A Comparison of Methodologies to Investigate the Influence of Light on the Atmosphere of a Space

Figures & data

Fig. 1 Schematic floor plan of virtual 3D space.

Table 1. Overview of light characteristics and luminance calculations of the light conditions. Michelson contrast was calculated by (L_max − L_min)/(L_max + L_min).

Light condition	Electric light					Daylight	Mean luminance (cd/m^2)	Michelson contrast
	Light source white light	Intensity white light (lux)^a	CCT (K) white light	Colored accent light LED spot (RGB value)	Colored accent light wallwasher (RGB value)	Type
Neutral	Fluorescent	95	3500	n/a	n/a	No	58.2	0.77
1	Fluorescent	95	5500	n/a	n/a	No	58.2	0.77
2	Fluorescent	95	2750	n/a	n/a	No	58.2	0.77
3	Spots	95	2750	n/a	n/a	No	55.4	0.91
4	Spots	95	2750	Orange (RGB = 100, 50, 0)	Blue (RGB = 0, 0, 22)	No	58.2	0.92
5	Fluorescent/spots	330	5500	Blue (RGB = 0, 0, 100)	Cyan (RGB = 0, 100, 100)	No	102.5	0.68
Neutrald1	Fluorescent	95	3500	n/a	n/a	Diffuse	58.2	0.75
1d1	Fluorescent	95	5500	n/a	n/a	Diffuse	58.2	0.75
2d1	Fluorescent	95	2750	n/a	n/a	Diffuse	58.2	0.75
3d1	Spots	95	2750	n/a	n/a	Diffuse	55.4	0.80
4d1	Spots	95	2750	Orange (RGB = 100, 50, 0)	Blue (RGB = 0, 0, 22)	Diffuse	58.2	0.82
5d1	Fluorescent/spots	330	5500	Blue (RGB = 0, 0, 100)	Cyan (RGB = 0, 100, 100)	Diffuse	108.0	0.68
4d2	Spots	95	2750	Orange (RGB = 100, 50, 0)	Blue (RGB = 0, 0, 22)	Direct 1 (azimuth = 0°, elevation = 45°)	41.6	0.86
4d3	Spots	95	2750	Blue (RGB = 0, 0, 100)	Cyan (RGB = 0, 100, 100)	Direct 2 (azimuth = 22.5°, elevation = 60°)	55.4	0.94

^aIlluminance level measured vertically at the eye in the real space the visualization resembled and was built after.

Fig. 2 Visualizations of the light conditions used in the experiments. The first column depicts all light conditions without daylight. The second column depicts the same light conditions in the presence of diffuse daylight. The third and fourth columns depict light condition 4 in the presence of two types of direct daylight.

Fig. 3 This graph depicts luminance measurements of light condition 2 taken in the 40° high band in the real space on which the light effect in the visualization was based. Luminance measurements with and without diffuse daylight are depicted, where the y-axis presents the luminance (in cd/m²) and the x-axis presents the horizontal dimension of the 40° high band in the space. This figure has been published before in Stokkermans and others [2015].

Table 2. Overview of atmosphere dimensions and their items.

Dimensions	Coziness	Liveliness	Tenseness	Detachment
Items	Cozy	Lively	Tense	Formal
	Pleasant	Stimulating	Threatening	Business-like
	Intimate	Exiting	Terrifying

Fig. 4 Schematic representation of the total effect model (shown at the top of the figure) and the mediation/moderation model (shown at the bottom of the figure). The total effect and mediation effects are indicated by solid lines and moderation effects with dashed lines.

Table 3. Cronbach’s alpha values for the four atmosphere attributes measured in experiments 1 and 2.

	Cronbach’s alpha values
	Original experiment	Present experiment
Coziness	0.85	0.78
Liveliness	0.87	0.88
Tenseness	0.83	0.87
Detachment	0.89	0.90

Fig. 5 Graph per measured attribute depicting the results of a random versus blocked presentation of the daylight conditions. Each data point represents the mean over all five electric light conditions with the error bar representing the 95% confidence interval.

Table 4. P-value and effect size (η_g²) per attribute for presenting the daylight conditions randomly versus blocked.

	Random daylight presentation			Blocked daylight presentation
	P-value	Effect size (ηg^2)	Interpretation effect size	P-value	Effect size (ηg^2)	Interpretation effect size
Brightness	0.710	0.000	Small	0.471	0.002	Small
Perceived color	0.001	0.050	Medium	0.023	0.009	Small
Perceived uniformity	0.946	0.000	Small	0.332	0.002	Small
Coziness	0.083	0.010	Small	0.590	0.000	Small
Liveliness	0.657	0.000	Small	0.919	0.000	Small
Tenseness	0.179	0.004	Small	0.082	0.005	Small
Detachment	<0.001	0.084	Medium	0.002	0.026	Small

Fig. 6 Effect of paired-comparison versus rating-scale methodology. Each graph (per attribute) presents the mean score of the rating-scale methodology and the z-score of the paired-comparison methodology for all light conditions.

