Investigating the Behavioral Effects of Longitudinal Exposure to Immersive Consumer Virtual Reality
The project was funded from September 1st, 2017 to August 31st, 2021 by NSF Award #1717937.
As systems for consumer virtual reality (VR) become more available, people will start to spend more and more time in VR environments for education, work, entertainment, and communication. This project will look at how people change as they use VR systems over time, as well as at how more experienced VR users compare to novice users and how well existing studies of novice behavior will predict experienced user behavior. The team will first analyze data from online forums where people talk about their experiences in VR to develop an understanding of how people have already begun to use VR systems. They will then compare how people explore and become skilled in simulated environments using VR versus using traditional interfaces, using the Minecraft environment as a testbed and collecting data about people’s behavior in the environment, physiological markers, and self-reports about their experiences. Finally, they will conduct a longitudinal study of people performing common VR tasks both before and after spending a considerable amount of time using VR. The goal is to use these multiple data sources to study how long-term exposure to VR affects both people’s behavior and their engagement with VR, and to use that to advance both psychological understandings of how people interact with virtual environments and the design of such systems. In doing the work, the team will develop tools to collect behavioral data from Minecraft and release both those tools and the datasets collected for the benefit of other researchers and teachers.
The project is organized in three main phases. The first phase will use thematic coding of the first six months of posts from forums devoted to each of the main consumer VR technologies, along with interviews of contributors to those forums guided by those themes, to identify how people react as they experiment with VR technologies and incorporate them into their lives, as well as key differences between available implementations that affect those reactions. The second study will ask participants to use both VR and standard desktop interfaces for six hours exploring Minecraft, which supports a wide variety of possible behaviors and provides a good API for collecting data about both those behaviors and the environment. These data, along with video and biometric recordings, will be analyzed around five core elements of VR experiences: navigation through, interaction with, and perception of the environment itself; and feelings of both individual presence in the environment and social presence of other people and characters. In the third study, the team will ask VR novices to complete a series of VR tasks that correspond to the five core areas, then to use a VR version of Minecraft for at least 30 hours over the course of several months, then to complete the tasks again. This will allow them to compare novice and experienced users' ability to navigate and interact in VR environments, as well as effects of VR expertise on key values claimed for VR such as presence and key barriers such as simulator sickness.