In order to effectively utilize the capabilities of virtual reality (VR) in supporting the desired learning outcomes, careful consideration in the design of instruction for VR learning is crucial. In line with this concern, previous work proposed an instructional design model that prescribes instructional methods to guide the design of VR-based learning environments. This article provides a thorough elaboration on how formative research is employed to enhance the earlier model. The study has successfully generated five new hypothesized principles to enhance the robustness of the instructional design model through the formative research process. The newly derived hypothesized principles also provide insights into the design of various experimental studies for testing them in the effort to form a more comprehensive guide for the design of VR-based learning environments.

No abstract

Designing a usable learning application is one of the key factors in ensuring effective learning. This article introduces modified group usability testing (MGUT) as a feasible framework for evaluating the usability of nonimmersive virtual reality (VR) learning applications. Conventionally, usability testing of such learning applications often employs the one-to-one approach in which an evaluator conducts testing with several individual participants. As opposed to the one-to-one approach, the group approach involves several-to-many participants performing tasks simultaneously, with several evaluators observing and interacting with participants. This article describes the complete step-by-step procedure for conducting MGUT to uncover usability problems of a VR learning application that aims to educate its users on fire safety and prevention. It also proposes methods to analyze these usability problems. The effectiveness and efficiency of MGUT was then compared with DGUT, the original group testing technique and cooperative evaluation (CE), which is a typical one-toone approach. Results indicate that all three techniques are able to reveal usability problems of different usability factors and show similar capability to discover the most critical and serious problems. MGUT is more effective than DGUT as it can collect additional usability problems of various factors and of different levels of severity. MGUT is as effective as CE as both techniques can identify usability problems which are more or less comparable in terms of quantity and quality. As for efficiency, MGUT and DGUT are more efficient than CE as these group testing approaches require lesser testing time, lesser effort in terms of the intensive interaction with participants although with slight more effort in the preparation of the physical setting. In addition, it is also obvious that MGUT and DGUT involve richer participation than CE. MGUT is also more feasible than DGUT as it allows some flexibility in the computer arrangement setting.

The study employs an exploratory approach to examine the satisfaction of normal and dyslexic learners toward different web text modes. As an average of 10 % of the population shows some traits of dyslexia, presenting web text solely based on the guidelines for normal web users will put users with dyslexia at disadvantage. Due to the ubiquitous use of the web for online learning purposes and the availability of tremendous amount of text on the web, this investigation intends to derive appropriate guidelines for presenting web text that could accommodate both groups of learners. This qualitative study uses a multiple case study design and data are mainly collected via observations and guided interviews. The study reveals that existing dyslexia-friendly text guidelines are also appropriate for normal learners and the use of screen reader, an assistive technology that reads text aloud, does not fit every dyslexic and normal learner.

This paper proposes an hybrid Artificial Neural Network (ANN) with Self-Organizing Map (SOM) and modified Adaptive Coordinates (AC) for multivariate dimension reduction and data structures exploration. SOM, being a prominent unsupervised learning algorithm, is often used for multivariate data visualization. However, SOM only preserved input space inter-neurons distances and not in the output space because of SOM rigid grid. SOM grid provides little information for visual exploration of the clustering tendency of the multivariate data. Modified AC is therefore proposed to remove SOM's map rigidity and provides better data topology preserved visualization. Empirical study of the hybrid yielded promising topology preserved visualizations for synthetic and benchmarking datasets.