Regional Differences in the Perception of Daylit Scenes across Europe Using Virtual Reality. Part II: Effects of Façade and Daylight Pattern Geometry

Figures & data

Table 1. Overview of studied factors and their levels

Factor type	Level
Within-subject	Façade geometry: 4 variations
Between-subject	Sky type: clear with high sun angle, clear with low sun angle, overcast sky
Between-subject	Spatial context: socializing or working
Between-subject	Country: Norway, Switzerland, Greece

Fig. 1. The selected façade geometry variations (top row) and the corresponding examples of contemporary architecture they were based on (bottom row). Façade variations were manipulated to have the same perforation ratio. From left to right: [1] Faena Aleph Residences, Foster + Partners, Buenos Aires, Argentina, 2012; [2] Selcuk Ecza Headquarters, Tabanlıoglu Architects, Istanbul, Turkey, 2013; [3] Kew House, Piercy & Company, Richmond, United Kingdom, 2014; [4] Federation Square, Lab Architecture Studio, Melbourne, Australia, 2002.

Fig. 2. Perspective view representing a participant’s viewpoint in virtual reality in scenes with Pattern 1 and 2 for all context and sky type combinations. Note that there is a narrow west-facing window between the two ceiling levels. The difference in brightness of this window between different sky conditions in these images is caused by external reflections and the applied tone-mapping algorithms.

Table 2. Radiance material properties for the main surfaces

Surface	Type	R	G	B	Reflectance	Specularity	TVis
Ceiling	Plastic	0.95	0.94	0.92	94%	0
Floor	Plastic	0.31	0.31	0.31	31%	0
Columns	Plastic	0.54	0.52	0.46	52%	0
Walls	Plastic	0.94	0.93	0.90	93%	0
Façade	Plastic	0.25	0.25	0.25	25%	0
Furniture (social)	Plastic	0.70	0.70	0.70	70%	0
Furniture (work)	Plastic	0.75	0.75	0.75	75%	0
Doors	Plastic	0.09	0.09	0.09	9%	0
Window	Glass	0.654	0.654	0.654			60%

Table 3. Radiance simulation parameters

-ds	-ds	-dt	-dc	-dp	-st	-ab	-aa	-ar	-ad	-as	-lr	-lw
0.02	0.05	0.05	0.5	256	0.5	4	0.02	32	50,000	25,000	4	0.000003
Resolution: 12,960 × 12,960 pixels.

Table 4. Categories of stature, eye height, and corresponding participant height

Stature [m]	Eye height [m]	Shown to participants with height [m]
1.5308	1.427	≤ 1.58
1.6363	1.5325	> 1.58 and up to 1.69
1.7408	1.637	> 1.69 and up to 1.80
1.8518	1.748	> 1.80

Fig. 3. Perspective view representing a participant’s viewpoint in virtual reality in scenes with Pattern 3 and 4 for all context and sky type combinations.

Table 5. Verbal questionnaire used in the study

Perceptual impressions
Pleasantness	How pleasant is this space?*
Interest	How interesting is this space?*
Excitement	How exciting is this space?*
Calmness	How calming is this space?*
Visual appearance
Complexity	How complex is this space?*
Brightness	How bright is this space?*
Spaciousness	How spacious is this space?*
Satisfaction with the amount of view	How satisfied are you with the amount of view in this space?*

*A scale from 0 to 10, with 0 corresponding to “Not at all” and 10 to “Very,” was used for all marked questions.

Table 6. Demographic characteristics of participants in each country of study

		Norway	Switzerland	Greece
Sample size		150	118	138
Gender	Male	86 (57%)	60 (51%)	74 (54%)
	Female	64 (43%)	58 (49%)	64 (46%)
Age	Range [years]	20–49	20–50	19–44
	M (SD) [years]	24.8 (6.5)	27.4 (7.5)	24.2 (4.5)
Nationality	Country of study*	123 (82%)	35 (30%)	130 (94%)
	Other	27 (18%)	83 (70%)	8 (6%)
Training	Architecture	37 (24%)	10 (8%)	47 (35%)
	Lighting	10 (6%)	0 (0%)	6 (4%)
	Virtual reality	5 (3%)	2 (2%)	3 (2%)
	None of the above	98 (7%)	106 (90%)	82 (59%)

*Norwegian, Swiss, and Greek nationality for participants in Norway, Switzerland, and Greece, respectively.

