=Paper=
{{Paper
|id=Vol-3090/paper42
|storemode=property
|title=Measurement of Presence by a Presence Counter Based on Breaks in Presence
|pdfUrl=https://ceur-ws.org/Vol-3090/paper42.pdf
|volume=Vol-3090
|authors=Natalya Averbukh,Boris Velichkovskiy
|dblpUrl=https://dblp.org/rec/conf/ims2/AverbukhV21
}}
==Measurement of Presence by a Presence Counter Based on Breaks in Presence==
https://ceur-ws.org/Vol-3090/paper42.pdf
Measurement of Presence by a Presence Counter
Based on Breaks in Presence
Natalya V. Averbukh1 , Boris B. Velichkovsky2,3
1
Ural Federal University, 19 Mira street, Ekaterinburg, 620002, Russian Federation
2
Faculty of Psychology, Moscow State University, Mokhovaya bld. 11/5, Moscow, 125009 Russian Federation
3
Moscow State Linguistic University, Ostozhenka 38 bld. 1, Moscow, 119034, Russian Federation
Abstract
The paper is concerned with methods of measuring presence. The physiological and behaviour methods
are situation-specific whereas subjective method is used after virtual session. In this paper we focus on
the method that evaluates the presence counter based on breaks in presence (BiP). The advantage of this
method is its implementation during the virtual session, during the virtual experience.
Within the experiment, 22 participants walked around the virtual Asian park and searched for virtual
vases. They gave signals when BiPs occurred. The presence counter can be calculated based on this
data. To validate this approach the participant were asked to answer ITC-Sense of Presence Inventory
(ITC-SOPI) after the virtual session.
Our approach of computing the presence counter involves the Markov chain that is a simplest model
of stochastic process.
Presence counter based on BiP and the Markov chain was proposed by Slater and Steed. They
consider the discrete time model. In the paper we revisit this approach and, additionally, develop the
continuous time Markov chain based method of presence counter. The calculation of the correlation
between the presence counters based on BiP and results of ITC-SOPI showes that the counters relied
on the continuous time Markov chain are most sensitive. This paper shows that a BIP-based presence
counter can be used as an effective presence measure.
Keywords
Virtual reality, presence, mediated presence, presence measurement, breaks in presence, Markov chains,
presence counter
1. Introduction
The virtual reality is a special technology that makes it possible to create an interactive three-
dimensional environment. The presence is the main phenomenon in the study of virtual reality.
”...The phenomenon of Presence is that an individual experiences the illusion of being present
in the same reality with objects or subjects that are not in the directly observable reality of
the individual. It is necessary to make a reservation at once that in this context we are not
talking about the situation of full consciousness of the individual that the reality he feels is in
fact artificially created or caused to exist in another way” [1, p.38].
IMS 2021 - International Conference ”Internet and Modern Society”, June 24-26, 2021, St. Petersburg, Russia
Envelope-Open natalya_averbukh@mail.ru (N. V. Averbukh); velitchk@mail.ru (B. B. Velichkovsky)
Orcid 0000-0002-8232-6711 (N. V. Averbukh); 0000-0001-7823-0605 (B. B. Velichkovsky)
© 2021 Copyright for this paper by its authors. Use permitted under Creative Commons License Attribution 4.0 International (CC BY 4.0).
CEUR
Workshop
Proceedings
http://ceur-ws.org
ISSN 1613-0073
CEUR Workshop Proceedings (CEUR-WS.org)
424 PART 2: Internet Psychology
The authors of [2] call the experience of presence in the virtual world mediated presence. In
their opinion, the mediated presence is a sense of presence in the external world, in the realization
of which technology plays an important role. When a person experiences a strong mediated
presence, his/her experience shows that the technology has become a part of themselves, and
the mediated reality has become a part of the Other. In the context of this work, the presence
will be understood as an mediated presence.
The presence as a phenomenon can be recorded and measured. The question of selecting
measurement methods is extremely important. The subjective methods, primarily questionnaires
and inventoies, can be used. The behavioral and physiological measurement methods can also be
used. Note that the subjective methods depend on the subjective opinion of the subjects and they
do not measure presence at the moment of experience. On other hand, they are universal. The
behavioral and physiological methods depend on events that trigger behavioral and physiological
responses. Slater proposed a method based on breaks in presence (BiP) [3]. Most participants
in virtual environments do not experience being present all the time while interacting with VR.
They can hear sounds from the real world and feel the touch of objects from the real world.
These sensations can cause the consciousness moves from the virtual environment to the real
one. However, if a person reports a transition from the virtual environment to the real one, it
means that they have just felt themselves in the virtual environment. In this way, messages
about breaks in presence help to measure the presence. This method allows us to know that a
participant experiences the presence at the moment when he experiences it. Slater measured
the presence based on the CAVE technique and the movement around the chess table with 3D
chess. The movement was carried out by walking. Slater used a mathematical model of discrete
time Markov chains.
