Analyzing Accurate Egocentric Distance Estimates of University Students in Virtual Environments with a Desktop Display and Gear VR Display

Abstract

The perception of distances is crucial in both the real world and virtual environments. However, distances can be incorrectly estimated in the latter one, and they can be affected by technological and human factors. We created a virtual environment to take a step toward understanding this phenomenon. We assessed the egocentric distance estimation skills of 239 university students at 10 various distances between 25 cm and 160 cm at 15 cm intervals. A desktop display was used by 157 students, while the Gear VR display was used by 72 students. The effects of the following factors were analyzed: gender, height, dominant arm, previous VR experience, gaming hours per week, whether the participants wore glasses, their field of study, and display device. Logistic regression analysis was performed to assess their influences on the probabilities of accurate distance estimates, while linear regression analysis was conducted to examine their effects on estimation times. The results show that except for the factors of whether the participants wore glasses and their field of study, the probabilities of accurate distance estimates can be affected along with estimation times themselves.