A model to design interactive learning environments for children with ...

Educ Inf Technol (2007) 12:149–163 DOI 10.1007/s10639-007-9039-3

A model to design interactive learning environments for children with visual disabilities Jaime Sánchez

Published online: 23 June 2007 # Springer Science + Business Media, LLC 2007

Abstract Current interactive learning environments cannot be accessed by learners with disabilities, in particular for students with vision disabilities. Modeling techniques are necessary to map real-world experiences to virtual worlds by using 3D auditory representations of objects for blind people. In this paper, a model to design multimedia software for blind learners is presented. The model was validated with existing educational software for these users. We describe the modeling of the real world including cognitive usability testing tasks by considering not only the representation of the real world but also modeling the learner’s knowledge of the virtual world. The software architecture is also described by using this model, displaying the components needed to impact learning. Finally, we analyze critical issues in designing software for learners with visual disabilities and propose some recommendations and guidelines. Keywords Modeling methodologies . Children with visual disabilities . Virtual environments . User adapted interfaces

1 Introduction A great amount of educational software has been developed for supporting learners with disabilities. Software for blind people aims to increase their access to current computing materials based on graphic user interfaces such as games, educational software, and web navigation systems. Clearly, the task of developing software for people with visual disabilities has some complexities. For blind learners, we face the problem of constructing interfaces that do not rely on graphics. However, we can find some similarities in the process of designing and constructing computer software for people with different types of disabilities. This is the case when developing cognitive systems aimed at modeling and implementing the real world by a computer system. J. Sánchez (*) Department of Computer Science, University of Chile, Blanco Encalada 2120, Santiago, Chile e-mail: [email protected]

150

Educ Inf Technol (2007) 12:149–163

Problems in constructing learning software for people with disabilities arise when the model is presented to the user for interaction purposes. Virtual learning environments for sighted learners transmit this model to the learners by using graphics (with or without animation), sounds, and text, taking advantage of a wide spectrum of the computer’s multimedia capabilities. For people with visual disabilities this spectrum is limited. This forces the software designers to project the information kept in the model, provided by the student, through the available auditory channel. Additionally, nontraditional interaction modes such as haptic devices can be used. The same considerations are valid for the construction of the learner’s model. Several virtual reality systems and virtual environments combined with appropriate human–computer interfaces have been used to enhance sensual capabilities of people with sensory disabilities. This is the case of presenting graphic information by text-to-speech and 3D auditory navigation environments to construct spatial mental representations and to assist users in acquiring and developing cognitive skills (Savidis et al. 1996). A sonic concentration game described by Roth et al. (2000) consists of matching different pair levels of basic and derived geometric shapes. To represent geometric shapes, it is necessary to build a two-dimensional sound space. The concept allows the shape to be rendered by the perception of moving sound in a specific plane. Each dimension corresponds to a musical instrument, and raster points correspond to pairs of frequencies on a scale. AudioDoom (Lumbreras and Sánchez 1999; Sánchez and Lumbreras 1999; Sánchez 2001) allows blind children to explore and interact with virtual worlds by using spatial sound. The game was based on the traditional Doom game where the player moves through corridors discovering the environment and solving problems simulated with objects and entities that inhabit a virtual world. VirtualAurea (Sánchez 2001) was developed after interactive sound-based virtual environments proved to trigger the development of cognitive spatial structures in blind children. VirtualAurea is a set of spatial sound editors that can be used by parents and teachers to design an ample variety of spatial maps such as the inner structure of the school and rooms, corridors, and other structures of a house. The development of mobile applications for users with visual disabilities has been studied by some researchers. The research project called Blind Interactive Guide System (Na 2006) proposes an interactive guiding system based on radio frequency identification technology to provide information needed for navigating inside rooms to the user with visual disabilities. In another project, Hedgpeth et al. (2005), authors aim to enrich the lives of people who are blind by developing visually perceptive wearable computers. A PDAsized computer receives a video input stream from a wearable camera (which is mounted in a pair of glasses worn by the user) and communicates what it “sees” through tiny sound emitters in the earpieces of those glasses. This wearable device can also communicate with “smart objects” that populate ubiquitous environments. This paper proposes a model for developing virtual learning environments for learners with visual disabilities. The model includes various steps and recommendations by considering key issues for conceptualization and implementation. Special attention is paid to the feedback issue considered to be a critical point in existing software. As a result of using this model in software design and development processes, software products emerge that contain certain properties that validate a correct impact on visual-impaired student learning. These properties are classified and outlined in the architecture that is also presented. This paper is an extended version of an article presented at the IFIP 19th World Computer Congress, TC-3 Education stream (Sánchez 2006).