The main question of this study is whether it is possible to measure the presence by the
BiP method using Markov chains in the HMD technology and moving by walking for small
distances and teleporting for significant distances? Will the BiP-based presence counter measure
the experience that is usually regarded to as presence? What methodological features should
be taken into account? To address these question we extend Slater’s approach to continuous
time Marcov chain. Moreover the validity of the BiP-based presence counter was evaluated by
correlation with the ITC-Sense of Presence Inventory (ITC-SOPI) [4].
Notice that in this paper, we did not set out to link the presence with any individual charac-
teristics of the participants, we were only focused on the validity of the presence counter.
2. Background
A large number of publications have been devoted to the presence, since the 90s. From relatively
old reviews, for the purposes of this paper, it is interesting [5].
Now, a number of reviews of more recent works related to the concept of presence have been
published. The paper [6] provides a theoretical overview, and the paper [7] discusses methods
for measuring presence, and suggests, according to the authors, unused methods. The paper
[8] discusses about immersion, social presence and co-presence as the ability to be somewhere
together, to perceive and be perceived by other people, to interact with people. The paper [8]
also raises the question of realism and veracity.
IMS-2021. International Conference “Internet and Modern Society” 425
Close to this is the discussion of perceptual and social realism as important aspects of the
presence [9]. Perceptual realism occurs when the environment supports the action in it, when
the response of the environment is perceived as plausible and adequate. Social realism refers to
a more general concept: when an event that occurred in the environment is plausible, when it
can happen in the real world. There can be high perceptual and low social realism in a virtual
environment.
These concepts echo Slater’s earlier ideas of place illusion and plausibility [10].
The paper [11] mentions three main approaches to the study of the presence: the mediated-
objective school of thought approach, mediated-subjective school of thought approach and
inner presence school of thought approach. The first two schools describe the presence as an
essential element in mediated experience. The third describes the presence as a phenomenon
that does not require median systems (virtual reality technologies, etc.).
The schools of the mediated presence define presence as the perceptual illusion of immediacy.
[11] criticizes this approach, although pointes out that schools of mediated presence provide
good definitions of a number of concepts, such as immersion and involvement. According to
[11], the mediated presence approach does not answer important questions: why do we feel the
presence, what is its role? Trrberti and Riva, the researchers of inner presence school raise the
similar questions [12]. They also talk about the schools of mediated presence, which, according
to them, do not answer questions about the evolutionary cause of the presence, about its causes
and purpose.
Representatives of the third school in their works define the phenomenon of presence as a
conscious sense of being in the external world, as a phenomenon that controls the division into
the inner and the outer [13], [14], [15], [16], [17], [18], [19], [20], [12]. They suggest treating the
presence as a presence in any environment, not necessarily created by virtual reality technology.
They see the presence as a central part of conscious mental life [20]. The sense of presence,
according to these researchers [12], allows you to constantly adapt your own activities to the
external environment. This approach is also related to the concepts of intentions and actions:
the more the environment allows you to implement intentions, turn them into actions, the
stronger the experience of presence.
There are other works in which presence is considered more broadly than presence in a virtual
environment. For example, [21] assumes that the presence is associated with the successful
implementation of intentions, so that a person feels like a successful author of their own actions.
In [? ], the presence is considered as a personal state that occurs in both real and virtual
environments, and mediation is not only a technique for creating virtual reality, but in general,
any situation where actions are implemented indirectly, so presence is analyzed by the example
of driving a car to demonstrate the flexibility of this approach.
Another interesting work [22] is devoted to the relationship between the presence and
the perception of own movement in a virtual environment. In contrast to the researchers
of the school of inner presence, who suggest that the virtual environment creates limitless
possibilitiesunlimited opportunities [22] indicate that virtual environments, on the contrary,
are limited in comparison with reality, in particular, it is very difficult to move. In [22] the
presence and illusory self-movement in a virtual environment are linked. The experimental
material shows that the presence is related to how one’s own motion is perceived in a virtual
environment.
426 PART 2: Internet Psychology
We should also mention the works [23], [24], [25], devoted to the relationship of metacognitive
processes and the presence. Metacognitive processes, such as monitoring and error correction,
prevent cybersickness, but can also prevent the appearance of the presence, since these cognitive
processes lead to the fact that a person pays attention to the low naturalness of virtual scenarios
[24]. Purposeful study of different classes of psychological factors of the sense of presence will
lead not only to progress in the construction of a general psychological theory of the sense of
presence, but also will allow us to find effective solutions to fundamental and applied problems
using virtual reality technologies [24].
The presence, regardless of the definition of which school the researcher uses, is a subjective
experience in which a person interacts in one way or another with the reality surrounding him
(for representatives of schools of mediated presence – virtual).
The presence measure is necessary in order to establish whether presence actually increases
interaction with the virtual environment system, and to better understand the factors that
may drive this phenomenon [26]. The question of a valid measure of presence was pointed in
[26]. The measuring of the presence in the framework of the subjective, the behavioral and the
physiological methods was consider since [27], [28].
The subjective method uses questions such as ”how real did the virtual environment seem to
you?”, ”was the virtual environment really the place you were in – or just a series of images
shown to you?”, [27], [28], [26], [29].
