Virtually Lost in Learning: Improving Navigational Efficiency in Virtual Reality Leads to Enhanced Learning

Abstract

Virtual Reality (VR), presented in many sci-fi movies and series such as Ready Player One and Star Trek, is often hoped to revolutionize training and teaching as it offers more immersion in the experience than any other medium. The benefits of VR and serious games for teaching are increased presence (i.e., players feeling being physically present in the game, engagement (i.e., heightened concentration, interest, involvement, and enjoyment), and cognitive interest (i.e., understanding topics and becoming more interested), which have all been linked to increased learning. However, the expected benefits of VR have not been fully proven, only showing that there are advantages when it comes to spatial information and a better in-game experience in the form of greater presence, improved engagement (van den Broek, 2012), and higher cognitive interest. When it comes to story-based information, there are conflicting results, with some research finding better learning performance, no difference, or even worse learning performance. As such, it could be argued that, currently, the benefit of using this technology is not worth the additional cost. To enable optimal learning from VR serious games, it must be investigated how to effectively develop these games to catalyze learning. To ensure that optimal learning occurs, this thesis beings by detailing a technique to ensure that the in-game story matches the gameplay, before evaluating how to assist players in VR serious games. The lostness measure was identified to evaluate player performance in VR serious games. Originally from the domain of hypertext, this measure is used to measure navigational efficiency in an information-searching task. Due to the node-based movement and information-searching tasks present in many VR serious games this proved to be a new useful game analytic. It helped show that removing the cognitive burden of navigation can improve learning but can also result in a less positive gaming experience. This is endorsed by the cognitive load theory, the amount of available memory available for cognitive data processing. Cognitive overload and underload can both hinder learning. For optimal learning performance, players should remain within the Zone of Proximal Development (ZPD). Based on the findings of navigation playing an important role in cognitive load and ZPD, an adaptive system is presented that makes use of textual specificity. This system works by presenting more specific instructions to narrow the search space, making it easier to find educational content, and, conversely, presenting less specific/generic instructions, making it more challenging to find educational content. This forces players into the center of the ZPD and gives them an appropriate level of challenge. It was found that, without having any impact on the in-game experience of a player, the adaptive system was able to reduce cognitive load and increase knowledge retention in both the story and spatial dimensions.