Educ Inf Technol (2007) 12:149–163

151

2 Model 2.1 Development of educational software We propose a model for creating virtual learning environments for people with visual disabilities. The modeling process starts with the definition of desired cognitive skills. Then, we create a virtual environment that includes a navigable world by using adequate modeling languages, dynamic scene objects, and acting characters. Scenic objects are characterized by graphic and acoustic attributes; character’s actions are based on deterministic and nondeterministic plans in the same way as in interactive hyperstories described by Lumbreras and Sánchez (1999). The learner explores the virtual world by interacting with appropriate interfaces and obtains interactive feedback. The learner’s actions, such as sound reproductions, are collected, evaluated, and classified based on student modeling and diagnostic subsystems. The modeling process follows the steps illustrated in Fig. 1 (Sánchez et al. 2004). We define cognitive skills in real-world situations, for example self-motivating activities, drill and practice applications, problem solving, and entertainment. Objects in fictitious world scenarios are constructed of geometric primitives. They are characterized by acoustic attributes and grouped into components with input and output slots. Control elements of the virtual world are represented by acoustic elements, known as icons and earcons. We develop an internal computer representation and define a geometric environment containing a 2D or 3D visual and acoustic model without considering the limitations of potential users. Modern object-oriented modeling languages are powerful tools for building virtual worlds using scene graphs, interaction, and animation facilities. We insist on the necessity of special editors for teachers and learners to create synthetic models

Fig. 1 A model for designing virtual learning environments for children with visual disabilities

152

Educ Inf Technol (2007) 12:149–163

independently. According to the problem and target users, problem-specific correspondences between graphic and acoustic attributes or properties can be used to reduce the model to its acoustic projection. The resulting model meets the requirements in terms of certain information channels that should not be used in such a way that cannot be explored by people with visual disabilities. Another way is to generate the model directly by using a special editor for learners with visual disabilities. The acoustic representation of the model for learners with visual disabilities uses spatial sound. Interaction and navigation are based on acoustic control elements. The learner explores the object space by navigating, interacting with suitable interfaces, interpreting, and reproducing the structures. This can be done through navigating without changing viewpoints or by using internal representations of users by giving them the illusion of being part of the virtual scene. A blind learner explores neighboring models by grasping them, tracking objects or listening to typical sounds. Sound-emitting objects help them to build a mental model of real or fictitious worlds. Additionally, the learner may build an external reconstruction of the mental model or try to rebuild the acoustic model as it was perceived and imagined after exploring the model space. We must be certain that conditions during the reconstruction process are always the same. Therefore, the interfaces involved are calibrated accordingly. Depending on the particular parameters and entities of the model, error measures between internal representations and the model reproduced by the learner are defined. The learner’s actions are collected, evaluated, and classified. The outcome is transformed to a user-adapted aural output. The use of this model in the development of virtual learning environments for users with visual disabilities ends up with software containing representations and functionalities that meet the visually impaired user requirements. To achieve this, we have defined an architecture to identify which elements of virtual learning environments are essential for a correct impact on the user’s learning. To represent the real world, we defined the virtual environment and scenes that users navigate. We established game rules and how different metaphors (scripts) connect to each other. The world was designed as a collection of scenes grouped in different zones with common characteristics (cities, forests, etc.). The description of the user cognitive impact was made by defining the minimum tasks set to be accomplished by the student, which resulted in software requirements (navigate some scenes, use orientation resources, solve certain problems, etc.). According to these requirements, different content was shown to the user, using a quest or mission-to-fulfill metaphor, exploring and interacting with the virtual world. The development process consisted of the computer implementation of the world and the user’s models. We used an incremental model in such a way that when the first cycle ended, the user could navigate some zones created in a coherent and simple manner. Thus, in each cycle, new zones, characters, and functionalities were added. The projections for different models were designed and new information inputs and outputs flowed from and to the user, mainly using 3D sounds. A usability evaluation was implemented to obtain data about the users’ acceptance of the design model embedded in a prototype and whether or not it represented their mental model. We also evaluated the accomplishment of the proposed cognitive tasks. Three instruments were considered for usability evaluation: end-users’ evaluation, heuristic evaluation, and facilitators’ evaluation. Below, we discuss the main results obtained in the software usability evaluation. The data obtained were analyzed and studied to figure out how the metaphors, models, and projections could be improved (Fig. 2).

