Description

Concept: Physical presence.
The BIPs method was developed with the goal to gather information during the course of a VE experience. In order to do this, Slater and Steed (2000) make a link to Gestalt psychology. They hypothesize that there are two alternate gestalts: state V (virtual world) and state R (real world). Presence in the VE is defined as the extent to which interpretation V is favoured. Users are asked to report transitions from V to R (from R to V is not possible). A probabilistic model is constructed to model these transitions. This model is then used to estimate the equilibrium probability (p) of being present in the VE. An additional post-experimental question is needed to discriminate between participants who were in the presence state for more than half the time and less than half the time.

Research

An experiment was carried out (Slater & Steed, 2000) to evaluate the measure (n =20, between-subjects design). Level of activity (low/high) and place (different from lab/same as lab) were varied. Subjects entered a VE in which they had to move 3D chess pieces. During the experience they reported transitions from V to R. Afterwards they filled out a presence questionnaire, which contained 5 questions based on the SUS. A significant correlation was found between p and the questionnaire. Participants reported both external (e.g. sensory information from the real world intruding) and internal (e.g. objects looking unnatural) reasons for BIPs.
Brogni, Slater, and Steed (2003) report an additional study into the relationship between BIPs and self-reported questionnaire-based presence. In this case, they did not calculate p but took a simpler approach and merely counted the number of BIPs. Participants (n=60, between-subjects design) were distributed over 6 different virtual urban environments, which they experienced by means of a Cave-like system. The experience lasted 4-5 minutes; during this time participants were asked to report BIPs by pushing a button. After the experience, participants completed a 4-item version of the SUS. A significant regression was found of the number of BIPs on the subjective presence score. A significant negative correlation was found between the subjective presence score and the number of BIPs, meaning that more BIPs are associated with a lower subjective presence score.

Sensitivity: not reported
Reliability: not reported
Validity: correlation with subjective presence items.

Primary Source

Description

Concept: Physical presence.
Duration estimation, or the human ability to indicate how much time has elapsed, is proposed as a corroborative measure of presence. In previous literature, presence has been associated with both longer (Waterworth & Waterworth, 2001) and shorter (Lombard, 2000) experienced duration.
Subjects are asked to estimate the duration of the time interval they needed to complete tasks (in minutes and seconds). They are also asked to judge on a 6-point scale whether they think they completed the tasks in a short time.

Research

In an experiment exploring the relationship between duration estimation and presence, subjects (n=42, mixed design) navigated through a 3D maze using a route navigation system. Independent variables were type of info (map or text), which was varied between-subjects, and range of info (complete route, per subgoal, or per decision point), which was varied within-subjects. Dependent variables were time taken to complete the task, duration estimation and subjective presence, which was measured by 4 items on a 6-point scale.
A significant positive correlation was found between subjectively judged speed of task completion and sense of presence. No significant correlation was found between duration estimation and sense of presence.

Sensitivity: not reported
Reliability: not reported
Validity: validity of subjectively judged speed of task completion is supported by the correlation with subjective presence scores. Validity of duration estimation is not supported.

Primary Source

• IJsselsteijn, W., De Kort, Y., & Bierhoff, I. (2001). Duration Estimation and Presence. Paper presented at Presence 2001, Philadelphia, USA, 9-11 October 2001.

Other Literature

• Lombard, M. (1995). Direct responses to people on the screen: Television and personal space. Communication Research, 22, 288-324.
• Waterworth, E. L., & Waterworth, J. A. (2001). Focus, locus and sensus: The 3 dimensions of virtual experience. CyberPsychology and Behavior, 4, 203-214.

Description

Concept: Physical presence.
There are 2 alternatives for the relationship between presence and motion sickness:

1. A higher degree of presence provokes a greater degree of conflict between the visual and the proprioceptive senses, which leads to a higher degree of motion sickness
2. Motion sickness may distract the user and lower the sense of presence

Research

Wilson, Nichols, & Haldane (1997) used the SSQ in two presence experiments. In the first experiment (n=20) subjects filled out the Witmer & Singer PQ and the SSQ after immersion in a VE wearing a HMD. Only the interface subscale of the PQ showed a significant negative correlation with SSQ scores. In the second experiment (n=24, between-subjects design), subjects participated in a “duck-shooting” VE using either HMD or desktop. Both presence scores (items taken from several existing questionnaires) and SSQ scores were significantly higher in the HMD condition, and a significant positive correlation was found between these measures in this condition.

Sensitivity: SSQ scores discriminated between different systems
Reliability: not reported
Validity: similar pattern and correlation with subjective presence items

Primary Source

• Kennedy, R. S., Lane, N. E., Berbaum, K. S., & Lilienthal, M. G. (1993). Simulator sickness questionnaire: An enhanced method for quantifying simulator sickness. International Journal of Aviation Psychology, 3(3),203-220.

Other Literature

• Wilson, J. R., Nichols, S. & Haldane, C. (1997). Presence and side-effects: Complementary or contradictory? In M.J. Smith, G. Salvendy, & R. J. Koubek (Eds). Design of Computing Systems: Social and Ergonomic Considerations, Proceedings of the Seventh International Conference on Human-Computer Interaction, (HCI International '97) (pp. 889-892). San Francisco, USA.

