Walking > walking-in-place > flying, in virtual environments

Proceedings of the 26th annual conference on Computer graphics and interactive techniques - SIGGRAPH '99 (1999), pp. 359-364, doi: 10.1145/311535.311589
Experiments
Tasks
Findings

Do participants experience a higher sense of presence when walking normally, versus walking-in-place and point-and-fly?

Variables and Constants

    Biomechanical Symmetry
  • kinematic - walking/walking-in-place move thighs, legs and feet to step in place. flying does not.
  • kinetic - walking: vertical and ground forces felt. walking-in-place: vertical forces only. flying: none
  • anthropometric - the walking techniques have similar body segments in place, point-and-fly does not.
    Control Symmetry
  • transfer function - point-and-fly has very little fidelity, walk-in-place has medium fidelity. real-world walking has high fidelity.
Independent Variables
  • interaction technique - real-world walking, walking-in-place, and point-and fly with respect to the floor plane.
Constants

Components relevant to input, such as the tracker, and graphics, relevant to the HMD were constant for all interaction techniques.

System Info

Displays
  • head-mounted display - virtual research v8 head mounted display with true vga resolution of (640x3) x 480 pixels per eye — 307,200 triads. this display consists of two 1.3 inch active matrix lcds with a field of view of 60 degrees diagonal at 100% overlap and aspect ratio 4:3.
Input Hardware
  • joystick - with 4 buttons, tracked
  • tracker - on the ceiling, tracked joystick and hmd. the system had an overall latency of about 100 ms with a lag of about 500 ms for walking-in-place.
Software

The experiments used a Silicon Graphics Onyx2 with one graphic pipe, two raster managers, four 195 MHz R10000 processors and 2 GB of main memory.

Participant Info

A total of 33 “naive” subjects participated in the study. The requirement was that they have no knowledge of the goals of the experiment. Each subject was paid $10. Naive subjects were grouped into flyers, virtual walkers, and real walkers, each with 6 men and 5 women. Another 11 subjects (10 men, 1 woman) were “expert” users who had experienced immersive virtual reality on several occasions and were generally working in the area of computer graphics.

Total # Age Range Gender Balance
44 -

This task was to grasp the green box in the training room and carry it to the chair in the virtual pit room. Picking up the green box automatically opens the door between the two virtual rooms. Subjects were free to choose the path to the chair, either going along the ledge, left or right, or moving directly to the chair over the pit.

Interaction and Environment

Interface

Interaction varied as a variable. The virtual environment consisted of the Training Room and the Pit Room. The Pit Room consisted of a 2-foot wide virtual ledge above a 20-foot (virtual) drop to a room below containing living room furniture. There was also a chair on the wooden ledge on the far side of the room.

Dimensionality Scale Density Visual Realism
3D Large Medium Medium
Metrics

  • cybersickness - kennedyv et al.,'s 1993 simulator sickness questionnaire was used
  • presence - presence was measured in three components: subjective presence, observed behavioral presence (through behaviors found with video analysis), and physiological presence.
  • user task perception - ease of locomotion
  1. There was a significant interaction between kinematic, kinetic, anthropometric, transfer function, and interaction technique on user task perception for a travel - maneuvering task.

    Real world walking was easier than walking in place and point-and-fly in navigating through VE's.

    Specificity: Neither

  2. There was a significant interaction between kinematic, kinetic, anthropometric, transfer function, and interaction technique on presence for a travel - maneuvering task.

    Point-and-fly led to lower sense of presence than real-world walking and walking-in-place for navigating through a VE.

    Specificity: Neither