Educ Inf Technol (2007) 12:149–163

153

Fig. 2 Architecture of the resulting software

2.2 Modeling the resulting architecture Metaphor of the real world (model) According to the cognitive skills to be developed, real world metaphors are designed, as well as the activities that learners have to attain the cognitive goals by considering their interest and motivation. Editor The editor contains tools to construct an internal model based on 2D/3D graphic representations or auditory representations. For learners with visual disabilities, models can be generated by drag and drop actions on icons and images from a gallery. The equivalent representation of entities is in a text form or phonetic transcription. Children with visual disabilities choose the internal world objects through sensitive tablets or haptic devices. Computer representation of the real system The computer representation of the real system corresponds to the computer representation of the problem or the real world metaphor; it is knowledge modeling. At this point are functions, parameters, and variables of the state of the system describing the situation of the represented world and how the transition from one state to another will be made by considering the interaction of the learner with the software and reflected on entry variables. Strategy This component provides the strategies used to model the state of the learner’s knowledge. They are taken from the field of artificial intelligence applied to intelligent tutoring. One of them is the overlay model (Kass 1989), which treats the learner’s knowledge as a subset of an expert knowledge. The differential model (Clancey 1987) extends the previous model by dividing the learner’s knowledge into two categories: knowledge that the learner should know and knowledge that is not expected to be known

154

Educ Inf Technol (2007) 12:149–163

by learners. The perturbation model (Kass 1989) supposes that learners should possess potentially different knowledge in quantity with respect to an expert. The model can represent the knowledge and beliefs of the learner beyond the ranks of the expert’s model. Learner model This component represents what the system thinks about the state of the student’s learning at a certain point. It contains knowledge and skill representations the learner should construct, the variables of the state of the learner representing the level of learning at a certain moment, and the rules about how to upgrade this information given the interaction with the system and reflected in the change from one real world model state to another. Thus, the learner model is constructed by making inferences from individual knowledge and by analyzing the performance (Dillenbourg and Self 1992). Evaluation This component defines the difference between the knowledge model represented in the software and the knowledge model of the learner generated by the strategies. Thus, an error measure is produced and projected to the interface as student feedback. System projection This is the main component to certify that the software can be fully assimilated by children with visual disabilities. It is in charge of projecting most interactions, state variables, and feedback from and to the software. Below, we present the attributes to adequately develop a learning virtual environment for children with visual disabilities. During our research study, we applied this model to evaluate current software for blind learners. To accomplish this, we grouped the main components of this model into four heuristics: metaphor, learning, interaction, and interfaces. When applying these heuristics, we found strengths and potential errors in software interaction, especially in meeting the uniqueness of this type of application for blind learners. One example of this evaluation can be found in Sánchez and Baloian (2005). Some software previously designed in our research was modeled with this architecture (Sánchez and Sáenz 2005). For example, the software called Theo and Seth (see Fig. 3a) is an application to enhance mathematics skills in blind users comprising the following components: model, strategy, computer representation, model of learner, evaluation, learner, and interface. In model, according to the cognitive skills to be developed, diverse content is presented to the learner by using a metaphor consisting of a farm containing different elements to help children to solve arithmetic problems. In strategy, Theo and Seth has embedded a strategy to solve the problems posed that are randomly presented but deterministic. Computer representation is developed with a functioning methodology and global variables, allowing coherence to be maintained between results and exercises presented to the user. Model of learner is the part of the system in charge of saving the answers of the user and provides the grading results of the problems solved by the learner. Evaluation evaluates the student’s learning by analyzing the final results obtained by him or her and comparing them with the correct results. Learner obtains feedback from the system. Interface is the main component and receives and interprets the parameters entered by the user through the keyboard by certifying whether the system can be understood by a blind user. It is in charge of generating the projection of all interactions, state variables, and feedback from and to the system. The output is represented through audio and images. Other software called AudioChile and AudioVida (see Fig. 3b, c) are virtual environments that can be navigated through 3D sound to enhance spatiality and immersion

