Supporting presence in collaborative environments by haptic force feedback

ACM Transactions on Computer-Human Interaction (2000), pp. 461-476, doi: 10.1145/365058.365086
Experiments
Tasks
Findings

Does haptic force-feedback in a collaborative desktop virtual environment increase perceived virtual presence, perceived social presence, perceived task performance, and task performance?

Variables

    Haptic Fidelity
  • haptics (on/off) - haptic feedback from the phantom device or none (the device acted solely as a 3d mouse)
System Info

Displays
  • monitor - two 21" screens were used
Input Hardware
  • haptic - phantom haptic device (one a type, one t type)

Two GN Netcom Headsets with two earpieces and a microphone each; video and audio recording of the experiment; single master computer with bifurcated video signal to show each user the same virtual environment.

Participant Info

The participants were students from Lund University in Sweden, had no prior knowledge of the interface used, did not know each other, and had not had face-to-face communication before the experiment. Participants were paired up, one female and one male, for the experiment.

Total # Age Range Gender Balance
28 20 - 31

The two users lifted and moved the cubes together (by pressing in on either side of a cube and moving their devices together) in order to form a larger cube (Task A). They then created two lines out of the cubes (Task B). They then made two vertical piles out of the cubes (Task C). They then each explained one-half of the pattern from an illustration, then built the whole pattern (Task D). They then navigated together around the pattern created in task D (Task E). These tasks are grouped together here because the findings were presented in the paper for all five tasks together.

Interaction and Environment

Interface

Two users (one male and one female) in different locations interacted with the same virtual environment in order to accomplish the task, communicating through an audio headset. Users could change their location in the virtual environment by moving the PHANToM pen. In the haptics condition, they could hold onto each other by pressing a button on the pen, but in the no haptics condition, did so symbolically by keeping their cursors connected. Users manipulated the cubes through one-point interaction, so both users needed to be touching the same cube in order to move it, by pressing opposite sides simultaneously.

Eight cubes of four different colors (two of each) were placed in the simple room-like environment, initially on the floor. Cubes could be moved and had form, mass, damping, and surface friction (in the haptics condition) but did not rotate. The users were represented by different colored spheres. Each user saw the same view of the environment and could manipulate it through moving their PHANToM device pen.

Dimensionality Scale Density Visual Realism
3D Small Medium Low
Metrics

  • presence - 32-question questionnaire for perceived virtual presence; 8-question questionnaire for perceived social presence
  • time - total time taken to accomplish all 5 tasks for each pair of participants
  • user task perception - 14-question questionnaire
  1. There was a significant direct effect of haptics (on/off) on time for a object manipulation task.

    Subjects took a longer total time to complete the five tasks in the no haptics condition than in the haptics condition (35 minutes vs 24 minutes).

    Specificity: Somewhat general

  2. There was a significant direct effect of haptics (on/off) on presence for a object manipulation task.

    Subjects rated their perceived virtual presence higher in the haptics condition than in the no haptics condition (mean value of 5.4 vs 4.4 on the seven-point Likert-type scale).

    Specificity: Somewhat general

  3. There was a significant direct effect of haptics (on/off) on user task perception for a object manipulation task.

    Subjects rated their perceived task performance higher in the haptics condition than in the no haptics condition (mean value of 5.9 vs 5.1 on the seven-point Likert-type scale).

    Specificity: Somewhat general