mediate world exploration and construction, its mapping of a user to any character he or she chooses and the provision of shared virtual worlds. Through its interactive environment, repetition and one-to-one experimentation, VR can help improve knowledge retention and this makes it appropriate for educational and training purposes. In the case of this study, a virtual environment can be used to simulate a real situation that is too dangerous, complex, or expensive to train in. There is potential for increasing safety standards, improving efficiency, and reducing overall training costs.


395

A particular advantage over traditional teaching and training technologies, such as books and video, is that the learner is active and can improve skills and understanding through practice. For example, Smith and Ericson [6] developed a VR-based learning environment to teach fire safety to school children. Students were able to identify home fire hazards and then practised escaping from a simulated fire in the virtual environment. Their studies also found out that students were more engaged by the learning environment without comprising skills acquisition. Kizil and Joy [14] on the other hand, developed a system to help prepare miners for dangerous situation that could not be addressed through traditional training methods. In terms of road safety education, the use of VR technology is slowly gaining momentum though still very limited. One rare example would be the learning system developed by Tolmie et al. [8]. The system provides simulation training by using a game-like scenario. Learners are asked to navigate through several town settings via an avatar or character. The activities in the virtual environment were mainly about making decisions on when or how to cross the street. When a decision was made, the computer demonstrated the consequences of that decision. However, this system lacks depth in terms of the environment details as it uses mainly plain comic-like graphics and it does not offer firstperson point of view, which could be vital in influencing learner’s judgment. Hence, it is the intention of this study to address these by creating a more life-like VR learning environment for the teaching of road safety skills.

4 ViSTREET System Design As mentioned, ViSTREET is a desktop VR-based learning environment for teaching school children pedestrian road safety skills. Based on the input gathered from pertinent literature, the VR-based learning environment focuses on three broad and related areas of pedestrian skills as suggested by Tolmie et al. [8]: i. Safe-place finding – perception of the dangers posed by aspects of road layouts. ii. Roadside search – awareness of potential and actually vehicle movements and the implication for road crossing. iii. Gap-timing – co-ordinating road crossing with vehicle movements Opinions from researchers at MIROS on these skills were also obtained as to check on their relevance towards the road safety curriculum used in Malaysian schools. In general, each skill is addressed by a distinct module consist of VR-based scenarios, which share the same town settings. 4.1 Instructional Design In ensuring that ViSTREET is able to achieve the intended goal of delivering road safety instructions in a more authentic manner, special attention were given to the selection of the guiding instructional design model. A more recent instructional design model or theoretical framework that specifically addresses instructional issues in VR-based learning environment, known as VRID [15] was chosen. This theoretical framework provided guidance on the methods for facilitating learning as well as provided assistance in deciding the design of ViSTREET. The framework comprises

396


Fig. 2. Theoretical framework for designing a desktop VR-based learning environment [15]

macro-strategy and micro-strategy (see Fig. 2). The macro-strategy combines the concept of integrative goals proposed by Gagné and Merrill [16] and the model of designing constructivist learning environment proposed by Jonassen [17] The microstrategy, on the other hand, is based on the cognitive theory of multimedia derived by Mayer [18] which is used to guide the design of the instructional message. Macro Strategy. Goals that are to be achieved from learning are presumed to be the starting point of the instructional design process. Thus, the VRID model starts with identifying the instructional goal which is a combination of several individual objectives that are to be integrated into a comprehensive purposeful activity known as enterprise. This is the concept of integrative goal proposed by Gagné & Merill (1990). These individual objectives may fall in the category of verbal information, labels, intellectual skills, or cognitive strategies. In the case of the ViSTREET learning environment, based on the three broad areas of pedestrian skills identified, the types of learning and the corresponding learning objectives (refer to Table 1) were formulated. In addition, the integrative goals for this learning environment are identified as the learner’s capabilities to interpret the basic skills of safe walking on various traffics scenarios (for example, crossing a busy street and walking to the park safely). The instructional designer then needs to design instruction that enables the learners to acquire the capability of achieving this integrated outcome, which is called the enterprise scenario. As mentioned by Chen et al. [15], the enterprise scenario is similar to the problem posed in a constructivist learning environment. The VRID model stresses on the importance of posing an appropriate problem which includes three integrated components: the problem context, problem presentation, and problem manipulation. The incorporation of these components in the ViSTREET learning environment is depicted in Table 2.