The behavioural method tracks the user’s behaviour, for example, whether the user Shield
himself/herself when an object is thrown at his head, [27], [28], [26], [29].
The physiological methods record changes in the heart rate, skin temperature, GSR, respira-
tory rate, etc. The experimenter looks at whether a person is stress in a stress situation in the
virtual environment [27], [28], [26], [29].
In 2000 Slater, together with his colleagues, proposed approach to measuring presence [3].
According to Slater, traditional methods of measuring presence have certain disadvantages. In
particular, the responses to the questionnaire can ensure that only conscious and voluntary
responses are integrated over time. Behavioural measures, such as responding to the sudden
appearance of an object, require events in the environment. The physiological measurements
offer, according to Slater, a promising way forward. However, the physiological methods
are used in cases where the environment causes anxiety (for example, go to a pit). That is
why they are only suitable for a very small number of applications. What the physiological
response is appropriate for observing a virtual chair, a room, or a deck of a ship? Relying on
the physiological responses to specific events or objects in a virtual environment does not offer
a general solution [30].
The main idea of the approach proposed by Slater and his co-authors can be described as
follows. A person who is in the virtual reality environment switches between being present in
the virtual environment and being present in the real environment [30].
Initially, Slater and Steed intended to use the mathematical model of Markov chains to
calculate presence based on BiP. Later [31], the Slater team rejected this method as too complex.
Natural noises was the cause of BiP in operation [3]. The noises was modeled by experimenters
in later works [31].
Let as briefly describe the approach of [3] that provides main idea of method used in this
paper. In the immersive virtual environment, the participant receives a continuous stream of
IMS-2021. International Conference “Internet and Modern Society” 427
sensory data. This data is mostly visual, coming from a virtual environment. But also often
this data is auditory from the real world or real haptic and kinesthetic data (for example, the
weight of the HMD). Sometimes, the sensor data of the virtual environment shows disturbances.
Or a close-up of an object shows its texture mapping. Additionally data from the real world
can intrude: the phone rings, there is a sudden movement of air when the door opens, the
temperature changes, the cable is wrapped around the leg. Sometimes, the participants’ internal
mental processes make them realize that they are actually in the virtual environment when the
participant in the laboratory or exhibition hall, and not in the illusory place presented to them
by the virtual environment [3].
In other words, two alternative gestalts are available to each person at any time. The first
gestalt is ”I am in the place depicted by the VE system” Bellow, we following [3] denote it by
“V”. The second gestalt is ”I’m in the lab in the computer science building, wearing a HMD”.
This state is denoted by “R”. At every moment, the individual can occupy only one state. Being
in the virtual presence in the virtual environment can be considered as the degree of preference
of the state V [3].
However, during the experience of being in the virtual environment, as expected, the individ-
ual usually experiences transitions between state V and state R. These time instants when the
individual switches from one interpretation to another, in particular, from V to R, are studied. It
is not possible to ask participants to report transitions from R to V because this will require them
to immediately exit the presence state. However, according to the authors [3] the experimenter
can ask participants to report transitions from V to R. Since the transition from V to R implies
the feeling of the presence, we can use this information to evaluate the being of presence.
On the other hand, the information of transition from R to V is unavailable. Thus, Slater
proposed to consider two cases: low and high presence. The low presence case implies that the
state “R” is basic, when the high presence case means that primary individual occupies the state
“V”.
In their work, Slater and Steed provide the mathematical calculations that allow using the
technique of Markov chains to derive the formula for counting the experienced presence
depending on the reported BiPs. As a consequence of this formula, the probability that the
participant experienced the presence is computed. The value of this probability serves as a
presence counter. This formula is modified for conditions involving high presence and for
conditions involving low presence [3].
Slater and Steed used the discrete time Marcov chain assuming the time unit equal to 10
seconds. Participants reported BiP orally.
In this framework, the low presence condition means that the transition from R to V occurs
just before BiP. Whilst, the high presence is modeled by immidiate recovering of V after BiP.
To distinguish the low and high presence cases Slater and Steed used discriminator question.
3. Methods
3.1. Participants
The study involved 22 people, including 11 women and 11 men. With the exception of two
people (men and women), all participants had or pursued the university degree.
428 PART 2: Internet Psychology
Figure 1: General view of a virtual scene.
3.2. Equipment: hardware and software
The environment uses the vr HTC Vive headset. It connects with the computer via cables. Apart
from a VR headset, the Vive system is equipped with special hand-operated controllers and with
two infrared cameras for tracking a person in the environment. In order to implement both
systems, the Unity 3D development environment was used, along with the C# programming
language, a SteamVR plugin etc.
The most elaborated and suitable for the experiment set of visual resources was chosen
at the design stage in order to create the effect of presence and immersion into the virtual
environment.
Visually, the scene is a reconstruction of a classical Far-Eastern mountain monastery (see
Figure 1).
3.3. Procedure
The participants were asked to provide their name, offered to wear HMD, and, having seen
the virtual environment with them, follow the opened area, which is an Asian-style park. The
movement was carried out by means of physical movement in a limited space and by means of
successive teleportations over long distances. In the park, the participants had to fix specially
selected objects along the way – low rounded vases (see Figure 2). The objects were chosen
according to the principle of average notability and, at the same time, naturalness for any
location of the created scene. The task was to give meaning to the walk in the park, make it
purposeful and encourage the participants to move around the space of the park.