Table 7. Distribution of sample size per country, spatial context, and sky type

	Norway		Switzerland		Greece
Sky type	Social	Work	Social	Work	Social	Work
Clear, high sun angle	26	26	20	20	21	23
Clear, low sun angle	25	25	20	20	27	23
Overcast sky	24	24	18	20	22	22
Sum	75	75	58	60	70	68

Table 8. Results of the LMM analysis for the main factors of interest Country, Façade geometry, Spatial context, and Sky type on all dependent variables

Predictor	Attribute	df	F	p-value
Country	Pleasantness	2, 391.82	0.37	0.69
	Calmness	2, 391.79	2.89	0.06
	Interest	2, 391.90	2.09	0.13
	Excitement	2, 391.88	5.46	0.005
	Brightness	2, 391.94	10.72	<0.0001
	Complexity	2, 391.87	0.26	0.77
	Spaciousness	2, 391.95	1.60	0.20
	Amount of view*	2, 391.88	1.85	0.16
Façade geometry	Pleasantness	3, 1215.93	29.20	<0.0001
	Calmness	3, 1215.95	46.98	<0.0001
	Interest	3, 1215.88	98.01	<0.0001
	Excitement	3, 1215.89	86.96	<0.0001
	Brightness	3, 1215.82	74.26	<0.0001
	Complexity	3, 1215.90	175.91	<0.0001
	Spaciousness	3, 1215.81	21.12	<0.0001
	Amount of view*	3, 1215.89	188.35	<0.0001
Spatial context	Pleasantness	1, 403.60	6.87	0.009
	Calmness	1, 402.71	6.09	0.01
	Interest	1, 407.38	0.16	0.69
	Excitement	1, 406.48	0.39	0.53
	Brightness	1, 412.45	1.52	0.22
	Complexity	1, 405.81	0.53	0.47
	Spaciousness	1, 413.43	3.16	0.08
	Amount of view*	1, 406.51	0.17	0.68
Sky type	Pleasantness	2, 400.97	1.00	0.37
	Calmness	2, 400.29	0.70	0.50
	Interest	2, 403.88	0.71	0.49
	Excitement	2, 403.18	0.62	0.54
	Brightness	2, 407.78	2.51	0.08
	Complexity	2, 402.66	0.99	0.37
	Spaciousness	2, 408.53	1.50	0.22
	Amount of view*	2, 403.20	0.03	0.97

*Satisfaction with the amount of view.

Table 9. Marginal and conditional R² of the LMM for each studied attribute

Attribute	R^2marginal	R^2conditional
Pleasantness	0.052	0.382
Interest	0.112	0.517
Excitement	0.117	0.499
Calmness	0.083	0.373
Complexity	0.198	0.531
Spaciousness	0.066	0.604
Brightness	0.121	0.613
Satisfaction with the amount of view	0.185	0.539

Table 10. Estimated marginal means (EMM), standard errors (SE), and 95% confidence intervals (CI) per attribute and country

	Norway			Switzerland			Greece
Attribute	EMM	SE	95% CI	EMM	SE	95% CI	EMM	SE	95% CI
Pleasantness	5.48	0.13	[5.17, 5.80]	5.40	0.13	[5.09, 5.71]	5.55	0.12	[5.26, 5.83]
Calmness	5.24	0.13	[4.93, 5.55]	5.27	0.13	[4.96, 5.58]	5.61	0.12	[5.32, 5.89]
Interest	5.57	0.14	[5.23, 5.91]	5.27	0.14	[4.93, 5.60]	5.64	0.13	[5.33, 5.95]
Excitement	5.12	0.14	[4.79, 5.64]	4.72	0.13	[4.39, 5.05]	5.33	0.13	[5.03, 5.64]
Brightness	5.56	0.13	[5.25, 5.88]	5.96	0.13	[5.65, 6.27]	6.36	0.12	[6.08, 6.64]
Complexity	5.12	0.14	[4.78, 5.46]	5.12	0.14	[4.79, 5.46]	5.01	0.13	[4.70, 5.32]
Spaciousness	7.15	0.13	[6.85, 7.45]	7.25	0.12	[6.96, 7.55]	7.44	0.11	[7.17, 7.71]
Amount of view*	4.88	0.16	[4.50, 5.26]	4.86	0.16	[4.49, 5.23]	5.21	0.14	[4.87, 5.56]

