Emotional Qualities of VR Space [605647]

Emotional Qualities of VR Space

Asma Naz1 Regis Kopper2 Ryan P. McMahan3 Mihai Nadin4
University of Texas at Dallas , USA Duke University , USA University of Texas at Dallas , USA University of Texas at Dallas , USA

ABSTRACT
The emotional response a person has to a living space is
predominantl y affected by light, color and texture as space-making
elements . In order to verify whether this phenomenon could be
replicated in a simulated environment, we conducted a user study
in a six -sided projected immersive display that utilized equivalent
design attributes of brightness, color and texture in order to assess
to which ext ent the emotional response in a simulated environment
is affected by the same parameters affecting real environments.
Since emotional response depends upon the context, we evaluated
the emotional responses of two groups of users : inactive (passive)
and active ( performing a typical daily activity ). The results from
the perceptual study generated data from which design principles
for a virtual living space are articulated. Such a space, as an
alternative to expensive built dwellings, could potentially support
new, minimalist lifestyles of occupants, defined as the neo-nomads ,
aligned with their work experience in the digital domain through
the generation of emotional experiences of spaces. Data from the
experiments confirmed the hypothesis that percei vable emo tional
aspects of real -world spaces could be successfully generated
through simulation of design attributes in the virtual space. The
subjective response to the virtual space was consistent with
corresponding responses from r eal-world color and brightness
emotional perception. Our data could serve the virtual reality (VR)
community in its attempt to conceive of further applications of
virtual spaces for well -defined activities.
Keywords : Architectural design, affective space, neo-nomad s,
aesthetics .
Index Terms : I.3.7 [Computer Graphics]: Three -Dimensional
Graphics and Realism —Virtual Reality
1. INTRODUCTION
This paper reports on an experimental study of virtual
environments (VEs) explored as an architectural tool to evaluate a
new design concept of a real -world dwelling for information -age
professionals, also known as neo-nomads . The severe housing
crisis of the Silicon Valley in recent years has led architects and
interior designers to search for new concepts of compact living to
accommodate the transitional lifestyles of these technology –
dependent, mobile professionals [1 9]. The dwelling envisioned in
this p aper consists of a single living space with variable design
parameters of color, light and texture used to generate multiple
emotional experiences of spaces that could potentially support daily
living of occupants . In a user study, a six -sided immersive Cave
Automatic Virtual Environment (CAVE) -type display was used to
simulate the living space that adapted equivalent design parameters of color, brightness and te xture to create a set of emotional
experiences of spaces. The study measured perceived emotional
dimensions of virtual spaces in nine categories: warmth, coolness,
excitement, calmness, intimacy, spaciousness and comfort, as well
as spatial preferences for two activities : work and rest. Since
emotional responses to space depend on context, the influence of
activity on perception was also assessed by evaluating responses
from two groups of users : an active group that performed a typical
daily activity and an inactive (passive) group.
The study presented a hypothesis that the real -world
phenomenon of affective space creation of traditional architecture
could be replicated in a simulated environment through
modification of design parameters of color, brightne ss and texture.
The study goal was to assess the extent to which design parameters
influenced emotional responses to virtual spaces. Subjective
responses in the VE were analyzed to find correlation wi th
corresponding normative real -world emotional space pe rception
related to color, brightness and texture.
The outcome of the study is in the form of data relevant for
design of a real -world dwelling that has the capacity to generate
variable spaces with affective dimensions in order to support
minimalist lifestyles for young professionals. The major
contribution of this study lies in its definition of virtual reality (VR)
as a viable evaluation platform for new architectural concepts
where variable design parameters can be adapted to cr eate
experiential spaces with emotional or psychophysiological
quali ties. Such data could have real -world implications in decision –
making processes for architects, builders, interior designers and
clients, as well as facilitate better communication in term s of time
and resources.
2. RELATED WORK
In this section , we review prior literature on VR in the context of
architecture and affective dimensions, and also provide an
overview of affective design parameters for living spaces.
2.1 VR as Architectural Design Evaluation Tool
Virtual Reality technology has been used as an evaluation tool in
architecture, mainly to assist in decision -making during design
process to explore unbuilt concepts and difficult -to-realize projects,
as well as communal, educational or rec reational designs that need
participatory inputs from architects, designers, clients and users in
a collaborative setting [ 7]. Research on architectural applications
of VR was initially envisioned by Sutherland [ 24] as virtual reality
consisting of realist ic living spaces. The first head -mounted display
(HMD), design ed by Sutherland, had an architectural space as one
of the first immersive VEs [ 24] followed later by the first room –
sized VR —CAVE [ 23], with four 10 -foot projection screens and
interactive user interfaces. In CAVE, a prototyping system called
CALVIN introduced multiple perspectives for architectural design
at both ground and global (above -ground) level s [16].
In VR architectural studies, walk -throughs allow users to travel
freely in the VE, and prototyping systems provide user-centric
designs during early development phases [ 7]. Two design studies
during the mid -1990s conducted walk -through studies to evaluate
architectural spaces for unnecessary design element s [17]. In
another study conducted at the University of Washington, working 1 asma.naz@utdallas.edu
2 regis.kopper @duke.edu
3 rymcmaha@utdallas.edu
4 nadin@utdallas.edu

