A
psychological
approach
to
evidence‐based
architectural
design
in
domes7c
spaces
Colin
Ellard1,

Deltcho
Valtchanov1,
Thomas
Seebohm1,
Jus;n
Perdue2,
Sarah
Susanka3,
Mark
Zanna1
1University
of
Waterloo,
Waterloo,
Ontario
Canada,
2Stantec
Architecture,
Toronto,
3Susanka
Studios,
Raleigh,
NC
Introduc7on
In
this
experiment,
we
demonstrate
a
principled
method
for
assessing
the
comfort,
usability,
and
affec;ve
proper;es
of
a
series
of
domes;c
spaces.

We
combine
high‐level
visualiza;on
with
immersive
virtual
reality,
mo;on
tracking
and
structured
interviews
to
assess
the
quali;es
of
a
variety
of
styles
of
domes;c
spaces.

This
ini;al
study
is
meant
as
a
“proof
of
concept”
for
future
studies
in
a
variety
of
built
seVngs
ranging
from
ins;tu;onal
and
health
care
environments
to
the
urban
scale.



Method
3d
visualiza;ons
of
three
exis;ng
domes;c
structures
were
constructed
using
3d
Studio
Max
and
V‐ray.

The
structures
were
Frank
Lloyd
Wright’s
Jacobs
House,
Sarah
Susanka’s
Milstein
House,
and
a
typical
North
American
suburban
house.

Par;cipants
were
provided
with
immersive
experiences
of
the
models
using
head‐mounted
displays,
infrared
mo;on
tracking,
and
mobile
compu;ng.

The
models
were
realized
at
actual
size
in
a
gymnasium.

Par;cipants
explored
the
spaces
un;l
they
felt
familiar
with
them,
and
then
they
were
asked
a
series
of
ques;ons
about
their
preferences
for
par;cular
loca;ons.


Figure
1.

A
screenshot
of
the
Milstein
House
visualiza7on


Results

Par;cipants
were
asked
about
their
overall
preferences
for
certain
aspects
of
each
of
the
models.
Results
suggested
that
designers’
inten;ons
were,
to
some
extent,
manifest
in
the
opinions
of
par;cipants.

For
example,
the
Milstein
House,
which
was
designed
to
enhance
both
opportuni;es
for
socializing
and
privacy
[1],
appears
to
have
met
both
of
these
objec;ves.


Table
1.
Mean
responses
of
par7cipants
to
general
ques7ons
pertaining
to
affec7ve
proper7es
of
the
three
VR
models.

Par7cipants
were
asked
to
rank
each
of
the
three
models
on
a
scale
of
1‐5
for
the
parameters
listed
in
the
table.

The
final
ques7on
asked
par7cipants
how
likely
they
would
be
to
purchase
each
of
the
homes
if
they
had
the
opportunity
to
do
so.


Using
Ajanachara
so^ware
[2]
the
spa;al
proper;es
of
the
home
designs
were
analyzed
to
determine
such
proper;es
as
isovist
size
(area
of
space
observable
from
a
given
loca;on),
and
other
higher
order
proper;es
of
spa;al
shape.

Analyses
were
converted
to
contour
plots
to
visualize
how
spa;al
proper;es
varied
between
loca;ons
in
the
space.

The
figure
below
shows
an
example
of
such
a
contour
plot
from
the
Jacobs
House.



Figure
2.

Spa7al
analysis
of
Jacobs
House.

Colour
coding
corresponds
to
size
of
visible
region
of
space
from
each
loca7on
in
the
space.

Light
colours
indicate
large
visible
regions
of
space
and
dark
colours
indicate
small
visible
regions
of
space.


Figure
3.

Par7cipants
were
asked:
“If
this
were
your
house
and
you
were
having
a
party,
where
would
you
go
to
oversee


Conclusions
The
findings
described
here,
although
very
preliminary
in
nature
suggest
the
viability
of


a
method
for
assessing
the
affec;ve
proper;es
of
architectural
spaces
using
a
combina;on
of
immersive
virtual
reality,
mo;on
tracking,
and
tradi;onal
structured
interview
methods
of
experimental
psychology.

In
future
work,
we
hope
to
add
to
these
findings
by
including
measurements
of
physiological
reac;ons
during
explora;on
of
such
spaces,
including
eye
movements,
heart
rate,
skin
conductance
and
EEG.


References


[1]
Susanka,
S.
(1998).
The
not
so
big
house:
A
blueprint
for
the
way
we
really
live.
Taunton
Press:
Newton
CT.



[2]
Wiener,
J.M.,
Franz,
G.,
Rossmanith
N.,
Reichelt
A.,
Mallot
H.A.,
&
Bulthoff
H.H.
(2007).

Isovist
analysis
captures
proper;es
of
space
relevant
for
locomo;on
and
experience.

Percep:on,
36(7),
1066‐1083.



This
work
was
supported
by
funding
from
the
Social
Sciences
and
Humani:es
Research
Council
of
Canada.

The
authors
gratefully
acknowledge
the
assistance
of
Gloria
Sicuro.
Ela
Malkovsky,
Hannah
Masterman,
Leanne
Quigley,
and
Jeff
Bruce
in
collec:on
of
data
and
contribu:on
of
ideas
to
this
experiment.


Par;cipants
were
asked
to
indicate
their
preferred
loca;ons
in
the
spaces
for
par;cular
ac;vi;es
(surveying
a
party,
undertaking
quiet
conversa;on
or
thought,
etc.)
in
order
to
assess
their
impressions
of
the
affec;ve
proper;es
of
spaces.

Preliminary
findings
suggest
that
preferences
correlate
with
spa;al
proper;es
having
to
do
with
visible
sizes
and
shapes
of
spaces.