Educ Inf Technol (2007) 12:149–163

155

Fig. 3 Screenshots of software developed based on the proposal, a Theo and Seth, b AudioChile, and c AudioVida

throughout the environment (Sánchez and Sáenz 2006). This software was developed based on the model described above. Each software has a model, editors, knowledge representation model, strategy, evaluation, and projection. The model (metaphor) represents labyrinths. A person has to navigate through these labyrinths searching for cues and information to solve a problem posed. Editors consist of the virtual world, including the definition of its properties, objects, and personages. The knowledge representation model is based on the mapping of real world behaviors in a virtual world, state variables (physics, kinetics, and luminous), and actions. The strategy is in charge of following up the actions performed by the user. AudioChile provides a compass that perceives the user’s orientation within the virtual world. AudioChile provides different types of food for interaction. The child must interact with the food and decide whether or not to eat it. The learner model is such that, as the user goes through the adventure, he/she perceives and processes more information to solve the problem to reach the end of the story (AudioChile) or to get the maximum score (AudioVida). The evaluation in AudioChile consists of an evaluation of the compass to provide information to the user to find diverse objects and personages within the virtual world. AudioVida evaluates decisions of learners during interaction with the game. Learners get high scores when taking good decisions and low scores with erroneous decisions. Projection consists of 3D images and sound virtual worlds. It explores diverse fine representation qualities to get a high standard of use as well as to solve the problem posed.

156

Educ Inf Technol (2007) 12:149–163

2.3 Workflow AudioDoom was used as an exemplary application for blind children. Thus, we validated our model by describing in more detail the modeling workflow. The whole testing is described by Baloian et al. (2002). AudioDoom allows blind children to explore and interact with virtual worlds through spatial sound (Lumbreras and Sánchez 1998, 1999). AudioDoom has been usability tested with more than 40 Chilean children aged 7–12 years in school settings for blind children (Sánchez 2001). We followed the modeling steps introduced above. First, a computer-based representation is derived from the real world scenario by means of abstraction and reduction without considering the limitations of potential users. Then, the computer-based model is reproduced (projected) as an appropriate acoustical internal model explored by people with sensory disabilities through the use of available communication channels. Appropriate editors support the modeling process. Important model entities and parameters must be identified at this stage. By interacting with the model, the learner makes an internal or external reconstruction, which is later evaluated (see Fig. 4a, b). This can be done by using appropriate error measures depending on the learner, the computer-based (internal) representation, and the reconstructed model. Finally, the degree of similarity is derived from the error measure and the result is displayed on some learner-adapted output and used for modeling the learner’s knowledge. The modeling workflow in AudioDoom starts with the computer representation of the virtual game world (a simple labyrinth with one main corridor and two secondary corridors including entities and objects), which is projected to an internal model consisting only of sounds. At this stage, the role of volume, frequency, melody, and rhythm in representing different forms, volumes, and distances is analyzed. Learners interact with this model by “virtually walking” through the labyrinth with keyboard, mouse, and ultrasonic joystick. Sound-emitting objects help them to build a mental model of the labyrinth. Finally, they make concrete mental models with Lego blocks and try to rebuild the internal model as it was perceived and imagined after exploring the spatial structure. Different types of blocks represent objects of the virtual world in the computer representation. The concrete reconstruction is checked by a human tutor against the game world or by a camera and image processing algorithms to look for any spatial correspondence with the computer representation of the original world model by evaluating the error measure. As we can see, despite object representation, interfaces, perception modes, and error measures, there are important tasks that should be undertaken when developing systems for blind

Fig. 4 a Virtual game world embedded AudioDoom, b reconstructed external model by blind learners