and testing design methods in VR improved design quality of
students [1 7]. In the late 19 90s, Fröst and Warren used a more
sophisticated, immersive CAVE system with multiple participants
and real -time image generation to compare traditional paper and
pencil design sessions with VR technology for the initial
conceptual and schematic design proces s in architecture [12].
Results showed that VR proved to be a better tool to conceptualize,
analy ze, test an d construct architectural designs. Over the years, the
benefits of VR have been acknowledged in the architectural design
process as an engaging, interactive system with a high degree of
realism [1 7].
2.2 VR Studies in Measuring Affective Dimensions
Assessment of spatial quality of architecture in VR has involved
the measurement of both qualitative and quantitative factors of
architectural space design. Quantitative measures are mostly
related to size, proportions, scale and distance perceptions, wher eas
qualitative factors are affective and environmental aspects of space.
To quantify emotional and environmental preferences, some basic
models were established in environmental psychology [ 9]. These
models used semantic differential scales (non -numeric al scales) for
affective appraisals of space constructed with pairs of oppositional
adjectives as descriptors of environmental aspects of mood [ 14].
The adjectives are part of everyday language that convey both
emotional and physical meanings to characteri ze perceived
architectural space dimensions. These adjectives relate to size,
openings, temperature, color brightness, style, form and structural
details of architectural spaces, such as: pleasant and unpleasant,
bright and dull, arousing and calm, narrow and spacious, dark and
bright, open and enclosed, and warm and cold [ 10].
Some studies conducted in VR have compared qualitative virtual
perception to corresponding real -world perception in order to find
out how similar these spaces feel [1 5]. These studi es established
quantitative relationship s between perceived emotional aspects and
simulated volume of space : scale, lighting, architectural details and
other artifacts [1 5].
Attempts have been made to incorporate VR as part of
architectural design curric ula for design scheming, development,
testing and review phases. One such example was the College of
Architecture and Planning (CAP) at Ball State University , which
used HMD -based VR to assist students understand spaces in terms
of everyday semantics with affective dimensions and manipulate
space -making features to create spaces with meaning [7]. Students
reproduced real -world functional spaces in the VE to compare how
they are similar in negative or positive appreciations [ 7]. Certain
studies show that , similar to generic pictorial features, spatial
features also elicit emotional responses [ 9]. Previous studies have
used semantic scales to quantify and establish correlations between
everyday language that describes perceived experiential qualities
and cert ain spatial features of interior rooms : size, shape,
dimension, color and saturation [ 11]. Another VR -based perceptual
study with the Elumens Vision Station used a series of still images
of interiors to establish relationship between physical openness
(window, room size) and affective qualities of space [9].
As far as we could find in the literature, very few studies and 3D
design tools have focused on the qualitative factors of architectural
spaces. Not enough research has been conducted on sensory and
psychological dimensions of space perception. Architectural design
is inherently spatial, and space perception is a sensory experience
that carries meanings wi th emotional dimensions. Affective
architecture articulates color, light, materiality, form and texture to
create sensory perceptive, visually engaging spatial experiences to
elicit emotional responses from the occupant [1 9]. The use of VR technology is yet to be explored in the
psychological and metaphysical dimensions of natural light,
materiality, color and texture that bring subjective meanings,
essences and sensuousness into space -making as an essential part
of the architectural design process.
This paper explores the direct emotional response to the space
design parameters —a phenomenon that we believe has not been
explored in any previous VE studies related to affective
architecture. The human -scale immersive CAVE is appropriate for
this study for i ts large -scale simulations, high resolution,
photorealistic real -time projections on all six screens, ability to use
mixed reality elements, and 360 -degree fully immersive view for
an enhanced sense of presence for the viewer.
2.3 Affective Space Design Par ameters
The visual space perceptual process depends on the properties of
architectural elements —color, light, material, texture, size and
shape —as well as their interrelationships [ 5]. Standard architecture
and interior design practice s use certain design principles to create
affective dimensions of space. The affective aspects of spaces are
psychophysiological dimensions of perceived spatial experience
that elicit or influence an occupant’s moods and emotions, feelings,
preferences and attitudes. Subjectiv e expression and interp retation
of spatial expression are driven by context : culture, imagination,
memory, thoughts or previous emotional states. Mood and
experienced spatial environment are interdependent [ 10].
2.3.1 Color
Warm and cool are sensorial emotions related to hue [ 13]. Warm
colors are at the red end of the color spectrum, consisting of red,
orange, yellow and their combinations. Cool colors are at the
opposite end of the spectrum , consisting of blue, green, purple and
their combinations. Artists, ar chitects, interior designers, scientists,
physicians and psychologists have worked with the science of
color —i.e., its impact on the human physiology, psyche and
behavior, its natural and cultural associations, and interactions
between colors. Color acts a s an emotional, physical and
physiological stimulant. Its perception is contextual. Studies have
acknowledge d certain affirmations regarding affective aspects of
color that are related to arousal and dominance [11]. Warm colors
feel arousing or exciting, w hile cool colors feel calming [ 11]. The
causes behind emotional and physiological responses to a color are
manifold and complex, yet interrelated. Perception of red, as a
symbol of romance, passion, power, criticism, anger, warning or
violence may stem fro m its natural or biological associations with
health, blood, stamina, vigor, heat or flame [2]. Similarly, blue may
soothe a nd calm as the color of the sky. (On the other hand, the idea
of “feeling blue” as an expression of sadness may have biological
roots with emotional and cultural associations. ) Cultural
connotations of a color can have strong impact on human attitude
and behavior. Experiments have demonstrated that student
engagement in class could be improved by avoiding the u se of red
in grading papers , and male aggression in jail cells could be reduc ed
by painting interior walls pink, a color that is culturally recognized
as feminine [2].
Color has an impact on perception of spatial depth as well [26].
Warm colors are known as “advancing colors” that appear to be
closer, and cool colors are known as “receding colors” that appear
to be farther away [ 13]. Artists and architects from the Bauhaus and
De Stijl movements explored perception of pure colors, their
complex interactions with geometric shapes and forms , and spatial
experiences related to color [8]. Architect Le Corbusier conceived
spaces with planar surfaces of pure and vibrant colors and
examined their interactions with exposed materials to e licit
emotional responses [ 6].