*Satisfaction with the amount of view in the space.

Fig. 4. Estimated marginal means, standard deviations, and pairwise comparisons of reported brightness across countries. Asterisks represent statistical significance in the pairwise comparisons. The notation **** indicates p < .0001.

Table 11. Estimated marginal means (EMM), standard errors (SE), and 95% confidence intervals (CI) per attribute and façade geometry variation, color-coded from the lowest (white) to the highest (dark gray) mean. Results are averaged over the levels of the other factors in the LMM

Attribute		Pattern 1	Pattern 2	Pattern 3	Pattern 4
Pleasantness	EMM (SE)	5.43 (0.10)	5.99 (0.10)	4.97 (0.10)	5.51 (0.10)
	95% CI	[5.23, 5.62]	[5.8, 6.29]	[4.77, 5.17]	[5.32, 5.71]
Interest	EMM (SE)	4.50 (0.10)	5.61 (0.10)	5.66 (0.10)	6.20 (0.10)
	95% CI	[4.30, 4.70]	[5.41, 5.82]	[5.46, 5.86]	[5.99, 6.40]
Excitement	EMM (SE)	4.15 (0.10)	5.23 (0.10)	5.07 (0.10)	5.78 (0.10)
	95% CI	[3.95, 4.35]	[5.03, 5.43]	[4.87, 5.27]	[5.58, 5.98]
Calmness	EMM (SE)	5.69 (0.10)	5.99 (0.10)	4.89 (0.10)	4.92 (0.10)
	95% CI	[5.49, 5.89]	[5.79, 6.19]	[4.69, 5.09]	[4.72, 5.12]
Complexity	EMM (SE)	3.77 (0.10)	4.82 (0.10)	5.77 (0.10)	5.98 (0.10)
	95% CI	[3.57, 3.98]	[4.62, 5.03]	[5.56, 5.97]	[5.77, 6.18]
Spaciousness	EMM (SE)	7.34 (0.08)	7.45 (0.08)	6.93 (0.08)	7.40 (0.08)
	95% CI	[7.17, 7.51]	[7.29, 7.62]	[6.76, 7.09]	[7.23, 7.57]
Brightness	EMM (SE)	6.10 (0.09)	6.28 (0.09)	5.25 (0.09)	6.21 (0.09)
	95% CI	[5.93, 6.27]	[6.11, 6.46]	[5.08, 5.43]	[6.03, 6.38]
Amount of view	EMM (SE)	5.30 (0.11)	5.50 (0.11)	3.32 (0.11)	5.81 (0.11)
	95% CI	[5.07, 5.52]	[5.28, 5.73]	[3.10, 3.55]	[5.59, 6.04]

Fig. 5. Estimated marginal means and standard error per country and per façade geometry variation for the attributes pleasant, interesting, exciting, and calming. Pairwise comparisons are shown between façade variations (averaged across levels of country, sky type, and context). Asterisks represent statistical significance in the pairwise comparisons: *** indicates p < .001 and **** indicates p < .0001.

Fig. 6. Estimated marginal means and standard error per country and per façade geometry variation for the attributes complex, bright, spacious, and satisfaction with the amount of view in the space. Pairwise comparisons are shown between façade variations (averaged across levels of country, sky type, and context). Asterisks represent statistical significance in the pairwise comparisons: *** indicates p < .001 and **** indicates p < .0001.