397

Table 1. Types of Learning and Corresponding Learning Objectives for ViSTREET Learning Environment Types of Learning Labels

Verbal Information Intellectual Skills

Cognitive Strategies

• • • • • • • • • • • •

Learning Objectives State the name of various traffic signs State the name of various line markings State the various types of roads (two-way, double lane and single lane). Describe the meaning of various traffic signs Describe the meaning of line markings (e.g. zebra crossing) Differentiate the use of various traffic signs Differentiate the use of various line markings Identify the basic rules when crossing the road Identify the basic rules when walking on busy streets Identify the safe zone for road crossing Reflect on the consequences of not following the right pedestrian road safety skills. Reflect on the actions taken when exploring the virtual road scenario

Table 2. The Description of Enterprise Scenario in the Learning Environment Enterprise Scenario Component Problem Context

Descriptions •

Problem Representation

•

Problem Manipulation Space

•

Constructivist learning environment must describe in the problem statement all of the contextual factors that surround a problem to enable the learners to understand the problem. In ViSTREET, the learning goal is presented to the learners when they begin exploring the environment. Constructivist learning environment must provide an interesting, appealing and engaging problem representation that is able to perturb the learner. In ViSTREET, a story-based problem is presented to the learner together with the virtual environments. This helps the learner to build a mental representation of the problem. Based on the given story line, the learner needs to complete several tasks, fulfilling each objective. Constructivist learning environment must allow active manipulation space for a problem. Learners must manipulate something and obtain feedback as how their manipulations affect the environment. In ViSTREET, the virtual traffic scenarios serve as the problem manipulation space that allows the leaner to navigate through the world in a “walking mode” using input devices such as a mouse or keyboard. The “waking mode” in the virtual environment is similar to the movement in the real world, allowing a better representation of the problem.

398


In addition, constructivists believe that individual’s knowledge is a function of one prior’s experiences, mental constructs, and beliefs that are used to interpret events and objects. Therefore, the instructional design has to provide various supports that may assist the learners to construct their knowledge and engage in meaningful learning in the learning environment. These support tools include related cases, information resources, cognitive tools, conversation and collaboration tools, and social or contextual support. Related cases refer to a set of related experiences or knowledge that the learner can refer to. This is mainly because skills or knowledge learned in a particular context are easily repeated by learners as long as they are in a similar context. In the ViSTREET learning environment, the various traffic simulations provide real-life representations that the learners could easily relate to. For instance, the learning environment is based upon a three-dimensional simulation of a small town, complete with buildings, roads, traffic signs, and vehicle movements, which allow the learners to explore the various consequences of their action such as crossing the road. This allows them to learn the safety skills better as opposed to learning it through text or static image presentation. Information resources refer to the rich sources of information that help learners to construct their mental models and comprehend the problems. In the ViSTREET learning environment, hyperlinks to related resources (for example, traffic signs explanation, road safety tips, and description of traffic scenarios) are given. These resources are easily accessible via the menu provided together with the learning environment. Thus, the learner is free to access these resources while trying to solve the given problem. Cognitive tools are tools that can scaffold the learners’ ability to perform the task. Conversation and collaboration tools allow learners to communicate, collaborate and share ideas. Social or contextual support stresses on the importance of considering contextual factors, such as physical, organizational, and cultural aspects of the environment. In the ViSTREET learning environment, cognitive tools are provided. The virtual simulated traffics act as a visualisation tool where learners can visualise a dynamic three-dimensional representation of the problem. Apart from that, the learning environment also provides screenshots of physically impossible viewpoints of the traffic scenarios. These include a plan view map that offer bird’s eye view of various parts of the scenarios and a tracer icon that shows the position of the learner on the plan view map in real-time. Another important component stressed by the VRID model is instructional activities. There are three types of instructional activities: modelling, coaching and scaffolding. Modelling. In the ViSTREET learning environment, both behavioural modelling and cognitive modelling are incorporated. Behavioural modelling is provided by virtual characters in the learning environment that practises good pedestrian safety skills. Cognitive modelling in the learning environment, on the other hand, promotes reflection of such behaviour in the form of text-based narrative. Coaching. In terms of coaching, the learning environment provides it using feedback messages. These messages appear as the learner is exploring the virtual scenarios, and can guide the learner as he or she is learning the intended skill. These feedback messages are presented in a more appealing manner in order to highlight them to the leaner.