2.3.2 Light
Light and color are inseparable . Perception of color in an interior
space depends on direct and indirect illumination, type of light and
interactions between color and light [13]. Light can still appear
“warm” or “cool” based on the feeling it evokes when perceived
with color and space as a whole [ 13]. Additionally, l ight can
provide materials with a sense of depth. Cool and light colors feel
more spacious than warm and dark colors [ 11].
2.3.3 Texture
Texture is the feel, a ppearance and consistency of a surface [ 25]. In
this paper, texture refers to the surface property of rendered
materials that construct perceived space. Visual perception of
spatial texture is informed by past tactile experience of material
weight, shape a nd resistance [20]. The distinct character of spatial
texture of a perceived space derives from the texture of each and all
surrounding surfaces perceived as a whole [21]. Roughness and
smoothness of texture are characteristics of texture graininess,
varying in density, size, orientation and depth of grain. Affective
qualities of perceived texture are inherited from material quality
and temporality : age, origin, inherent permanence or
impermanence, origin and construction process. A rough texture of
a brick or stone may appear richer in character, exciting and
“honest” in its origin, while smooth texture, such as plaster, may
appear comparatively dull or boring [ 21]. Depen ding on material
quality, darker or lighter color tone, the perceived character of
texture may be hard or soft, heavy or light, warm or cool [ 21].
Depending on visual weight and tactility of material and its texture,
a space may feel spacious or intimate, warm or cool.
3 USER STUDY
We performed a user study to assess affective responses to
perceived spatial experiences. Participants were invited to a test
session inside a simulated living space rendered in a 6 -sided
CAVE -type system. Each participant was shown a set of virtual
spaces and w as asked to rate them quantitatively through a set of
questions. Half of the participants performed an activity during
experiment, while the other half remained inactive .
3.1 Method
The study used a mixed design with three within-subject factors —
color, brightness and texture —and one between -subjects factor —
idling/performing an activity. The within -subject design was used
to establish quantitative correlations between variable design
parameters (independent variables) and emo tional responses to
spaces. The design parameters each had two levels: color (orange
and blue), brightness (dark and light) and texture (rough and
smooth). The dependent variables were the quantitative user ratings
of affective dimensions of spaces. The between -subjects factor was
used to assess the impact of activity on perception.
Participants were randomly distributed over two experimental
groups : one which required participants to perform an activity
(active) and one where participants remained idle (i nactive) . Table 1
summarizes the participant pool and distribution over the
experimental groups. In the active condition, sixteen users
performed a minimal -effort daily activity of folding real clothes
from a laundry basket and piling them up on a real table. In the
inactive condition, the users remained still.

Table 1. Between -subjects design of the user study

Experimental
Condition
Activity
Participants Male/Female
Ratio
Active Yes 16 9/7
Inactive No 16 8/8 3.2 The Virtual Environment
3.2.1 Space Layout and Apparatus
A single, square living space (3.35 m x 3.35 m x 3.35 m) was
simulated in the Duke immersive Virtual Environment (DiVE), a
projection -based, 6 -sided CAVE -type display at Duke University.
Screen resolution was 1920×1920. The living space was extended
beyond the actual size of display ( 2.9m x 2.9m x 2.9m ) to stimulate
depth perception due to binocular disparity. The living space was
enclosed by four walls, a ceiling and a floor. It consisted of a mixed
reality (MR) setup, where both real and virtual furniture were
rendered to provide a sense of scale and enhance presence. A study
table and a s mall bed were simulated and placed abutting two
opposing walls (figure 1). A real chair and a round table were
placed inside the virtual room flanked with the simulated study
table on its left and the bed on its right. A real laundry basket full
of unfolde d clothes was placed on the ground in front of the chair
(figure 2).