The stay in the environment was limited by 7 minutes. After the first five minutes, the
participants could leave the environment at any time they wanted. The participants were
instructed to say the word ”here” loudly every time they felt BiP. During the experiment, an
IMS-2021. International Conference “Internet and Modern Society” 429
Figure 2: Detecting a vase and indicating it with a laser pointer .
audio recording was conducted. The time at which the subjects uttered the word ”here” was
recorded on an audio recording.
During the experiment, the number of vases found was also recorded. The participant could
see them on the counter on his hand (see Figure 2). A large or small number of vases found
were not rewarded or punished in any way.
As in the original paper by Slater and Steed the discriminator question about the experience
of the presence was asked after the experiment. The participants were asked where they felt
during the session. Also, the reasons for pronouncing the signal ”here” were specified. If the
participant never said ”here!”, then the question ”why?” was asked. Did the participant always
feel like they were here, in a real room? Did the participant feel like they were in an Asian park
all the time?
The Russian version of the ITC-Sense of Presence Inventory (ITC-SOPI) was offered after the
experiment. It is described in the paper [4]. This inventory is based on four factors:
• Sense of Physical Space;
• Engagement;
• Ecological Validity;
• Negative Effects.
These four factors cannot be combined into a single measure of presence. According to [4],
the main determinants of the Sense of Physical Space factor are the variables of the media
form, i.g., the properties of the virtual environment. The questions on this scale relate to
the participants’ experiences about the reality of what is happening, the ability to touch the
elements of the environment, to interact with them, ect. The second factor that is considered is
Engagement. One of the points of this factor directly examines how attractive respondents find
430 PART 2: Internet Psychology
the content. Other questions that make up this scale relate to excitement and emotionality. The
answers depend on the media content, but are also amplified by the media form.
The questions that include the third factor, Ecological Validity, concern the plausibility and
realism of the content, as well as the naturalness of the environment. The amount, degree, and
sequence of sensory stimulation, according to the authors, improves perceived naturalness and,
in turn, increases scores on this scale. The high immersiveness of the media form (i.e., how it
allows the participants to experience immersion), gives an increase of Ecological Validity. The
higher imeressiveness is the less the impact of the content on the perception of authenticity is.
The difference in photorealism also leads to differences in the scale of Ecological Validity.
The questions related to the fourth factor, negative effects, are less related to the first three
factors than to each other. In [4] Negative Effects were not strongly correlated (positively or
negatively) with Engagement or Ecological Validity. However, they had a low but significant
positive correlation with the Sense of Physical Space. Some negative effects, such as headache,
eye strain, fatigue may be associated with the media form. Further, content can affect on the
Negative effects: if it is perceived as boring, the participants may give appropriate ratings of
fatigue or even headache [4].
3.4. Presence counter based on BiP
The presence counter was calculated based on the probability formula of experiencing presence,
which was derived using Markov chains. The formulas given in [3] were applied. Time was
discrete in the work [3]. The length of the time interval was 10 seconds. Also, the presence
counter at the interval length of 5 seconds was calculated to increase the sensitivity of the
method. In addition, formulas for continuous time were derived.
As in paper [3] we study the stationary distribution. Notice that [32] the stationary distri-
bution exists and moreover the Markov chain converges to it exponentially. The approach
proposed in (Slater, Steed, 2000) implies that we are to solve the inverse problem: we compute
the stationary distribution using the information on breaks in presence, and then we are to
find the matrix of transition probabilities that provides the measure of presence. Following
[3] we consider two cases. First is the low presence case. Roughly speaking, it means that the
participant leaves the state of presence once he/she reaches it. The second case we examine is
the high presence situation. It implies that the state no-presence is leaved once the participant
reaches it.
𝑝𝐿 (𝑏) is the probability of presence, corresponding to the stationary distribution, for conditions
of low presence at discrete time [3]:
𝑏−𝑘
𝑝𝐿 (𝑏) =
𝑛−1
where
𝑏 is the number of BiPs,
𝑛 is the number of time intervals.
𝑘 is the number of ”close BiPs”, that is, BiPs in adjacent time intervals.
𝑝𝐻 (𝑏) is the probability of presence for high presence conditions at discrete time [3]:
IMS-2021. International Conference “Internet and Modern Society” 431
𝑛−1−𝑏
𝑝𝐻 (𝑏) =
𝑛−1
where
𝑏 is the number of BiPs,
𝑛 is the number of time intervals.