Description

Concept: Physical presence.
Hoffman, Hullfish, and Houston (1995) have used the approach of source monitoring in order to measure presence. Virtual reality monitoring is defined as the decision process by which people distinguish between real, virtual, and imagined events, as represented in memory. ”Real, virtual and imagined environments differ in quality (e.g., in the amount of cognitive effort they require). These differences are preserved in memory, and can later serve as cues to where the memory originated”.
The Memory Characteristic Questionnaire (MCQ) was designed to assess this qualitative difference between experiences in memory. The questionnaire contains 21 items on a 7-point scale, and includes metamemory judgments concerning presence, attention, coherence, field-of-view, and similarity among environments. The items are listed in Appendix A.

Research

The MCQ was used in an experiment comparing real, virtual, and imagined worlds. Participants (n=16, within-subjects design) encountered 24 worlds (arrangements of shapes on a chess board) which could be real, virtual, or imagined. Afterwards, they completed a Virtual Reality monitoring test (recognition of encountered worlds) and the MCQ, in which they answered each item 3 times (for real, imagined and virtual worlds each). The results section discusses only 1 item of the MCQ, the one related to cognitive effort. Cognitive effort is highest for imagined worlds, lower for virtual worlds and lowest for real worlds. Results for other MCQ items are given in Appendix E and show several significant differences.

Sensitivity: several items discriminated between real, virtual, and imagined worlds
Reliability: not reported
Validity: not reported

Primary Sources

Description

Concept: Physical presence.
Darken, Bernatovich, Lawson, & Peterson (1999) argue that selective attention is an important component of presence. Their approach is based on Held & Durlach’s notion of presence as an “alternate experience”. In order to be present in an alternate world, attention must be focused there rather than on the real world.

Research

In an experiment by Darken et al. (1999), participants (n=70, between-subjects design) experienced a virtual world; a videotape (Wallace and Gromit) was simultaneously shown. Independent variables were type of visual display (flatscreen, mini-CAVE or HMD), presence/absence of sound, and whether or not participants were primed to the presence of the dual task. Attention/engagement was measured through quizzes about content of the virtual and real (videotape) world experiences. Subjects also completed the Witmer & Singer PQ. Attention scores were highest in the mini-CAVE condition, lower in the flatscreen condition and lowest in the HMD condition. For the virtual world attention scores these differences are not significant, for the real world attention scores they are. In the condition where sound was present, attention scores for the virtual world were higher (as expected), but scores on real world attention were also higher. Authors explain this by suggesting that the presence of sound allows users to divide attention along a multi-modal axis. A significant relation (regression analysis) was found between attention scores and PQ scores.
Nichols, Haldane, & Wilson (2000) measured background awareness by recall of background music. In their experiment (n=24, between-subjects design) using a VE representing a duck-shoot fairground stall, independent variables were system (HMD or desktop) and audition (full or none). Presence was also measured subjectively by 3 items. There was a negative correlation between 1 presence item (visiting) and background awareness scores. No difference was found between awareness scores in the different conditions.

Sensitivity: one of Darken et al.’s measures (the real world attention score) distinguished between different systems. Nichols et al.’s measure did not discriminate between different conditions.
Reliability: not reported
Validity: Darken et al. found a significant correlation between their measure and PQ scores.

Primary Sources

• Darken, R. P., Bernatovich, D., Lawson, J., & Peterson, B. (1999). Quantitative measures of presence in virtual environments: The roles of attention and spatial comprehension. CyberPsychology and Behavior, 2, 337-347.
• Nichols, S., Haldane, C., & Wilson , J. R. (2000). Measurement of presence and its consequences in virtual environments. International Journal of Human Computer Studies, 52, 471-491.

Description

Concept: Physical presence.
Similar to their attentional measures, Darken, Bernatovich, Lawson, & Peterson (1999) base this approach on Held & Durlach’s notion of presence as an “alternate experience”. If presence is the sense of being in another place, the amount of spatial information that is remembered from the VE can be used as a presence measure.

Research

Darken et al (1999) used pointing, map building, and landmark selection spatial tests. Participants (n=40, between-subjects design) experienced a virtual world with a mini-CAVE system. The independent variable was sound; the 4 conditions were: no sound, semantic info only, spatial info only, or both spatial and semantic info. The PQ was also used as a presence measure. The presence of sound in any form was found to increase PQ scores and landmark selection scores, but not the scores on other spatial tasks. No relation was found between PQ scores and spatial comprehension.
Dinh, Walker, Song, Kobayashi, & Hodges (1999) used 4 spatial layout questions and 5 object location questions to assess spatial memory of a VE depicting an office (n=322, between-subjects design). Level of visual detail, olfactory stimulation, ambient auditory stimulation and tactile stimulation were varied as independent variables. No effects were found on the spatial layout scores. A significant positive effect of tactile cues and olfactory cues was found on object location scores.