Figure 1 : Simulated furniture in VE.

Figure 2: Real furniture and props placed inside virtual environment.

The chair was placed at a distance of approximately 210 cm from
the facing wall. The door to the CAVE -type display was behind it.
A calibrating gridline on the floor of the display was used to
provide reference lines and mark the location of the real furniture
and props for each test session. While the user was seated, eye

height was approximately 1.25 -m above the floor. No travel or
other interaction was involved in the test, except for natural head
movement while seated. Participants wore active stereo shutter
glasses and sat on a chair throughout the testing session. An
Intersense IS-900 ultrasonic tracker (head) was used to measure
location and orientation of participants for proper computer
simulation . This spatial setting, including size of space, furniture
layout, configuration and the activity props remained constant
across the two experimental conditions.
The concept of a multi -functional single space draws its
inspiration from the efficiency or studio apartments , and “micro –
apartments ” of overcrowded , high-density big cities, such as San
Francisco and New York [22]. Many tiny apartments rented by the
digital age population have single , multi -functional living areas as
small as 9.3 m2 (100 sqft) [2 7]. These spaces accommodate
overlapping activities with multipurpose, compact furniture and
storage for minimalist belon gings of neo -nomads [ 4].
3.2.2 Space Layout and Apparatus
Three types of texture maps were used for the simulated virtual
surfaces of the living space in order to represent three commonly
used materials. These were stone, drywall and carpet (fabric). A
stone tex ture map was applied only on one wall (facing the user)
that acted as the main accent wall of the room. The surrounding
three walls and ceiling were rendered with drywall texture. A carpet
texture was applied on the floor. All texture maps represented the
actual scale of real materials (figure 3). The texture maps of the
simulated furniture had a predominantly neutral shade to avoid any
color conflict or interaction with simulated surface colors.

Figure 3: Texture mapping of simulated space —stone, drywall and
fabric (carpet).
3.2.3 Simulating Design Parameters
The real -world color, light and texture (surface property of
materials) of an interior space were represented by the color,
brightness and texture gradient of texture maps of all six enclosing
surfaces of the virtual living space:
a) Color —Orange (HSB 18,58,84) and blue (HSB 200, 48, 55);
two complimentary hues were selected from the opposite
spectrum of the color wheel .
b) Brightness —Two levels of brightness were created —bright and
dark. Color tones of texture maps were used to represent real –
world illumination. Increasing Saturation and lowering
Brightness from the HSB (Hue, Saturation, Brightness) values
of each color darkened the color tones. Light and dark color tones represented bright and dark brightness levels. The hue
remained the same (figure 4).
c) Texture —Image -based texture maps of each material were
modified to create two types of texture graininess for each
material —rough and smooth (figure 5). Roughness in texture
meant increased depth and sharpnes s of texture grains of a
material. High resolution, photorealistic images of materials
were manipulated in contrast, highlight and sharpness to create
texture grains that closely represented rough and smooth
characteristics of real surface properties of ma terials. Random
variations for tiling, light and shadow were made for a more
natural and realistic look.

Figure 4: Color tones of orange and blue were darkened to represent
two levels of brightness —bright (left) and dark (right).

Figure 5: Image -based texture maps of stone, drywall and fabric
(carpet) were modified to create two texture gradients —rough (top
row) and smooth (bottom row).
3.2.4 Generating Virtual Test Spaces
Eight unique test spaces were created with two types of three design
parameters , as shown in Table 2. Each space shared a unique
combination of each design parameter (color, brightness and
texture). The depiction of all test spaces is shown in figure 6.

Table 2: Each of the eight test spaces is composed of one
characteristic of each design parameter.

Design Parameters
Virtual Test Spaces
Color
Type Brightness
Levels Texture
Gradients
Orange Dark Rough Orange – Dark – Rough
Smooth Orange – Dark – Smooth
Light Rough Orange – Light – Rough
Smooth Orange – Light – Smooth
Blue Dark Rough Blue – Dark – Rough
Smooth Blue – Dark – Smooth
Light Rough Blue – Light – Rough
Smooth Blue – Light – Smooth

The eight living space configurations were randomized and
presented to users in the virtual environment for assessment. The
order of display for the test spaces was randomized to minimize
biases that could arise from tedium or familiarity. In each
experim ental group, each of the sixteen participants was allocated
a unique randomized sequence. The randomization sequences were
paired between groups.