Furthermore, we extend the approach of [3] assuming the continuous time. In this case, the
dynamics of probabilities is determined by so called Kolmogorov equation that is an ordinary
differential equation [32]. As above, we consider the inverse problem. We compute the matrix
of transition rates by using the stationary distribution computed by BiPs. We also follow [3]
and examine two cases: low presence and high-presence. However, the continuous time setting,
implies that we are to introduce the relaxation time that is an external parameter. For the low
presence case, it is assumed that the participant occupies the presence state once it is reached
for the averaged time interval of the length 𝜇 −1 . The case of high presence is opposite. Here
we assume that the participant will reach the presence state once he/she leaves it for the time
interval of the averaged length equal to 𝜇 −1 . To adjust the continuous and discrete time models,
one is to let 𝜇 −1 equal to the time unit for the original discrete time model proposed in [3].
𝑝𝐿 𝑉 is the probability of presence for low-presence conditions at continuous time:
𝑏
𝑝𝐿 𝑉 =
𝑇 ∗𝜇
where
𝑏 is the the number of BiPs,
𝑇 is the total time of the virtual reality session,
𝜇 = 0.1 with a relaxation time taken equal to 10 seconds.
𝜇 = 0.05 with a relaxation time taken equal to 5 seconds.
𝑝𝐻 𝑉 is the probability of presence for conditions of high presence at continuous time:
𝜇
𝑝𝐻 𝑉 =
𝜇 + 𝑏−𝑘
𝑇
where
𝑏 is the the number of BiPs,
𝑇 is the total time of the virtual reality session,
𝜇 = 0.1 with a relaxation time taken equal to 10 seconds.
𝜇 = 0.05 with a relaxation time taken equal to 5 seconds.
3.5. Mathematical and statistical methods
In this paper, the mathematical model of Markov chains is used.
In this paper, we use the Pearson correlation coefficient for
• calculation of the relationship between the scales of the ITC-SOPI questionnaire based
on the responses of the subjects to the Russian version of the ITC-SOPI;
432 PART 2: Internet Psychology
• calculation of the relationship between the ITC-SOPI scales and different variants of
calculating the probability of presence based on BiP;
• calculation of the relationship between the presence measures and the effectiveness of
the search for vases, which is expressed in the number of vases found.
4. Results and discussions
The results of the ITC-SOPI questionnaire were calculated for all participants of the experiment.
A BiP-based presence counter was calculated for each participant. The probability of presence is
calculated when calculating the presence counter. If the participant gave a definite answer that
he/she felt a presence most the time, then the formula for high presence condition was applied.
If the participant gave a definite answer that he/she felt no presence or felt little presence, then
the formula for low presence condition was applied. If the participant did not give a definite
answer, then both formulas were applied.
Presence counter was calculated for the group of participants in the experiment 8 times
according to the following principles:
The presence counter was calculated under the following frameworkes
• Time interval/relaxation time is equal 5 sec or 10 sec;
• Discrete or continuous time Markov chain is used;
• The high or low presence assumption was applied to the undecided participants
And always the high presence assumption was applied to the participants reported a high
presence. Always the low presence assumption was applied to the participants reported low
presence or no presence.
This gives the 8 ways of calculation of presence counter
It is nessecery to compare the presence counter with the results of the ITC-SOPI to determine
the validity of the presence counter.
The ITC-SOPI results were also calculated. The Pearson’s correlation analysis was performed
between the scales of the questionnaire. See Table 1.
Table 1: Correlations between the ITC-SOPI scales, 𝑟𝑐𝑟𝑖𝑡 = 0.42 (𝑝 ≤ 0.05) for the first level of
significance, 𝑟𝑐𝑟𝑖𝑡 = 0.54 (𝑝 ≤ 0.01) for the second level of significance
Sense of Physical Space Engagement Ecological Validity Negative Effects
Sense of – – – –
Physical
Space
Engage- 0.70 – – –
ment
Ecological 0.77 0.61 – –
Validity
Negative -0.02 -0.04 0.13 –
Effects
IMS-2021. International Conference “Internet and Modern Society” 433
Table 1 shows that the first three factors of ITC-SOPI significantly correlate with each other
for the second level of significance. This corresponds to the results described in [4].
The Pearson’s correlation analysis was conducted between the results of the inventory and
the probabilities of experiencing presence, see Table 2.
Table 2: Correlations between the results of the questionnaire and the probability of presence
based BiP, 𝑟𝑐𝑟𝑖𝑡 = 0.42 (𝑝 ≤ 0.05) for the first level of significance, 𝑟𝑐𝑟𝑖𝑡 = 0.54 (𝑝 ≤ 0.01)
for the second level of significance
Probabilities of Sense of Physi- Engagement Ecological Va- Negative
presence cal Space lidity Effects
The length of the time interval/relaxation time is 10 seconds
𝑝𝐿 (𝑏) for uncer- 0.509 0.318 0.511 0.063
tain and low
presence;
𝑝𝐻 (𝑏) for high
presence
𝑝𝐻 (𝑏) for uncer- 0.487 0.392 0.482 0.333
tain and high
presence;
𝑝𝐿 (𝑏) for low
presence
𝑝𝐿 𝑉 for uncer- 0.554 0.336 0.541 0.043
tain and low
presence;
𝑝𝐻 𝑉 for high
presence
𝑝𝐻 𝑉 for uncer- 0.551 0.416 0.520 0.241
tain and high
presence;
𝑝𝐿 𝑉 for low
presence
The length of the time interval/relaxation time is 5 seconds
𝑝𝐿 (𝑏) for uncer- 0.577 0.350 0.541 -0.015
tain and low
presence;
𝑝𝐻 (𝑏) for high
presence
𝑝𝐻 (𝑏) for uncer- 0.349 0.348 0.403 0.355
tain and high
presence;
𝑝𝐿 (𝑏) for low
presence
434 PART 2: Internet Psychology
𝑝𝐿 𝑉 for uncer- 0.657 0.378 0.592 -0.051
tain and low
presence;
𝑝𝐻 𝑉 for high
presence
𝑝𝐻 𝑉 for uncer- 0.538 0.428 0.523 0.355
tain and high
presence;
𝑝𝐿 𝑉 for low
presence
Table 2 shows that the probabilities of presence significantly and positively correlates with
the two scales of the questionnaire. These scales are Spatial Presence and Ecological Validity.