Educ Inf Technol (2007) 12:149–163

157

people. A critical task in this modeling process is the reduction of the original model to only acoustic output. 2.4 Testing We can describe in more detail two steps of the modeling process. To process an external model and to evaluate error measures, we assume that a blind child reconstructs AudioDoom’s maze structure by using Lego blocks. Each block is individually marked by black bars and dots. A blind user perceives different blocks, selects the appropriate ones, and rebuilds the mental image by using the blocks one by one. After each step, a picture is taken with a digital camera placed on a fixed position over the scene. We highlight a typical state of the reconstruction process and indicate the next step by adding a new Lego block on the lower wall. Figure 5 shows the situation before and after the next step. Reducing the colors to black and white makes it possible to apply lowlevel image processing routines to detect the new Lego block. After a calibration of the two pictures, we can localize the new block through an exclusive-or operation on both images resulting in one or two dashes, or in new dots otherwise. Starting from these new picture elements, we calculate the position and type of the new Lego block. Finally, the learner’s model representation is updated. Thus, by certain low-level operations on succeeding pictures the external model can be transformed into an internal representation, which is used to feed an error measure function. A degree of fidelity is derived and displayed by text-to-speech. For a more sophisticated approach using Lego Robotic Command Explorer robots, see Ressler and Antonishek (2001). In AudioDoom, the differences concerning parameters such as volume, articulation, stress, and intonation must be visualized. From these visual patterns, important parameters

Fig. 5 Reconstructed Lego model

158

Educ Inf Technol (2007) 12:149–163

are detected by means of picture processing and in connection to the stored reference patterns. An intelligent component of the system interprets these results, displays a visualization of the error function, and transforms them into instructions for the learner. Whereas the wave amplitude image provides, in the xy-plane, any feedback concerning volume and rate, omitted syllables are characterized by a lack of colored areas, intonation, and articulation by the hue and saturation parameters, as well as the shape of the curves. AudioDoom is manipulated by using a wireless ultrasonic joystick, called The Owl. By using this device, the child can interact and move through the environment by clicking in different voxels of the surrounding space. According to the sound position, the child must coordinate the haptic/kinesthetic device with the perceived sound position. This scheme of action–reaction is strongly stimulated in the child because of the strong haptic–acoustic correlation embedded in the system. To deal with this issue, we designed AudioDoom to be mainly used with an ultrasonic joystick with three degrees of freedom and spatialized sound (see Fig. 6). For designing and developing information-equivalent interfaces for children with visual disabilities, we followed a similar model workflow as described above in two new systems. AudioBattleShip (Sánchez et al. 2003) is an interactive version of the board game Battleship, providing different interfaces for both sighted and blind people to enhance collaboration and cognition (see Fig. 7). Ebbinghaus’s forgetting function experiments were implemented in a project concerning the historical replication of key experiments in psychology (Biella et al. 2003). AudioMemory (Sánchez and Flores 2004) is software based on the classic memory game board where the user interacts through keyboard, joystick, and tablets. A few keyboard keystrokes are used. These devices have been used in different applications previously developed by the research group after usability testing them with children with visual disabilities. Microsoft SideWinder, a force feedback joystick, was used. This tool allows grading the user position in the grid and provides direct feedback with different forces. Counter forces to the movement are generated per token position change and vibratory forces indicate that the user is near the grid edge: up, down, left, and right. Force feedback joysticks allow direct interaction with diverse degrees of freedom. A plastic graphic grid is posed on the tablet to define the position of each token. A pen is used to point and select interface elements (see Fig. 8).

Fig. 6 AudioDoom input device for blind players

Educ Inf Technol (2007) 12:149–163

159

Fig. 7 AudioBattleShip input device for blind players

3 Guidelines and recommendations Virtual learning environments can be used to simulate features of the real world not physically available to users due to a wide variety of reasons. They become more realistic through multimedia displays that include haptic and auditory information. According to Colwell et al. (1998) and Paciello (1998), there are several domains in which virtual learning environments can be used to build educational software for people with visual disabilities: (a)

In education, a virtual laboratory assists students with physical disabilities in learning scenarios. Possible applications concern problem solving, strategic games, exploring spaces or structures, and working with concrete materials. Special virtual learning environment interfaces, such as head-mounted devices, the space mouse, and gloves, are often included. (b) Training in virtual learning environments deals with mobility and cognitive skills in spatial or mental structures. (c) Rehabilitation is possible in the context of physical therapy—a recovery of manual skills or learning how to speak and listen to sound can be targeted. (d) Access to educational systems is facilitated via dual navigation elements such as earcons, icons, and haptic devices.