(a) Blue –Light –Smooth (b) Orange –Light –Smooth

(c) Blue –Dark –Smooth (d) Orange –Dark –Smooth

(e) Blu e–Light –Rough (f) Orange –Light –Rough

(g) Blue –Dark –Rough (h) Orange –Dark –Rough

Figure 6: Screenshots of each of the eight test spaces composed of
one characteristic of each design attribute .
3.2.5 Participants
A total of 32 participants were recruited for the study. Respondents
were undergraduate, graduate and post -graduate students from
various disciplines, as well as professionals (faculty and staff) at
Duke University . Recruitment was done via email and poste r
announcements on the university campus . Participation was
voluntary. Twenty -nine participants (91%) ranged in age from
18—40 (Mean =27, SD= 4.8). Three participants (9%) were in the age range of 50 —76. There were twenty -seven (84%) students and
five (16%) professionals. Among the 27 students, five participants
(16%) were in undergraduate level, thirteen (40%) were in master’s
level and fourteen (44%) were in doctoral or postdoctoral level.
3.2.6 Procedure
The whole experiment took approximately half an hour for each
participant to complete. Participants were required to sign an
informed consent form and fill out a demographic questionnaire
before the test session commenced.
Participants were seated inside the virtual environment. Each of
the eight test spaces w ere displayed in a pre -defined random order
for each participant. Each space was displayed for 20 seconds
before a set of questions was asked regarding that space. While
inactive participants were allowed to stare at each space, active
participants were as ked to perform their task of folding clothes
during those 20 seconds. They were asked to stop once the
questions were asked. Their verbal responses were recorded.
3.2.7 Questionnaire
Participants were asked a set of questions verbally for each virtual
test space displayed. The questions consisted of three categories of
spatial dimensions used to measure the capacity of perceived
spatial experience to influence moods or emotions, in terms of size,
temperature and arousal level .
Participants were asked to rate each space in terms of specific
psychophysiological aspects of space on a scale of 1 to 10. Here, 1
indicated the most negative or lowest rating and 10 indicating the
most positive or highest rating. The aspects were: warm, cool,
spacious, intimate, exciting, calm and comfort ing. They were also
asked to rate each space for two activities : rest and work.
It was important to ensure that participants understood the
semantics or contextual meaning of the spatial terms used in the
questionnaire without being primed. Prior to the virtual session, the
context of these psychophysiological aspects of spaces w as
explained to them in terms of pairs of oppositional adjectives:
spacious and intimate, warm and cool , and exciting and calm.
Participants were instructed to understand exciting spaces as the
opposite of spaces that are calming. Intimate spaces were explained
as the opposite of spacious spaces. Cool was described as a term
opposite of warm to be understood in the context of eliciting a
feeling of a warm or a cool temperature in the spaces.
4 RESULT
Data analysis was performed in IBM SPSS 24. A mixed -design
factorial ANOVA with three repeated -measures factors and one
between -subjects factor was performed. Data was visually
inspected and verified to approach normality . Data is reported as
statistically significant at p<.05 0. Statistically significant
differences are denoted by an asterisk (*) in figures 7 -12.
4.1 Main Effects
Figure 7 shows a summary of the main effects of color. Statist ically
significant main effects of color were found on perceptions of
warmth, coolness, excitement, calmness, intimacy, comfort , and
environments for working and resting . Orange was found
significantly more warm (F(1,30)=153.162 , η2=.836 ), exciting
(F(1,30)=22.087 , η2=.424) and comfortable (F(1,30)=31.772 ,
η2=.514) than blue at p<.001 for all measures. Orange was also
found significantly more intimate than blue (F(1,30)=4.960,
p<.05 0, η2=.142), and significantly preferable for resting
(F(1,30)=10.732 , η 2=.263) and working (F(1,30)=10.247 , η2=.255)
than blue both at p<.005. Blue was found to be significantly more
cool (F(1,30)= 97.117 , p<.001, η2=.764), and calm (F(1,30)=4.918,
p<.050, η2=.141) than orange.

Figure 7: Main effects of color. Error bars represent standard errors.

A significant main effect of brightness was observed only on
perception of spaciousness . As shown in Figure 8, bright space was
found more spacious than dark space (F(1,30)=5.615, p<.05 0,
η2=.158). Also, a near -significant trend points in the direction that
dark spaces were preferred over bright spaces for resting
(F(1,30)=4.169, p= .050, η2=.263).

Figure 8: Brightness effects. Error bars represent standard errors .

No significant main effects of texture and activity were found on
users’ perceptions of emotional spaces. Although a strong trend
towards significance (p= .061) was found that suggested that active
participants perceived spaces more intimate than inactive
participants did.
4.2 Interaction Effects
We fou nd significant interaction effects between color and
brightness on perceptions of warmth (F(1,30)=11.262, p<.005 ,
η2=.273 ); coolness (F(1,30)=9.060, p<.05 0, η2=.232); excitement
(F(1,30)=7.922, p<.05 , η2=.209), and intimacy (F(1,30)=6.753,
p<.05 0, η2=.184).
Pairwise comparisons show that for orange spaces, changing
brightness made a significant difference to perception in specific
situations. Orange felt significantly warmer and more intimate in
dark environments, whereas orange felt significantly cooler in a bright environment (figure 9). On the contrary, blue felt
significantly more exciting in bright environments, but significantly
calmer in dark spaces. Orange felt significantly more intimate than
blue only in dark spaces.