The highest correlation values, significant for the second level of significance, were obtained
when comparing the first and third ITC-SOPI scales with the probabilities of presence, calculated
using a formula designed for low presence condition for the participants who did not give a
clear answer or gave an answer about low presence, and using a high presence formula for
the participants who gave a certain answer about high presence, for continuous time with a
relaxation time taken equal to 5 seconds. This shows that this formula allows to create the most
sensitive presence counter. However, when calculating the probabilities of presence using a
formula designed for high presence conditions for the participants who did not give a clear
answer and for the participants who reported high presence, and using a low presence formula
for the participants who reported low presence or did not experience it, for continuous time
with a relaxation time taken equal to 5 seconds, a correlation of the probability of presence with
the Engagement scale was obtained.
Since, see Table 1, the first three ITC-SOPI scales correlate with each other, this result seems
to be natural.
The relationship between the presence counter, which represents the probability of presence,
and the ”Sense of Physical Space” scale seems obvious. The presence counter is based on BiP and
the participants were instructed to report BiP every time they felt like they were in a real room,
not in an Asian park. Therefore, it is obvious that the presence counter should correlate with
the scale associated with the participants’ experiences about the reality of what is happening,
the ability to touch the elements of the environment, to interact with them.
The focus on the environment, as measured by the second scale, ”Engagement”, can also affect
whether a participant feels like they are in a virtual environment or in a real room. However,
this issue needs further clarification.
Apparently, the naturalness and plausibility of the environment, as measured by the ”Ecolog-
ical Validity” scale, also influenced how the participant felt in the virtual environment.
It should be understood that the application of the formula of high presence conditions for
undecided participants does not in itself give more valid results. The greater validity of the
results, in which the probability of presence in undecided participants is calculated using the
low-presence condition formula, is due to the fact that this formula seems to reflect the presence
they experienced more than the high-presence condition formula. The correlation between the
IMS-2021. International Conference “Internet and Modern Society” 435
two indicators suggests that one will change in the same way as the second. In our case, the
probability of presence calculated using BiP is expected to increase or decrease in the same
way as the ITC-SOPI scales. The method of calculating the probability of presence has this
property to the greatest extent, when the formula of high presence conditions is used for those
participants who indicated a high presence, and the formula of low presence conditions is used
for those who indicated a low presence or no presence and for those who were undecided. But
of course we get the significative correlation between ITC-SOPI scale “Engagement” and the
presence counter using the formula of high presence conditions for those participants who
indicated a high presence and for those who were undecided, and the formula of low presence
conditions for those who indicated a low presence or no presence.
To clarify the optimal formula, additional studies will be required on an extended sample
group and with a more strict discriminant question.
The number of vases found by the participants was also calculated, but it was not related to
either the results of the presence counter or the results of the ITC-SOPI, see Table 3.
Table 3: Results of calculating the correlation between the number of vases found and marked
by the subject during the experiment and the presence indicators, 𝑟𝑐𝑟𝑖𝑡 = 0.42 (𝑝 ≤ 0.05)
for the first level of significance, 𝑟𝑐𝑟𝑖𝑡 = 0.54 (𝑝 ≤ 0.01) for the second level of significance
Presence indicators Correlation between the number of found
vases and the presence indicator
The length of the time interval/relaxation time is 10 seconds
𝑝𝐿 (𝑏) for uncertain and low presence; 0.143
𝑝𝐻 (𝑏) for high presence
𝑝𝐻 (𝑏) for uncertain and high presence; -0.271
𝑝𝐿 (𝑏) for low presence
𝑝𝐿 𝑉 for uncertain and low presence; 0.163
𝑝𝐻 𝑉 for high presence
𝑝𝐻 𝑉 for uncertain and high presence; -0.248
𝑝𝐿 𝑉 for low presence
The length of the time interval/relaxation time is 5 seconds
𝑝𝐿 (𝑏) for uncertain and low presence; 0.124
𝑝𝐻 (𝑏) for high presence
𝑝𝐻 (𝑏) for uncertain and high presence; -0.285
𝑝𝐿 (𝑏) for low presence
𝑝𝐿 𝑉 for uncertain and low presence; 0.168
𝑝𝐻 𝑉 for high presence
𝑝𝐻 𝑉 for uncertain and high presence; -0.245
𝑝𝐿 𝑉 for low presence
Results of ITC-SOPI
Sense of Physical Space 0.242
Engagement 0.002
Ecological Validity 0.164
436 PART 2: Internet Psychology
Negative Effects -0.214
The effectiveness of vases search is not related to the presence, as shown in table 3. It can be
assumed that the presence does not affect the performance task related with the objects search
and the objects search does not affect the presence.
