Evaluating Tutorial-Based Instructions for Controllers in Virtual Reality Games

Abstract

Virtual reality (VR) has disrupted the gaming market and is rapidly becoming ubiquitous. Yet differences between VR and traditional mediums, such as controllers that are visible in the virtual world, enable entirely new approaches to instruction. In this paper, we present four studies, each using a different VR game. Within each study, we compared three different modalities of tutorials: Text (text-only), Text+Diagram (text with controller diagrams), and Text+Spatial (text with controller tooltips appearing on top of the player's virtual controllers). Data from our studies show that the importance of tutorial modality depends greatly on game type. In a third-person shooter, Text+Spatial led to significantly higher controls learnability than Text and Text+Diagram, and also led to significantly higher performance, player experience, and intrinsic motivation than Text. In a puzzle game, Text+Spatial led to significantly higher controls learnability and performance than Text. Additionally, Text+Diagram led to significantly higher controls learnability than Text. However, in a wave shooter and a rhythm game, differences between conditions were negligible on all measures. Our studies show that game type is an important factor to consider when designing tutorial modality.