Figure 9: Interactions: Color and brightness. Error bars represent
standard errors .

There was a significant interaction effect between brightness and
texture on a participant’s perception of environment preferable for
working (F(1,30)=4.815, p <.050, η2=.138). Pairwise comparis ons
show that in dark environments, smooth texture was found more
preferable for working than rough texture, whereas no significant
differences were found in bright spaces (figure 10).
Figure 10: Interactions: Brightness and texture on environment for
working . Error bars represent standard errors .

A significant interaction effect between color and activity was
also found on the perception of intimate spaces (F(1,30)= 5.221,
p<.05 0, η2=.148). For inactive participants, changes in color made
a significa nt difference in the perception of intimacy of space , while
no differences were observed for active participants. Inactive
participants found orange significantly more intimate than blue,
whereas no such significant difference was observed for active
participants. Active participants found blue significantly more
intimate than inactive participants did, while the same effect was
not observed with orange spaces (figure 11).

Figure 11: Interactions: Color and Activity on intimacy of space. Error
bars represent standard errors .

Significant interaction effect of brightness and activity was found
only on the perception of warmth (F(1,30)=4.395, p<.05 0,
η2=.128 ). For active participants, changes in level of brightness
made significant difference in the perception of warmth . Active
participants rated dark spaces more warm than bright spaces. No
such significant difference was observed for inactive participants
(figure 12).

Figure 12: Interactions: Brightness and activity on warmth of space.
Error bars represent standard errors .

Finally we found a significant four -way interaction effect among
color, texture, brightness and activity on perception of intimacy of
space (F(1,30)= 4.651, p<.05 0, η2=.134).
5 DISCUSSION
Comparisons were drawn between pairs of oppositional
adjectives —warm and cool, exciting and calm, spacious and
intimate —to find out if perceptions of spatial aspects represented
by oppositional adjectiv es elicit contrasting emotional responses.
Data revealed that orange was found significantly more warm and
exciting than blue, whereas blue was found significantly more cool
and calming than orange, regardless of level of brightness. It can be
deduced from this data that spatial qualities represented by
oppositional adjectives evoked contrasting emotional responses.
This data provides evidence that the VE can generate spatial aspects
of warmth, coolness, and excitement and calmness that were
perceivable to participants. Thus, the hypothesis posed at the
beginning of this paper as a basis for this study can be confirmed.
However, comparisons of data between oppositional adjectives
spaci ous and intimate reveal a different result. While color had a significant impact on the perception of intimate spaces, it did not
have any significant impact on perception of spaciousness . In fact,
brightness was found to have a significant effect on perce ived
spaciousness of space . From this data, it can be concluded that the
participants chose to consider a different or broader semantic for
the word intimate in their understanding of space. It is possible that
they did not perceive the term intimate as a directly oppositional
adjective for the term spaciousness of spatial experience.
The study was able to replicate certain real -world perceptions
related to color and brightness in the virtual environment . In real –
world color perception, warm colors are gen erally found more
exciting than cool colors, whereas cool colors are found more
calming than warm colors. Also, in real -world perception of space,
warm colors appear to be closer in distance than cool colors. The
data gathered in this virtual test revealed that orange was perceived
warmer, more exciting and intimate than blue. On the other hand,
blue was perceived more calm and cool than orange. This correlated
with real -world color experience. Additionally, the user stud y could
also replicate the real -world phenomenon of brightness having a
significant impact on perception of spaciousness .
6 FORMULATING DESIGN PR INCIPLES
The role of VR in the aesthetic practice of space design is the use
of “sensory substitution methods” to replicate the multi -sensorial
real world by understanding how to override real-world sensory
perceptions and in that process, provide affordance for a new
aesthetics that conforms to the medium [ 18]. In this study, the real –
world spatial experiences that are essentially multi -sensory and
three -dimensional are simulated within the technological
constraints of VR as visual ly perceptible experience s with
equivalent graphical methods. The study quantifies the aesthetic
parameters of architectural space -making that are pertinent to
design. The u nderlying aesthetics of a medium influences its
capacity and efficiency [ 18]. This fundamental aesthetics affects
VR’s effectiveness in allow ing compelling space crea tion that
triggers human emotional response and potential activ eness
through perception of sensorial spatial elements.
A set of design guidelines has been developed as the underlying
aesthetics for space design in which one or more parameters can be
modified to create desired spatial qualities with affective
dimensions. These principles can be used by architects and interior
designers as a foundation for living space design for neo-nomadic
minimalist lifestyles.
For desi gn principles, the user study findings of established
correlations between design parameter s and perceptions of
psychophysiological spatial aspects are used as a basis. Three
categories of psychophysiological spatial aspects have been
selected, each with a pair of oppositional affective dimensions:
temperature (warm and cool), size (spacious and intimate) and level
of arousal (excitement and calmness). As it was found in the study
that spatial qualities represented by oppositional adjectives evoked
contrast ing emotional responses, inferences could be drawn to
compare each pair to establish the design principles. For example,
data for cool and calm spaces c ould be inferred from relevant data
found for warm and exciting spaces respectively. Relevant data for
intimate spaces can also be inferred from data on spaciousness .
Here, the term intimate is considered as an exact opposite adjective
of the term spaciousness .
For surface color modifications of space , a color spectrum or a
traditional color wheel is used t o identify ranges of warmer or
cooler shades of any given color [1]. Here , pure red is considered
the warmest color and pure blue is considered the coolest color.
Although the study presented only two colors —orange and blue —
the design guidelines are formed with the assumption that orange
represents warm colors within the range of red, orange and yellow
on the warmer side of the color spectrum , and blue represents cool

