pattern matters in interior space perception

PATTERN
MATTERS IN
I N T E R I O R S P A C E P E RC E P T I O N
Christoph von Castell, Heiko Hecht, Daniel Oberfeld
Johannes Gutenberg-Universität Mainz, Department of Psychology, General Experimental Psychology
MOTIVATION
Interior-design experts:
o Horizontally striped walls make a room appear wider and lower, vertically striped walls make a room appear higher and narrower (e.g., Neufert & Kister, 2009)
Visual expansion in the orientation of striped patterns
Previous findings of surface-characteristics effects on the perceived size of small objects:
o Helmholtz-square (von Helmholtz, 1867): A horizontally striped square appears taller than wide, a vertically striped square appears wider than tall
o Oppel-Kundt-illusion (OKI; Kundt, 1863; Oppel, 1861): An exocentric distance between two vertical sticks appears larger when divided by additional vertical sticks
o OKI depends on the number of dividing sticks – e.g., Mikellidou & Thompson (2014): Maximum visual expansion with eight to 12 sticks, reversal of effect with one stick
Visual expansion perpendicular to the orientation of striped patterns
Who is right? Or, do interior space and objects behave differently?
METHOD
DESIGN AND PROCEDURE
Independent variables (IVs)
o Room width:
4.30, 4.50, 4.70 m
o Ceiling height:
2.90, 3.00, 3.10 m
Dependent variables (DVs)
o Perceived width
o Perceived height
Centimeter ratings in two separate blocks
Order of blocks balanced between subjects
EXPERIMENT 1
o Pattern orientation:
Plain, vertical stripes, horizontal stripes
(10 cm stripe width)
3 trials per combination of the IVs × 2 DVs =
162 trials per subject
STIMULI
APPARATUS
3D room simulations
o Variation of wall pattern
o Variation of room width
o Variation of ceiling height
o Constant room depth (6.00 m)
o Constant overall lightness of walls
(controlled by spectroradiometer)
Large projection screen
o Stereoscopic viewing
o Eye position horizontally and vertically
centered to screen
o Viewing distance: 2.00 m
o Virtual eye height: 1.30 m
o FOV: 66.00° horizontal and 52.00° vertical
EXPERIMENT 2A
o Number of vertical stripes (spatial frequency):
0, 4, 12, 36
3 trials per combination of the IVs × 2 DVs =
216 trials per subject
Observer’s (O) position relative to the projection
screen (S) and the variably wide virtual room
(rectangular frames). The grey shaded area was not
visible to the observer. Virtual dimensions are
printed in grey, real dimensions are printed in black.
EXPERIMENT 2B
o Number of vertical stripes (spatial frequency):
0, 36, 72, 144, 288
3 trials per combination of the IVs × 2 DVs =
270 trials per subject
SUBJECTS
o N = 20 subjects each
RESULTS
Separate rmANOVAs
for both DVs
o Room width in all experiments
Perceived width increased with
increasing room width
o Ceiling height in all experiments
Perceived width decreased with
increasing ceiling height
Wall pattern
×
Room width
×
Ceiling height
o Room width in all experiments
Perceived height decreased with
increasing room width
o Ceiling height in all experiments
Perceived height increased with
increasing ceiling height
Note: All other effects n.s.
CONCLUSION
o EXPERIMENT 1: Virtually no effect of pattern orientation on width and height estimates
At odds with both the experts' design recommendations and von Helmholtz's (1867) assumptions
Probably due to relatively low number of stripes?
o EXPERIMENT 2A and EXPERIMENT 2B: Considerable effect of the number of vertical stripes
Smaller width and height estimates for low-frequency (up to 12 vertical stripes) wall paints than for plain
(unstriped) wall paints
Larger width and height estimates for high-frequency (288 vertical stripes) wall paints than for plain
(unstriped) wall paints
Width estimates consistent with previous results on the number of dividing sticks in the OKI (e.g.,
Mikellidou & Thompson, 2014)
When you seek to visually enlarge your room's perceived size by means of a striped wall paint, we recommend working with a slender brush.
REFERENCES
Kundt, A. (1863). Untersuchungen über Augenmass und optische Täuschungen. Poggendorffs Annalen, 120, 118-158.
Mikellidou, K., & Thompson, P. (2014). Crossing the line: Estimations of line length in the Oppel-Kundt illusion. Journal of Vision, 14(8).
Oppel, J. J. (1861). Über geometrisch-optische Täuschungen (zweite Nachlese). Jahresbericht des physikalischen Vereins zu Frankfurt am
Main 1860/61 (pp. 26-37).
Von Helmholtz, H. (1867). Handbuch der physiologischen Optik (Allgemeine Encyklopädie der Physik, Vol. 9). Leipzig: Voss.
CONTACT: CASTELL@UNI-MAINZ.DE
TEAP 2016
HEIDELBERG
r epr int
We are grateful to Agnes Münch for programming the experiments.
This study was funded by the Deutsche Forschungsgemeinschaft: "Innenraumwahrnehmung", HE 2122/10-1
(Heiko Hecht) and OB 346/5-1 (Daniel Oberfeld).
Sc an for
ACKNOWLEDGEMENTS