Our idea is to support learners with visual disabilities in building conceptual models of real-world situations as sighted users do. Our approach is comparable to that introduced by Zajicek et al. (1998). We can identify five important common elements and aspects in modeling for virtual learning environments for children with visual disabilities: 1. The conceptual model is a consequence of mapping the real or fictitious world situation into a computer model using digital media by applying adequate modeling languages.

160

Educ Inf Technol (2007) 12:149–163

Fig. 8 AudioMemory input devices for blind players

2. The perceptual model is created by developing a perceptual model and a script for dynamic changes of the model. It can be perceived by the learner using only available information channels and considering the type of disabilities of end-users. It is important to provide surprising elements that call attention and enhance the perception process. The computer model description should be based on text. Explanation of graphic objects should be given in caption form. This text can be presented to the blind by using a text-to-speech plug-in and a Braille display. Intuitive correspondences between graphic and aural objects must be defined. Attention must be paid to the fact that only a small number of sounds can be memorized. Also, melodies that help to identify objects should be used. 3. The implementation design provides icons and earcons in parallel. If there are animated image sequences or video with sound, subtitles and moving text-banners should be used. Sound provides a rich set of resources that complement visual access to a virtual world. The four types of audio examined by Ressler and Wang (1998) are ambient background music, jingles, spoken descriptions, and speech synthesis. Ambient music can play and vary as the user moves from one room to another, providing an intuitive location cue. Jingles or small melodies should characterize special objects and control elements. Spoken descriptions of objects can play as the viewer moves closer to an object. Speech synthesizers can read embedded text from the nodes in a scene graph representation. Recent Web languages provide anchor node descriptions, EnvironmentNodes, or WorldInfo nodes. Internet-accessible speech synthesizers supply easy access to text-to-speech technology. Icons should be simple and big, with clear messages and fine color contrast with backgrounds. They should be easily interpreted by users through voice and sound to orient them and help them to understand their meaning. Buttons should be big and simple with associated voices and sounds for children with visual disabilities to orient them and map their function.

Educ Inf Technol (2007) 12:149–163

161

4. The implementation tools concerning special editors or languages such as Java, Java3D, Virtual Reality Modeling Language (VRML), OpenGL, and DirectX should be used. VRML defines a standard language for representing a directed acyclic graph containing geometric information nodes that may be communicated over the Internet, animated, and viewed interactively in real-time. VRML also provides behavior, user interaction, sensors, events, routes, and interpolators for interactive animation. User interaction, time-dependent events, or scripts can trigger dynamic changes in some properties of an object. VRML viewers are available not only as plug-ins to Internet browsers, but also as interactive objects that may be embedded in standard Office documents. However, the actual version does not yet support collaborative scenarios. As shown by Sánchez et al. (2003), a special platform for replicated collaborative applications such as MatchMaker (TNG) together with Java was successfully applied. This is a summarizing description of some tools that are currently available. The study focuses on those that are platform-independent and based on international standards. 5. The interface consists mainly of fonts, images, colors, icons, buttons, and audio. Fonts should be sans serif, avoiding the use of roman or serif typography, unless uppercase letters are used. Decorative fonts should be avoided. Size should be 18 points and up with medium line spacing and background contrasting colors. The best contrasting colors for people with low vision are yellow/blue and black/white. Images should have good contrast with backgrounds. They should be simple, clear, and precise, in such a way that users can easily recognize them with their residual vision. Iconic images should be preferred instead of photographic representations. Strong and contrasting blue and yellow spectrum colors should be used. These colors are better perceived by children with vision loss because they are diametrically opposed in both chrome and luminescence. Audio is critical in software for children with visual disabilities. It is the most relevant interface element in this type of software. Good quality audio should be included and clearly identified as to how and when it should be used for educational software purposes. Most interaction events should include audio such as to describe icons, buttons, menus, and contexts. Voices should be used depending on the target user. Software for children with visual disabilities should use interesting, motivating, and pleasant voices. Content refers to whenever software is designed for people with visual disabilities; the instructions should be audible. The interface for children with low vision should include clearly defined word messages with directions. Verbal directions should be clear, brief, and precise because the user should remember them when interacting with the software. Users should have the option to relisten to verbal directions when needed.

