To move or not to move? The effect of active versus passive pre-training on cognitive load and in-game performance in an AR game

Abstract: A distinct feature of educational games using augmented reality (AR) is that the game is played through physically interacting with the environment, whereas physical interaction is typically rather limited in other digital games. Understanding and performing the interactive game mechanics can be cognitively demanding. Adding pre-training could help students manage cognitive load during in-game performance. However, traditional approaches of pre-training (eg, paper-based or video sequences) might not be sufficient, given the crucial role of physical interactions in educational AR games. In the present study, primary and early secondary school students (N = 255) were randomly assigned to an active pre-training, which involved students practising the movements needed in the game or a passive pre-training, where students watched a video explaining the game. The aim was to investigate whether active pre-training reduces students' cognitive load and improves in-game performance. It was also examined whether these effects were dependent on students' visuospatial working memory capacity (WMC). Results showed no significant differences between the conditions regarding students' cognitive load and in-game performance. However, visuospatial WMC predicted students' dropout of the game. This suggests that observing the movements in the passive pre-training might be similarly effective as enacting the movements. Nevertheless, more research is needed to gain a better understanding of how different levels of WMC impact learning with educational AR games and how students with low visuospatial WMC can be efficiently supported. Practitioner notes What is already known about this topic Working memory capacity influences students' ability to successfully perform complex tasks. Not every learning medium is equally effective for students with different levels of working memory capacity. Pre-training in educational games can support learning and increases the academic performance of students. Educational games with AR can be cognitively demanding, since students need to understand the interactions with the used device in addition to the game content and mechanics. It is unclear how to sufficiently support students in AR environments. What this paper adds Both active and passive pre-training are equally effective. Students with lower working memory capacity were more likely to drop out during the educational AR game than students with high working memory capacity. Implications for practice and/or policy Both practising and watching movements may be helpful for students to learn the relevant game mechanics to interact with the AR environment. There are students who do not optimally benefit when AR games are implemented in education and may need additional support.