399

Scaffolding. As for scaffolding, the learning environment splits the learning problem into four sub-problems. The four sub-problems are linked to provide scaffold for the learner’s performance in approaching the learning problem. Such scaffolding is known as strategic scaffolding. In addition, the learning environment also provides learners with a help link that scaffolds the learner’s abilities to perform navigational task in the virtual environment, which could impede their learning. This type of scaffolding is known as procedural scaffolding [15]. Another type of scaffolding is known as conceptual scaffolding, in which in the learning environment various hints that guide the learner to available resources are given to assist them in solving the given task. Micro Strategy. To complement the macro-strategy, Mayer’s [18] principles of multimedia design is used as the micro-strategy to guide the design of instructional message in the learning environment for a more effective learning. The five principles of multimedia adopted are multimedia principle, spatial contiguity principle, coherence principle, modality principle and redundancy principle. In line with the multimedia principle, the content in the learning environment is presented using both words and pictures. Images or pictures are labelled accordingly rather than solely depending on texts. In terms of spatial contiguity principle, the texts used to describe the learning problem in the learning environment are presented near to a given snapshot (e.g. picture of the destination they are required to go). This is one example of enhancing spatial contiguity. As for coherence principle, guiding content in the learning environment is made clear and simple without overloaded texts, sounds or pictures. The remaining principles are applied when deemed necessary as the ViSTREET learning environment only uses ample amount of narration. 4.2 Development of Learning Environment The development of ViSTREET is based on the integration of Virtual Reality Modelling Language (VRML) and Hypertext Markup Language (HTML). These are selected due to its feasibility to be widely used without the need to expensive software. In ViSTREET, each skill to be taught is presented in one specific scenario. The combination of virtual road scenarios formed the complete learning environment. The learning environment was then integrated onto the web interface to allow better visualisation and flexibility in adding other useful web-based components. Fig. 3 shows the steps involved in developing one single virtual road scenarios. These steps are occasionally repeated to form the full learning environment. Select a Skill Focus. As mentioned in the earlier chapter, the VR-based learning environment for the present study was developed based on the suggestions given by the subject matter expert. Prior to developing the scenarios, a skill focus was selected. This is to help form the focus on the types of scenarios to be identified. Identify a Scenario. After the skill focus was selected, a specific scenario was identified with the assistance of the subject matter expert. The scenario was then divided into several sub-scenarios to provide proper scaffolding for the learner. Example of this scenario is walking to the school for co-curricular activity when learner’s father was unavailable to fetch him or her there.

400


Fig. 3. Steps for developing a virtual road scenario for the learning environment

Create a Storyboard. Based on the identified scenario, a storyboard was created. The storyboard include sketches of two-dimensional plan on grid paper, scripts for instructional events as well as sketches of relevant objects like characters, buildings and plans. Assemble a Scene. Following the storyboard, the virtual road scenario was assembled using the ParallelGraphics’ Internet Scene Assembler (ISA) version 2.0, a threedimensional VRML authoring tool with the help of Internet Space Builder software (a VRML object builder) To begin, a new scene in ISA was created. The newly created scene was blank and contains no background or any landscape on it. In order to set the environment, a pre-build scene developed using ISA, was imported and used as the core environment. The scene contains the basic landscape of a town setting with roads. The background of the scene was then changed to a more realistic sky look inserting the image via the Background tool object. Using the pre-build environment, related objects such as buildings, roads and plants were added to the scene for the specific scenario. ISA allows great integration with ISB objects. As such, all objects were created in ISB and saved in the ISB format before being added or imported to ISA. A two-dimensional plan view available in ISA also permits the moving and repositioning of objects. Fig. 4 shows some of the features available in ISA that are used to create and add virtual objects. Add Interactivity. Each object in ISA contains fields that hold the values of its parameters. These parameters can be changed to create animation and interaction. The way to change a field is to send an event to that field by means of a mechanism through which this event can be programmed to cause changes in other object. In ISA, this is known as route, which is the connection between an object generating an event and an object receiving the event. ISA allows object animation by changing the position, orientation and size of any object in the scene. Other properties such as colour, transparency, intensity can also be animated. To add further interactivity, several Javascript functions (using the JSFunction object in ISA) such as pop-up alert, sound control and fading effects are added