Table 5. Pearson correlation coefficient per attribute between the z-score per light condition resulting from the paired-comparison methodology and the mean score of each light condition resulting from the rating-scale methodology. Additionally, the median effect size ratio, or the effect size of the paired-comparison results divided by the effect size of the rating-scale results, is given for all attributes.

	Pearson correlation coefficient	P-value	Median effect size ratio (paired comparison/rating scale)
Brightness	0.971	<0.001	2.4
Perceived color	0.958	<0.001	1.5
Perceived uniformity	0.954	<0.001	1.5
Coziness	0.975	<0.001	2.1
Liveliness	0.966	<0.001	2.1
Tenseness	0.685	0.014	3.2
Detachment	0.976	<0.001	2.1

Table 6. Overview of the outcomes of the repeated-measures ANOVAs for the main and interaction effects of daylight and electric light condition.

Repeated measures ANOVA	Main/interaction effect	Attribute	Huyn-Feldt correction	df	Sum of squares	F	P	ηg^2
Effect of diffuse daylight for all electric light conditions	Daylight	Brightness	No	1	0.14	0.14	0.710	0.000
		Perceived color	No	1	15.3	13.7	0.001	0.050
		Perceived uniformity	No	1	0.00	0.01	0.946	0.000
		Coziness	No	1	3.31	3.26	0.083	0.012
		Liveliness	No	1	0.21	0.20	0.657	0.000
		Tenseness	No	1	1.44	1.92	0.179	0.004
		Detachment	No	1	40.8	30.0	<0.001	0.085
	Electric light condition	Brightness	No	4	111	17.4	<0.001	0.217
		Perceived color	No	4	458	81.6	<0.001	0.611
		Perceived uniformity	No	4	489	64.6	<0.001	0.550
		Coziness	Yes	3.0	189	33.4	<0.001	0.400
		Liveliness	Yes	3.2	206	34.9	<0.001	0.404
		Tenseness	No	4	33.6	9.34	<0.001	0.088
		Detachment	Yes	3.3	274	39.8	<0.001	0.383
	Daylight × Electric light condition	Brightness	No	4	19.5	5.4	0.001	0.046
		Perceived color	No	4	17.9	5.5	<0.001	0.058
		Perceived uniformity	Yes	3.4	26.8	8.0	<0.001	0.062
		Coziness	No	4	5.60	2.4	0.057	0.019
		Liveliness	No	4	17.2	6.5	<0.001	0.053
		Tenseness	No	4	9.3	3.3	0.014	0.025
		Detachment	No	4	15.8	5.2	0.001	0.034
Effect of diffuse and direct daylight for electric light condition 4	Daylight	Brightness	No	3	31.3	8.6	<0.001	0.168
		Perceived color	No	3	12.0	3.4	0.023	0.068
		Perceived uniformity	No	3	3.8	1.6	0.191	0.022
		Coziness	Yes	2.4	18.2	6.9	0.001	0.119
		Liveliness	Yes	2.3	6.7	3.2	0.040	0.056
		Tenseness	Yes	2.4	6.7	3.9	0.020	0.049
		Detachment	Yes	2.1	35.3	12.1	<0.001	0.138

Fig. 7 Effect of daylight on the light and atmosphere attributes for all light conditions. Error bars represent the 95% confidence interval. Asterisks indicate the daylight conditions that differ significantly (P < 0.05) from the no-daylight condition for that particular light condition and attribute.

Table 7. Results of the mediation analyses for the total effect of the independent measure electric light condition on the dependent measures coziness (C), liveliness (L), tenseness (T), and detachment (D). Regression coefficients are given for all combinations of electric light condition. An asterisk indicates a significant effect.

			Regression coefficients (comparison of two electric light conditions)
Path	Independent measure	Dependent measure	1–2	1–3	1–4	1–5	2–3	2–4	2–5	3–4	3–5	4–5
c	Electric light	C	−1.45*	−2.24*	−1.70*	−1.05*	−0.78*	−0.51	0.41	0.27	1.19*	0.92*
		L	−0.72*	−1.69*	−2.23*	−2.23*	−0.98*	−1.52*	−1.52*	−0.54*	0.54	0.00
		T	0.67*	1.12*	0.84*	0.46	0.45	0.18	−0.22	−0.27	−0.66*	−0.39
		D	1.25*	1.72*	2.64*	2.26*	0.47	1.39*	1.01*	0.92*	0.54	0.38

*P < 0.005, with Bonferroni correction applied to account for the 10 comparisons.

Fig. 8 Schematic representation of the results of the total effect analyses (top) and mediation/moderation analyses (bottom). An arrow is drawn if for one of the electric light condition comparisons a significant effect was found. The type of arrow (as indicated in the legend) indicates the number of significant effects that were found, where darker and thicker arrows represent a higher number of significant effects.

Table 8. Overview of the a-path effects of the mediation and moderation analyses: influence of the independent measure electric light condition with and without the impact of the moderator daylight on the mediators (that is, brightness (B), perceived color (Pc), and perceived uniformity (Pu)). Regression coefficients are given for all combinations of the electric light conditions (that is, 1 versus 2, et cetera). An asterisk indicates a significant effect.