colors within the range of green, blue and purple on the cooler side
of the color spectrum .
Table 3 shows the design principles formulated for an occupant’s
desired affective spatial qualities. Statistically significant
correlations form ed guidelines mainly for two design parameters —
color and brightness. Rules are also form ulated for active
occupants. Color is the primary design parameter that can be
modified to create feelings of warmth, coolness, excitement and
calmness . For warm er or more exciting spaces, a warmer shade of
warm color s can be used. For any given color, its warmer shades
are its adjacent colors that are located closer to red in the color
wheel . For example, yellowish -orange is a warmer color than
yellow. On the other hand, for cooler and calmer spaces, cooler
shade s of cool color s can be used . For example, greenish -blue is a
cooler color than green.
In addition to color, brightness levels can also be modified as the
secondary design parameter to further manipulate these feelings .
For a space that has warm colors, decreasing brightness level can
make it fee l warmer , and increasing it can make it feel cooler . On
the other hand, for a space that has cool colors, brightness level can
be increased to make it feel more exciting , and decreased to make
it feel calmer .
As a primary design parameter, brightness lev els can be modified
to increase or decrease feeling of spaciousness , and by inference,
the feeling of intimacy .

Table 3: Correlations between user’s desired spatial qualities and
corresponding design principles

Desired
Affective
Aspect s Design
Parameters
to modify Design Principles
Warm Color
Brightness • Use warm color s—variations of
red, orange or yellow
• Use warmer shades of warm
colors
• If existing color of space is warm ,
decrease brightness to make it
warmer
• For active occupants, decrease
brightness
Cool Color
Brightness • Use cool colors —variations of
blue, green or purple
• Use cooler shades of cool color s
• If occupant prefers warm colors,
increase brightness to make it
cooler
• For active occupants, increase
brightness
Spacious Brightness • Increase brightness
Intimate Brightness • Decrease brightness
Exciting Color
Brightness • Use warm colors —variations of
red, orange or yellow
• Use warmer shades of warm
colors
• If occupant prefers cool colors,
increase brightness
Calm Color
Brightness • Use cool colors —variations of
blue, green or purple
• Use cooler shades of cool colors
• If room already has cool colors,
decrease brightness 7 CONCLUSION & FUTURE WORK
The outcome of the study assisted in formulating aesthetic
guideline s for affective living spaces p ertaining to sensory
engagement . One major focus of future work is to explor e
architectural design elements that can potentially evoke feelings of
spaciousne ss in a tiny, shared living space intended for young
professionals. It includes virtually simulated sunlight and openings
(windows) as design elements to examine their affective
dimensions. Perception of interior surfaces can extend beyond the
traditional wall and reflect abstract representations of external
views, weather conditions, and real-time diurnal and seasonal
changes with the use of simulated color, light and texture.
This research can be taken further into the domain of sensor –
driven interactive living spaces. Interactive architecture is defined
as the ability to provide variable characteristics of living, working
and leisure spaces. While VR can serve as an architectural
evaluation tool to test, replicate and evaluate real -world spatial
situation s, its interactive capacity can be utilized to explore real-
time space -making as an aesthetic medium. The use of VEs can
provide a highly effective aesthetic tool that allows changing of
spaces to trigger human emotional responses in order to support
human activities through interaction. The immersive human -scale
CAVE -type display can work as an architectural design tool to
create a variety of spaces for testing and applying in real -world
interactive living spaces.
ACKNOWLEDGMENTS
Experiments at the Duke University DiVE were made possible by
a Dissertation Research Award from the University of Texas at
Dallas and by Duke University. Dr. Timothy J. Senior
(Neuroscience, Oxford University) suggested DiVE as an
immersive environment fo r experimental evidence of design
choices. Hypotheses informing the experiments were presented at
the international conference “Anticipation Across Disciplines”
(September 2014). The presentation was made possible through
funding from the Hanse Institute f or Advanced Study (Hanse
Wissenschaftskolleg, Germany) and the antÉ – Institute for
Research in Anticipatory Systems. The inter -institutional IRB is the
result of cooperation between the University of Texas and Duke
University.
REFERENCES
[1] G. Agoston. Color Theory and Its Application in Art and Design .
Springer -Verlag. 1986.
[2] A. Alter. Drunk Tank Pink . The Penguin Press, New York, 2012.
[3] A. Angulo. Rediscovering Virtual Reality in the Education of
Architectural Design: The immersive simulation of spatial
experiences, Ambiances: International Journal of Sensory
Environment , Architecture and Urban Space , September 8, 2015.
http://ambiances.revues.org/594
[4] N. Bowles. The Real Teens of Silicon Valley: Inside the almost –
adult lives of the industry’s newest recruits . The California Sunday
Magazine . [Online]. https://stories.californiasunday.com/2015 -06-
07/real -teenagers -silicon -valley. Accessed May 2015 .
[5] F. D. K. Ching. Architecture: Form, Space, and Order . Wiley,
Hoboken, 2007.
[6] J. de Heer. The Architectonic Colour: Polychromy in the Purist
Architecture of Le Corbusier . 010 Publishers, Rotterdam, 2009.
[7] G. Drettakis, M. Roussou, A. Reche, and N. Tsingos. Design and
Evaluation of a Real -World Virtual Environment for Architecture
and Urban Planning. Presence: Teleoperato rs and Virtual
Environments , 16(3), 2007.
[8] K. Frampton. Modern Architecture: A Critical History . Thames &
Hudson, 1980.