401

Fig. 4. Features available in creating and adding objects within ISA

Fig. 5. Screenshot of the learning environment showing examples of cognitive tools and problem representation

402


Fig. 6. Screenshot of the learning environment showing examples of problem representation, problem manipulation space and information resources

Publish a Scene. After all editing was done, the scene was published using the ISA publishing tool. All resources files of a scene were placed in a sub-folder specified in the publishing options. The completed scene was then embedded onto a HTML page together with its interface design. Fig. 5 and Fig. 6 show screenshots of the completed ViSTREET learning environment.

5 Conclusions This paper has presented the use of VR as a tool to enhance the teaching of road safety skills to school students in Malaysia. The design and development of the ViSTREET learning environment are also described. Demonstrating correct behaviour in a classroom setting does not necessarily translate to real-world situations. VR applications such as ViSTREET show potential for providing the means for teaching and training of road safety skills in a safe and authentic environment. They allow students to experience a more realistic virtual ‘‘on-site’’ experiences for high-risk safety training, which cannot be achieved through lectures or regular video presentation. Thus, the present study demonstrates a significant step towards improving road safety education for school students.

References 1. Louka, M.N., Balducelli, C.: Virtual Reality Tools for Emergency Operation Support and Training. In: Proceedings of TIEMS (The International Emergency Management Society), Oslo (2001) 2. Dickey, M.D.: Teaching in 3D: Pedagogical Affordances and Constraints of 3D Virtual Worlds for Synchronous Distance Education. Distance Education 24, 105–121 (2007)


403

3. Hamada, M.: An Example of Virtual Environment and Web-based Application in Learning. International Journal of Virtual Reality 7(3), 1–8 (2008) 4. Pantelidis, V.: Designing virtual environments for instructions: Concept and Considerations. VR in the Schools 2(4), 6–10 (1997) 5. Roussou, M.: Learning by doing and learning through play: an exploration of interactivity in virtual environments for children. ACM Computers in Entertainment 2(1), 1–23 (2004) 6. Smith, S., Ericson, E.: Using immersive game-based virtual reality to teach fire-safety skills to children. Virtual Reality 13, 87–99 (2009) 7. Ampofo-Boateng, K., Thomson, J.A.: Child pedestrian accidents: A case for preventative medicine. Health Education Research: Theory and Practice 5, 265–274 (1990) 8. Tolmie, A., Thomson, J., Foot, H., McLaren, B.: Child development and the aims of road safety education. Department of Transport, Road Safety Report (1) (1996) 9. Mustafa, N.M.: Overview of Current Road Safety Situation in Malaysia (2006), http://www.unescap.org/ttdw/roadsafety/Reports2006/ Malaysia_RSpaper.pdf 10. Moe, H.: Road Traffic Injuries among Patients Who Attended The Accident and Emergency Unit of the University of Malaya Medical Centre, Kuala Lumpur. JUMMEC 11(1), 22–26 (2008) 11. Road Safety Education portal, http://www.pkjr.org.my 12. Burdea, G.C., Coiffet, P.: Virtual reality technology, 2nd edn. Wiley & Sons, New Jersey (2003) 13. Vince, J.: Essential virtual reality fast: How to understand the techniques and potential of virtual reality. Springer, New York (1998) 14. Kizil, M.S., Joy, J.: What can virtual reality do for safety? St University of Queensland, Lucia (2001) 15. Chen, C.J., Toh, S.C., Wan, M.F.: The theoretical framework for designing desktop virtual reality based learning environments. Journal of Interactive Learning Research 15(2), 147– 167 (2004) 16. Gagne, R.M., Briggs, L.J., Wager, W.W.: Principles of instructional design, 4th edn. Harcourt Brace Jovanovich Publishers, Texas (1992) 17. Jonassen, D.H.: Designing constructivist learning environments. In: Reigeluth, C.M. (ed.) Instructional-design theories and models: A new paradigm of instructional theory, 2nd edn., pp. 215–239. Lawrence Erlbaum Associates, New Jersey (1999) 18. Mayer, R.E.: Multimedia learning. Cambridge University Press, Cambridge (2002)