4 Conclusion This paper presents an integrated model to design virtual learning environments for learners with disabilities. We provide an overview of the state-of-the-art in the field of existing educational software for learners with visual disabilities. We focus on transferring the real world into appropriate computer representations by introducing an integrated methodology for modeling the real world for these people. The model starts with the definition of targeting cognitive skills. Then, a virtual environment is created including a navigable world through the use of a modeling language, dynamic scene objects, and acting characters. The learner explores the virtual world through appropriate

162

Educ Inf Technol (2007) 12:149–163

interfaces and obtains immediate feedback. Actions such as sound reproductions are collected, evaluated, and classified based on student’s modeling and diagnostic subsystems. The usability of the model is illustrated with virtual learning environments for blind children. We believe that by using models such as the one presented here, the process of software design can be facilitated and improved. Some virtual learning environments have already been developed for learners with visual disabilities. Then, by following our model, we can make generalizations and recommendations for designing and improving educational software. The defined architecture for the proposed model allows identifying software components that need to be considered to assure a correct impact on the learning of visual impaired users. We have identified each of these components in the implemented software, which helped us to understand the user’s interaction with other software elements. Likewise, this allowed us to get an abstraction about how the user learns when using virtual environments. This is a necessary analysis to assure correct and effective application of the educational software. The development of computer systems for learners with visual disabilities should no longer appear as isolated handcrafted efforts. Rather, efforts should be made to systematize the construction of these systems. Recent advances in hardware and software design support this idea and provide hope that the technological and educational foundation for such systems has already been laid. In this paper, we present interfaces to provide user interaction and feedback. Mostly they are either haptic or sound-based interfaces. Current research has highlighted that the use of spatial sound provides an added value to educational software, allowing users to make a precise mapping of the virtual environment and thus improving the interaction. Guidelines and recommendations for software design and implementation for users with visual disabilities have been included. Their correct use will assure adequate tuning between the software and methods of user interaction. Guidelines were designed for all abstraction levels in the process of software design and development, involving problem modeling (in a real or fictitious situation), I/O interfaces, and implementation (design of virtual environments). In other words, this model can be used to improve and extend the use of certain software to assist the learning of learners with visual disabilities. Finally, we expect to contribute to the field with an explicit and functional model that can be generalized and replicated to help improve the learning and cognition of users with visual disabilities. Acknowledgement This report was funded by the Chilean National Fund of Science and Technology, Fondecyt, Project 1060797.

References Baloian, N., Luther, W., & Sánchez, J. (2002). Modeling educational software for people with disabilities: Theory and practice. In Proceedings of ASSETS 2002 conference (pp.111–118). July 8–11, 2002, Edinburgh. Biella, D., Luther, W., Mietzel, G., Musahl, H.-P., & Wormuth, L. (2003). Replication of classical psychological experiments in virtual environments (Replicave). Proceedings of ECEL03, October 6–7, 2003, Glasgow. Clancey, W.J. (1987). Methodology for building an intelligent tutoring system. In G. Kearsley (Ed.), Artificial intelligence and instruction—applications and methods (pp. 193–227). Reading: Addison-Wesley. Colwell, Ch., Petrie, H., Kornbrot, D., Hardwick, A., & Furner, S. (1998). Haptic virtual reality for blind computer users. In Proceedings of the third annual acm conference on assistive technologies (pp. 92–99). April 15–17, 1998, Marina del Rey.