Sensitivity: 2 out of 3 spatial tests used by Darken et al.(1999) were sensitive to the presence of sound. An object location questionnaire used by Dinh. et al. (1999) was sensitive to tactile and olfactory cues.
Reliability: not reported
Validity: inconclusive. In both studies, presence-enhancing factors influence some spatial measures, but not others.

Primary Sources

Description

Concept: Physical presence.
Mania, Troscianko, Hawkes, & Chalmers (2003) developed a methodology for assessing simulation fidelity of a VE based on human judgment of memory awareness states. Performance alone is seen as an imperfect reflection of the subjective experience that underlies performance in memory tasks. A distinction is drawn between different awareness states: “remembering”, which is accompanied by a specific recollection of the source, and “knowing”, which is a general sense not accompanied by such a recollection.
A questionnaire was designed to test memory recall of the positions and geometric shape of objects in the VE. A diagram of each wall in the room included numbered positions of objects. The questionnaire consists of 21 multiple choice questions; one for each object in the room. Every question includes 3 possible answers (box, sphere, pyramid), a confidence scale with 5 possible states (no confidence, low confidence, moderate confidence, confident, certain), and an awareness state report (remember, know, familiar, guess). Awareness state responses are taken to reflect the amount of visual mental imagery involved during retrieval.

Research

Participants (n=105, between-subjects design) were distributed over 5 conditions: real world, HMD mono head-tracked, HMD stereo head-tracked, HMD mono mouse, and desktop. Apart from the memory recall task participants also completed the SUS presence questionnaire. The recall task was completed 2 times: right after the experiment and one week later.
There was a significant main effect of condition upon awareness state. Although it was expected that the more naturalistic head-tracking interface would be associated with more correct responses, this was not the case: the amount of correct responses in the “remember” state was higher for HMD mono mouse condition than for both HMD tracking conditions. These responses correlated positively with confidence scores. SUS results are not reported.

Sensitivity: questionnaire scores discriminated between conditions, although not in the expected direction.
Reliability: not reported.
Validity: not reported.

Primary Source

• Mania, K., Troscianko, T., Hawkes, R., & Chalmers, A. (2003). Fidelity metrics for virtual environment simulations based on spatial memory awareness states. Presence: Teleoperators and Virtual Environments, 12, 296-310.

Description

Concept: Physical presence.
By manipulating the orientation of a VE, visual information is provided about spatial location, which is different from information provided by non-visual stimuli. Nemire, Jacoby, and Ellis (1994) propose that the extent to which the visual information biases participant’s estimate of spatial orientation can be used as a measure of simulation fidelity. Participants are asked to estimate their eye level when viewing a pitched VE through a HMD. The GREL measure is calculated based on actual eye level and estimated eye level. The formula is described in the paper.

Research

Participants (n=12, mixed design) wore a HMD showing a virtual box. The orientation of the VE was manipulated by pitching it around the horizontal axis. Participants saw 5 different angles, with varying optic structure of the visual stimuli: no grid, transverse partial grid, longitudinal partial grid, full grid. Participants, who had to remain motionless themselves, were asked to indicate their eye level. Data were compared to a similar experiment carried out in a physical environment. Results show that GREL was biased by direction of the VE pitch. For simple or transverse partial grid conditions, this bias was greater in the physical environment than in the VE; for longitudinal partial grid and full grid conditions the bias of the VE was not different from the bias of the physical environment.

Sensitivity: GREL scores were affected by tilting angle
Reliability: not reported.
Validity: not reported.

Primary Source

• Nemire, K., Jacoby, R. H., & Ellis, S. R. (1994). Simulation fidelity of a virtual environment display. Human Factors, 36, 79-93.

Description

Concept: Physical presence.
The subjective tilt angle measure used by Hatada, Sakata, Kusaka (1980) is based on the idea of “unification of display and observer space”, which produces a feeling of seamless continuity under which the observer is presented with the information contained in the displayed picture. Objective, visually obtained information influences the condition of the observer’s subjective coordinate axis, which can be used as a measure of the sensation of reality. A subject is placed in front of a hemispherical glass-bead concave 180 screen, looking at it in a vertical position.

1. Without a picture being shown, a tilted line target is presented to which the observer will adjust so that it will appear vertical
2. Still without a picture, the direction is determined at which the subject feels that the target line is vertical
3. The subject is shown a tilted picture for 15 seconds
4. The line target is shown again and it is determined at what tilt angle the observer feels the line is vertical.

Research

Earlier work has shown that the effects on the observer increase when the view angle is widened. View angle was found to have a similar effect on the ‘sense of reality”, measured by a subjective 7-step scale (no further explanation). The effect was also shown to be stronger when pictures contain pronounced perspective elements.

Sensitivity: method distinguished between different viewing angles and differences in content
Reliability: not reported.
Validity: similarity in results obtained by measuring subjective tilt angle and subjective sense of reality.

Primary Source

• Hatada, T., Sakata, H., & Kusaka, H. (1980). Psychophysical analysis of the sensation of reality. SMPTE Journal, 89, 560-569.