IEEE Consumer Electronics Magazine - July 2018 - 46

Our review of related literature revealed a lack of studies that specifically addresses the level of social presence
in the context of human-to-human interaction within
immersive VR environments. Riva et al. examined the VE's
affective impact; however, in their experiment, participants
did not encounter other characters, avatars, or agents during the VR experience. Other studies have focused on textbased mediums [15], [16] online, nonimmersive virtual
worlds [12], or the level of social presence based on the
agent's degree of anthropomorphism [17]. We assert, however, that VR's potential as an affective social medium of
communication that facilitates social interaction requires
an extensive disquisition.

social context and expeRimental objectives
In our day-to-day lives, social context is defined by a vast
number of factors, cues, indicators, and so on, which all individuals are expected to detect, abide by, and, therefore, engage
in socially and culturally appropriate behavior. In accordance
with Bourdieu's habitus, "...it acts within them as the organizing principle of their actions, ... this modus operandi informing all thought and action (including thought of action) reveals
itself only in the opus operatum" [18].
Since the first component of any interaction is the social setting, we argue that social presence may be affected by the social
context of a VE. Social context is based on, among other things,
design features, script-related tasks, and the presence of other
characters, thus affecting the subsequent social interaction that
is taking place within the VE. In other words, individuals
acknowledge behavior-making social conditions derived from
the design and script-related characteristics of the immersive
VE. Therefore, since the basis for our experiment was to generate social context within an immersive VE, we introduced a
variety of social components.
▼ The 3-D environment, designed for the purposes of this
study, was labeled as a clothing store; to communicate it as
such, a few signifiers were placed, e.g., shelves with folded
clothes, a cash register, and a curtain that created a dressing room.
▼ The use of mirrors and a dressing room are also social
components in the sense that there is no cause for participants to witness or confine the process of changing clothes
while in a VE.
▼ The additional character was defined as a salesman (rather
than another customer) to further validate the clothing
store component. The presence of a second character triggers the potential for social interaction and elicits the feeling of being watched, ultimately introducing a witness to
the socially inappropriate occurrence of public nudity.
Based on these components, participants were expected to
draw assumptions that would be in accordance with their
empirical understanding of what constitutes a typical clothing
store. Thus, a clothing store is a public space wherein the
presence of a salesperson is expected, and most importantly,
undressing while in this space is neither uncommon nor
frowned upon; in fact, a change of clothes is often necessary,
46 IEEE Consumer Electronics Magazine

July 2018

even though it still needs to be visually restricted from all of
the other individuals present. All typical clothing stores,
therefore, are expected to facilitate (or provide) at least one
dressing room.
Our experiment's primary goal was to examine whether
participants would engage in specific behavior due to the
social context. The naked VB was intentionally introduced as
an abnormal social element and an unexpected, script-related
occurrence. The intent of this abnormality was to challenge
participants' level of comfort and cause a state of alertness,
but mostly it was meant to confirm that they did, in fact,
acknowledge a social context in which full nudity and/or
public nudity was inappropriate.
Moreover, in a manner similar to everyday social interactions, individuals feel more compelled to abide by a social
norm when others are present or engaged in the interaction.
That is why the variance between the three experimental conditions was based on whether another character was present
and to what extent that character was realistic (i.e., NPC versus real actor).
Participants' reactions were closely observed and analyzed in
relation to the VR research-related concepts presented in the
"Related Research" section. The following hypotheses were
drawn with respect to the social context's impact on participants':
1) behavior and emotional state
2) level of presence and BOI.
Mirror choice, and the subsequent justification of that
choice, was regarded as a behavioral determinant since it preceded the VB's exposure, of which, participants had no prior
knowledge. Additionally, during the moment of the VB's
exposure, the participants' emotional state was recorded and
reviewed per each condition. Finally, the levels of presence
and BOI were expected to be higher when a salesman was
present, whereas the unexpected occurrence of the naked body
was expected to fluctuate the level of presence.

naked viRtual body expeRiment
ParticiPants
Our VR experiment was advertised on the campus of the University of the Aegean, Greece. No prerequisites were set, and
no script-related specifics were made public. The final sample comprised 54 volunteers (31 females), mostly undergraduate students between the ages of 19 and 39. The vast
majority of subjects had no previous experience with VR
technology, nor any social or professional association with
the research team or the laboratory where the experiment was
conducted. The subjects did not receive any compensation for
their participation.

ExPErimEntal DEsign
All of the 54 participants visited a virtual clothing store to
collect a preordered outfit. Eighteen of the participants were
arbitrarily assigned to group A (which included 11 females),
19 were placed in group B (11 females), and 17 were assigned
to group C (nine females).

Table of Contents for the Digital Edition of IEEE Consumer Electronics Magazine - July 2018