Educ Inf Technol (2007) 12:149–163

163

Dillenbourg, P., & Self, J. (1992). A framework for learner modeling. Interactive Learning Environments, 2 (2), 111–137. Hedgpeth, T., Rush, M., Black J. A., & Panchanathan, S. (2005). The iCARE iReader Project. CSUN’s 20th Annual International Conference “Technology and Persons with Disabilities”. March 14–19, 2005, Los Angeles, CA. Kass, R.(1989). Student modeling in intelligent tutoring systems—implications for user modeling. In A. Kobsa & W. Wahlster (Eds.), User models in dialog systems (pp. 386–410). New York: Springer. Lumbreras, M., & Sánchez, J. (1998). 3D aural interactive hyperstories for blind children. International Journal of Virtual Reality, 4(1), 20–28. Lumbreras, M., & Sánchez, J. (1999). Interactive 3D sound hyperstories for blind children. Proceedings of ACM CHI 1999 (pp. 318–325). Pittsburg. Na, J. (2006). The blind interactive guide system using RFID-based indoor positioning system. Proceedings of the ICCHP (pp. 1298–1305). LNCS 4061: Austria. Paciello, M. (1998). Making the Web accessible for the deaf, hearing and mobility impaired. In Yuri Rubinsky Insight Foundation. Retrieved from http://xml.coverpages.org/yuriInsightFound.html. Ressler, S., & Antonishek, B. (2001). Integrating active tangible devices with a synthetic environment for collaborative engineering. In Proceedings of the 2001 Web3D Symposium (pp. 93–100). Retrieved from http://www.itl.nist.gov/iaui/ovrt/people/sressler/tangible3.pdf. Ressler, S., & Wang, Q. (1998). Making VRML accessible for people with disabilities. In Proceedings of ACM ASSETS 1998, the Third Annual ACM Conference on Assistive Technologies (pp. 144–148). Retrieved from http://www.itl.nist.gov/iaui/ovrt/people/sressler/vrmlsigcaph.pdf. Roth, P., Petrucci, L., Assimacopoulos, A., & Pun, Th. (2000). Concentration game, an audio adaptation for the blind. In CSUN 2000 Conference Proceedings. Los Angeles: California State University. Sánchez, J. (2001). Interactive virtual acoustic environments for blind children. In Proceedings of ACM CHI 2001 (pp. 23–25). Seattle, April 2–5, 2001. Sánchez, J. (2006). A model to design multimedia software for learners with visual disabilities. Education for the 21st century—impact of ICT and digital resources, IFIP 19th world computer congress, TC-3 education (pp. 195–204). August 21–24, 2006, Santiago. Sánchez, J., & Baloian, N. (2005). Modeling audio-based virtual environments for children with visual disabilities. In Proceedings of the world conference on educational multimedia, hypermedia & telecommunications ED-MEDIA 2005 (pp. 1652–1659). June 27–July 2, 2005, Montreal. Sánchez, J., Baloian, N., & Flores H. (2004). A methodology for developing audio-based interactive environments for learners with visual disabilities. In Proceedings of the world conference on educational multimedia, hypermedia & telecommunications ED-MEDIA 2004 (pp. 540–545). June 21–26, 2004, Lugano. Sánchez, J., Baloian, N., Hassler, T., & Hoppe, U. (2003). Audiobattleship: blind learners collaboration through sound. In Proceedings of ACM CHI 2003 (pp. 798–799). April 5–10, 2003, Ft. Lauderdale. Sánchez, J., & Flores, H. (2004). Memory enhancement through audio. Published in the proceedings of the sixth international ACM SIGACCESS conference on computers and accessibility, assets 2004 (pp. 24–31). October 18–20, 2004, Atlanta. Sánchez, J., & Lumbreras, M. (1999). Virtual environment interaction through 3D audio by blind children. Journal of Cyberpsychology and Behavior, CP&B 2(2), 101–111. Sánchez, J., Sáenz, M. (2005). Developing mathematics skills through audio interfaces. In Proceedings of 11th International Conference on Human–Computer Interaction, HCI 2005. July 22–27, 2005, Las Vegas. Sánchez, J., & Sáenz, M. (2006). 3D sound interactive environments for blind children problem solving skills. Behaviour & Information Technology, 25(4), 367–378. Savidis, A., Stephanidis, C., Korte, A., Crispie, K., & Fellbaum, K. (1996). A generic direct-manipulation 3D-auditory environment for hierarchical navigation in non-visual interaction. In Proceedings of ACM ASSETS 96 (pp. 117–123). Zajicek M., Powell C., & Reeves C. (1998). A web navigation tool for the blind. In Proceedings of ACM ASSETS’98, the 3rd ACM/SIGCAPH on assistive technologies (pp. 204–206). Los Angeles.