[9] G. Franz, M. von der Heyde, and H. H. Bülthoff. An empirical
approach to the experience of architectural space in virtual reality —
exploring relations between features and affective appraisals of
rectangular indoor spaces. Automation in Construction , volume 14,
pages 165 -172, Germany, 2005.
[10] G. Franz. An Empirical Approach to the Experience of Architectural
Space . PhD thesis, Max Planck Institute for Biological Cybernetics,
Tübingen and the Bauhaus University, Weimar, 2005.
[11] G. Franz. Space, color, and perceived qualities of indoor
environments. In Environment, Health and Sustainable
Development (IAPS 19) , Seattle, 2006, pp. 1 -8.
[12] P. Fröst and P. Warren. Virtual reality used in a collaborative
architectural design process. In: Proceedings of IV2000 IEEE
International Conference on Information Visualization . IEEE
Computer Society, Brussels, 2000.
[13] M. Hårleman. Daylight Influence on Co lour Design: Empirical study
on Perceived Colour and Colour Experience Indoors . PhD thesis,
Royal Institute of Technology, Sweden. 2007
[14] R. G. Hershberger and R. C. Cass. Predicting user responses to
buildings. In Jack L. Nasar, editor, Environmental Aesthe tics:
Theory, research, and applications , Cambridge U niversity Press
Cambridge, 1988, pp. 195 –211.
[15] D. Henry. Spatial Perception in Virtual Environments: Evaluating
and Architectural Application . Master’s thesis, University of
Washington, 1992.
[16] J. Leigh and A. E. Johnson, "CALVIN: an immersimedia design
environment utilizing heterogeneous perspectives," Proceedings of
the Third IEEE International Conference on Multimedia Computing
and Systems , Hiroshima, 1996, pp. 20 -23. [17] M. P. Mobach. Do virtual worlds cre ate better real worlds? Virtual
Reality , 12(3), September 2008. pp. 163 -179.
[18] M. Nadin. Negotiating the World of Make -believe: The Aesthetic
Compass. Real-Time Imaging. 1(3), August 1995, pp. 173 -190.
[19] A. Naz. Interactive Living Space Design for Neo -Nomads:
Anticipation Through Spatial Articulation. In Mihai Nadin, editor,
Anticipation Across Disciplines , Springer International Publishing
Switzerland, 2016 , pp. 393 –403.
[20] J. Pallasmaa. The Eyes of the Skin : Architecture and the Senses . John
Wiley & Sons, 2012
[21] S. E. Rasmussen. Experiencing Architecture . MIT Press, 2000.
[22] N. J. Riley. S.F. supervisors back micro -apartments. SFGate . Nov
20, 2012. [Online]. http://www.sfgate.com/ bayarea/article/SFsuper
visorsbac kmicroapartments4055493.php
[23] J. Ryu, N. Hashimoto, M. Sato, M. Soeda, and R. Ohno. Application
of Human -scale Immersive VR System for Environmental Design
Assessment – A Proposal for an Architectural Design Evaluation
Tool. Journal of Asian Architecture and Building Engineering
(JAABE) , 6(1), May 2007
[24] I. E. Sutherland. The Ultimate Display. Proc. IFIP 65(2), 1965, pp.
506-508, 582 -583.
[25] M. Wiberg. Interactive Textures for Architecture and Landscaping .
IGI Global, Hershey, 2011.
[26] D. Yorgancioglu. Steven Holl: A Translation of Phenomenological
Philosophy in the realm of Architecture . PhD thesis, Middle East
Technical University, September, 2004.
[27] What is the Tiny House Movement? (n.d.). The Tiny Life . [Online].
http://thetinylife.com/what -is-the-tiny-house -movemen t/. Accessed
June 14, 2016.