5. Conclusion
In this study, we considered the measuring presence of the using primarily information available
during the session of interacting with the virtual reality that does not require an emotionally
rich content of the environment. Such a measure is a presence counter based on the calculation
of the probability of presence using the Markov chain model. During the session, the participant
reports the BiP, while continuing to perform the task. In this experiment, we used an environ-
ment that was an Asian park, where the participants could move around using teleportation
and find vases. The experiment showed that the number of vases found during the session is
not related to the presence measured during the experiment.
The paper shows that the presence counter based on BiP is a valid measure of presence
for HMD technology and environment containing moving by walking for small distances and
teleportation for significant ones. The presence counter provides the same measure of presence
as the ITC-SOPI, which showed the high level of internal consistency in the Russian-language
sample group. The formula based on continuous time seems to be more productive, especially
since time is actually a continuous quantity, not a discrete one. The obtained results can be
improved when we perform the study with largest group and clarify the discriminator question.
Thus, the results show that calculating the probabilities of presence using a formula designed
for high presence conditions for continuous time Markov chain with the relaxation time taken
equal to 5 seconds is optimal and allows to create the most sensitive presence counter. However,
the experimental procedure allows for uncertainty in the subjects’ responses to the discriminator
question whether they felt a presence. Therefore, it was necessary to use both the formula for
high-presence conditions and the formula for low-presence conditions for undecided participants.
It will be profitable to find a more strict discriminator question that does not give uncertainty.
Moreover future works include the improvement of instructions. The fact that the presence
counter calculated on the basis of BiP has a significant correlation with the three of ITC-SOPI
scales, that is, not only with “Sense of Physical Space”, but also with “Ecological Validity” and
with “Engagement”, looks very interesting. Obviously, the presence counter measures how
much the participant feels in the virtual environment and also how much they perceive the
environment as plausible and natural. The presence counter also measures how much the
participant is involved in the events of the environment.
Future research should also include virtual experiences related to different environments,
participants’ capabilities, and tasks. When the procedure for applying the presence counter
is clarified, it will be possible to raise the question of the relationship between the presence
deducted using the BiP-based presence counter and various individual characteristics of partici-
pants, such as gender of respondents, age, degree of familiarity with virtual reality, attitudes
associated with participation in the experiment and so on.
IMS-2021. International Conference “Internet and Modern Society” 437
References
[1] Y. P. Zinchenko, Virtual reality technologies in the system of post-non-classical psychology,
The world of psychology (2013) 31–42.
[2] J. A. Waterworth, E. L. Waterworth, F. Mantovani, G. Riva, On feeling (the) present: An
evolutionary account of the sense of presence in physical and electronically-mediated
environments, Journal of Consciousness Studies (2010) 167–188.
[3] M. Slater, A. A. Steed, Virtual presence counter, Presence: Teleoperators and Virtual
Environments 9 (2000) 413–434.
[4] J. Lessiter, J. Freeman, E. Keogh, J. Davidoff, A cross-media presence questionnaire: the
itc-sense of presence inventory, Presence: Teleoperators Virtual Environ (2001) 282–297.
[5] J. van Baren, W. IJsselsteijn, Measuring presence : A guide to current measurement
approaches, in: Deliverable of the OmniPres project IST-2001-39237, 2004.
[6] T. Hartmann, W. Wirth, P. Vorderer, C. Klimmt, H. Schramm, B. S., Spatial presence
theory: State of the art and challenges ahead, in: M. Lombard, F. Biocca, J. Freeman,
W. Ijsselsteijn, R. Schaevitz (Eds.), Immersed in Media: Telepresence Theory, Measurement
and Technology, Springer, London, 2015, pp. 115–135.
[7] J. Laarni, N. Ravaja, T. Saari, S. Böcking, T. Hartmann, H. Schramm, Ways to measure
spatial presence: Review and future directions, in: M. Lombard, F. Biocca, W. Ijsselsteijn,
R. Freeman, J. Schaevitz (Eds.), Immersed in Media: Telepresence Theory, Measurement
and Technology, Springer, London, 2015, pp. 139–185.
[8] R. Skarbez, J. F. Brooks, M. Whitton, A survey of presence and related concepts, ACM
Computing Surveys 50 (2017) 1–39.
[9] M. Lombard, M. T. Jones, Defining presence, in: M. Lombard, F. Biocca, W. Ijsselsteijn,
R. Freeman, J. Schaevitz (Eds.), Immersed in Media: Telepresence Theory, Measurement
and Technology, Springer, London, 2015, pp. 13–34.
