Case study: Reading Experience in VR (Work in progress)

This small research project was aimed to provide a guideline for how to display text in a different distance in virtual reality. This is a 2x3x3 experimental design that we compared the English reading experience use three type-faces, San-serif, Serif, and Landolt's C in three viewing distances, 0.5, 5 and 50 meters between two headsets, HTC Vive, and Oculus Rift CV1. Through the statistical analysis, the research suggested that placing the text in a middle range might provide a better reading experience, and participants' subjective feedback suggested that in general, Oculus Rift CV1 had provided a better experience in reading for both the text looks clearer and the headset was more comfortable to wear.

Introduction

 

Methods

Participants

12 students at the University of Washington participated in the study. Their vision was tested before the experiment that all participants have normal or corrected to normal vision and no color blindness or other vision related disability.  

General Procedures

The experiment consisted with a total of 180 trials, and required approximately 40 minutes to complete. The experiment was broken down into two experiment sections and one interview section with a five minutes break between sections. A vision test was conducted prior to the formal experiment to record the participant's vision data. During each section, the participant was asked to put on the VR headset and to response to the stimulus displayed in the headset. Six practice trials were presented at the beginning of each section and no data was recorded. For each formal section, the participant was instructed to response to the stimulus in the display, trails were randomized. The participant was asked to use Up and Down arrows keys to adjust the size of the stimulus, and press the spacebar when they consider the size of the stimulus was comfortable to read to proceed to the next trial. The display was changed during the break between sections and the sequence of used devices was counterbalanced. After the two experiment sections, the participant was interviewed with following questions:

  1. After wearing two headsets, did you noticed any difference between them?
    1. If so, what was the difference?
    2. Which display do you think has a better image quality?
    3. Which display do you think is better for reading the text in Sans-Serif? What about Serif? Why?
    4. Which font, serif or sans-serif (showing to two font on the screen) do you feel better for reading the text in VR? Why?
    5. If you could pick one of the two headsets to be used for office work, which one do you prefer? Why?
  2. Did you feel any discomfort while wearing the headset?
    1. If so, which headset caused discomfort?
    2. What kind of discomfort had you experienced?
    3. When did you feel discomfort?

Apparatus

The experiment program was created using Unity 5.5 (Unity Technologies, San Francisco, CA) and ran on a Windows 10 laptop (CPU: Intel® Core™ i7-4710HQ (Quad-Core, 6MB Cache, up to 3.5GHz w/ Turbo Boost) ; RAM: 16.0GB (15.9 GB usable); OS: Microsoft Window 10 Pro 1670, 64-bit Operating System, x64-based processor; GPU: nVidia GTX 970 4GB DDR5 Driver version 364.51), the HTC Vive and Oculus Rift was pre-calibrated before the session starts. Only one device was installed at the time, and swap with another device during the break between sessions. The participant was instructed to seat on the office chair with wheels that is placed ten feet from the table.

The two virtual reality headset used for the experiment were HTC Vive (HTC, with technology by Valve Corporation, Seattle, WA) and Oculus Rift CV1 (Oculus VR, Menlo Park, CA). HTC Vive equip with two 1080 x 1200 resolution OLED displays that supports the refresh rate of 90Hz and a field of view of 110 degrees. The tracking area is 5 x 5 feet with two sensors. HTC vive equip with two 1080 x 1200 resolution OLED displays that supports the refresh rate of 90Hz and a field of view of 110 degrees. The tracking area is 15 x 15 feet with two lighthouses. For this experiment, we used only one tracking sensor or lighthouse for each display since we do not need the user to move their head. The devices were pre-calibrated and tested to insure that there was no noticeable jitter or latency. 

The Landolt's C chart that was used to test the participant's vision was presented on a 24" vertical display (Display model: Dell 2407wfp) and the chair was set to 4 meters (13 feet) away from the display.

 

Stimulus

The experiment was a 2x3x3 experimental design with Device (HTC Vive, Oculus Rift CV1), type-face (Sans-serif, Serif, Landolt's C), and viewing distance (0.5, 5, 50 meters). The asset was created using Sketch. (Bohemian B.V.) Three type-faces were used in the experiment, San-serif (Calibri), Serif (Times New Romen), and Landolt's C. Every stimuli consisted of five characters of the same type-face, for Sans-serif and Serif, five characters were selected by random for every trials, for Landolt's C, five C will be shown in a different orientation. The location of the stimulus was randomized between 0.5, 5, and 50 meter from user's position in the virtual space. Ten trials were presented for each condition, five of which required the participant to scale the stimuli up and five required to be scaled down. The size of the stimuli depended on user's previous response that for scale-up trails, the preset size will be 0.8 times the size of the stimuli of the previous trial (of the same condition), and for scale-down trails, the preset size will be 1.2 times the size of the stimuli of the previous trail.

 

Results

A three-way ANOVA was used to analyze the data that the dependent variable was the font size and the Independent variables (factors) were Device (Rift, Vive), Viewing Distance (0.5m, 5m, 50m), and Type-face (Landolt’s C, Sans-Serif, Serif).

Discussion