[10] S. M., Place illusion and plausibility can lead to realistic behaviour in immersive virtual
environments, Philosophical Transaction of Royal Society B: Biological Sciences (2009)
3549–3557.
[11] N. A. Sonnenfeld, M. Meyers, J. P. Kring, Presence in transfer: The holistic perspec-
tive model, 2016. URL: https://www.researchgate.net/publication/308607726_Presence_in_
Transfer_The_Holistic_Perspective_Model.
[12] S. Triberti, G. Riva, Being present in action: a theoretical model about the “interlocking”
between intentions and environmental affordances, Frontiers in Psychology 2052 (2016)
21–28.
[13] G. Riva, Enacting interactivity: the role of presence, in: Enacting intersubjectivity: a
cognitive and social perspective on the study of interactions, Ios Press, Amsterdam, 2008,
pp. 97–114.
[14] J. A. Waterworth, E. L. . Waterworth, Presence in the future, in: Proceedings of the
11th Annual International Workshop on Presence. Padova, CLEUP Cooperativa Libraria
Universitaria Padova, Padova, 2008, pp. 61–65.
[15] G. Riva, J. A. Waterworth, E. L. Waterworth, F. Mantovani, From intention to action: the
role of presence, New Ideas in Psychology (2011) 24–37.
[16] G. Riva, F. Mantovani, From the body to the tools and back: a general framework for
438 PART 2: Internet Psychology
presence in mediated interactions, Interacting with Computers (2012) 203–210.
[17] G. Riva, J. A. Waterworth, Being present in a virtual world, in: M. Grimshaw (Ed.), The
Oxford Handbook of Virtuality, Oxford University Press, Oxford, 2014.
[18] J. Waterworth, G. Riva, Feeling Present in the Physical World and in Computer-Mediated
Environments, Palgrave Macmillan, London, 2014.
[19] G. Riva, F. Mantovani, E. L. Waterworth, J. A. Waterworth, Intention, action, self and
other: An evolutionary model of presence, in: M. Lombard, F. Biocca, W. Ijsselsteijn,
R. Freeman, J. Schaevitz (Eds.), Immersed in Media: Telepresence Theory, Measurement
and Technology, Springer, London, 2015, pp. 73–99.
[20] J. A. Waterworth, E. L. Waterworth, G. Riva, F. Mantovani, Presence: Form, content
and consciousness, in: M. Lombard, F. Biocca, W. Ijsselsteijn, R. Freeman, J. Schaevitz
(Eds.), Immersed in Media: Telepresence Theory, Measurement and Technology, Springer,
London, 2020, pp. 35–58.
[21] C. Redaelli, G. Riva, Flow for presence questionnaire, in: L. Canetta, C. Redaelli, M. Flo-
res (Eds.), Digital Factory for Human-oriented Production Systems. The Integration of
International Research Projects, Springer, London, 2011, pp. 3–22.
[22] B. E. Riecke, J. Schulte-Pelkum, An integrative approach to presence and self-motion
perception research, in: M. Lombard, F. Biocca, W. Ijsselsteijn, R. Freeman, J. Schaevitz
(Eds.), Immersed in Media: Telepresence Theory, Measurement and Technology, Springer,
London, 2015, pp. 187–235.
[23] B. Velichkovsky, Psychological factors of the emerging sense of presence in virtual
environments, National psychological journal 15 (2014) 31–38.
[24] B. Velichkovsky, Error monitoring and correction related to the sense of presence in virtual
environments, Moscow State University Bulletin. Series 14. Psychology (2016) 25–33.
[25] B. Velichkovsky, A. Gusev, V. Vinogradova, O. Arbekova, Cognitive control influences the
sense of presence in virtual environments with different immersion levels, Experimental
Psychology 9 (2016) 5–20.
[26] W. Sadowski, K. Stanney, Measuring and managing presence in virtual environments,
in: N. Mahwah (Ed.), Handbook of virtual environments: Design, Implementation, and
Applications, Lawrence Erlbaum Associates, 2002, pp. 791–806.
[27] B. E. Insko, Measuring presence: Subjective, behavioral and physiological methods, in: Be-
ing There: Concepts, Effects and Measurement of User Presence in Synthetic Environments,
Ios Press, Amsterdam, The Netherlands, 2003, pp. 109–119.
[28] D. Mestre, Immersion and presence, Preprint, 2005. URL: http://www.ism.univmed.fr/
mestre/projects/virtual%20reality/Pres_2005.pdf.
[29] P. Gamito, J. Oliveira, P. Santos, D. Morais, T. Saraiva, M. Pombal, B. Mota, Presence,
immersion and cybersickness assessment through a test anxiety virtual environment,
Annual Review of CyberTherapy & Telemedicine (ARCTT) 6 (2008) 83–90.
[30] M. Slater, Presence and the sixth senses, Presence 11 (2002) 435–439.
[31] A. Brogni, M. Slater, A. Steed, More breaks less presence, in: Presence: The 11th Annual
International Workshop on Presence, 2003, pp. 1–4.
[32] L. Koralov, Y. G. Sinai, Theory of Probability and Random Processes, Springer-Verlag Berlin
Heidelberg, 2007.