				Regression coefficients (comparison of two electric light conditions)
Path	Independent measure	Moderated by daylight	Dependent measure	1–2	1–3	1–4	1–5	2–3	2–4	2–5	3–4	3–5	4–5
a1	Electric light condition		B	1.35*	1.19*	0.96	0.62	−0.15	−0.38	−1.96*	−0.23	−1.81*	−1.58*
w1		Yes		−1.31	−1.54*	−0.54	−0.88	−0.23	0.77	0.42	1.00	0.65	−0.35
a2			Pc	3.38*	2.92*	2.54*	−0.35	−0.46	−0.85	−3.73*	−0.38	−3.27*	−0.29*
w2		Yes		−1.31*	−0.50	−1.00*	0.00	0.81	0.31	1.31*	−0.50	0.50	1.00
a3			Pu	0.31	2.73*	4.08*	2.92*	2.42*	3.77*	2.62*	1.35*	0.19	−1.15*
w3		Yes		0.50	−0.35	−0.62	−1.42*	−0.85	−1.12	−1.92*	−0.27	−1.08	−0.81

*P < 0.005, with Bonferroni correction applied to account for the 10 comparisons.

Table 9. Overview of the b-path effects of the mediation and moderation analyses: influence of the mediators (that is, brightness (B), perceived color (Pc), and perceived uniformity (Pu)) with and without the impact of the moderator daylight on the dependent measures coziness, liveliness, tenseness, and detachment. Regression coefficients are given for all combinations of the electric light conditions (that is, 1 versus 2, et cetera). An asterisk indicates a significant effect.

				Regression coefficients (comparison of two electric light conditions)
Path	Independent measure	Moderated by daylight	Dependent measure	1–2	1–3	1–4	1–5	2–3	2–4	2–5	3–4	3–5	4–5
b1	B		Coziness	0.14	0.05	0.29*	0.11	−0.01	0.26	0.01	0.17	−0.11	0.22
b2	Pc			−0.48*	−0.48*	−0.40*	−0.24	−0.51*	−0.45*	−0.29	−0.40*	−0.33	−0.26
b3	Pu			0.04	−0.01	0.21	0.02	0.03	0.16	0.05	0.11	0.05	0.11
w4	B	Yes		−0.03	0.01	−0.16	0.04	−0.12	−0.21	0.05	−0.26	−0.01	−0.19
w5	Pc	Yes		0.11	0.23	0.02	−0.13	0.30	0.05	−0.09	0.07	0.11	−0.08
w6	Pu	Yes		−0.18	−0.12	−0.30	−0.05	−0.14	−0.21	−0.07	−0.11	−0.03	−0.10
b1	B		Liveliness	0.31*	0.41*	0.42*	0.43*	0.34*	0.36*	0.33	0.48*	0.51*	0.51*
b2	Pc			−0.30	−0.27	−0.12	−0.11	−0.32	−0.13	−0.18	−0.15	−0.13	−0.03
b3	Pu			0.00	0.06	0.01	0.11	0.07	0.02	0.11	0.05	0.10	0.05
w4	B	Yes		0.038	−0.22	−0.21	−0.01	−0.12	0.10	0.24	−0.51*	−0.19	−0.17
w5	Pc	Yes		0.11	0.13	0.12	−0.02	0.15	−0.11	−0.15	−0.13	0.02	−0.19
w6	Pu	Yes		−0.09	−0.29	−0.17	−0.30	−0.16	−0.05	0.18	−0.10	−0.26	−0.19
b1	B		Tenseness	0.07	−0.01	−0.02	0.09	−0.03	−0.07	0.08	−0.18	−0.07	−0.19
b2	Pc			0.15	−0.10	0.05	0.03	0.13	0.13	0.32	−0.01	0.04	0.06
b3	Pu			−0.65*	−0.21	−0.16	−0.17	−0.36*	−0.28	−0.32	−0.08	−0.09	0.04
w4	B	Yes		−0.02	0.12	0.14	0.11	0.08	0.08	0.05	0.26	0.34	0.33
w5	Pc	Yes		0.08	0.18	0.30	0.18	0.00	0.15	−0.10	0.20	0.08	0.29
w6	Pu	Yes		0.41	0.07	−0.03	0.14	0.23	0.20	0.35	−0.04	0.05	0.08
b1	B		Detachment	0.30*	0.31*	0.25	0.16	0.25	0.19	0.02	0.21	0.02	−0.13
b2	Pc			0.09	0.00	0.04	0.39*	−0.20	−0.09	0.18	−0.15	0.01	0.14
b3	Pu			0.40	0.38*	0.57*	0.32	0.34*	0.48*	0.33	0.43*	0.33*	0.49*
w4	B	Yes		−0.11	−0.04	−0.01	0.07	−0.14	−0.03	0.11	−0.03	0.14	0.32
w5	Pc	Yes		0.19	0.39	0.24	−0.11	0.46	0.24	−0.01	0.38	0.23	−0.01
w6	Pu	Yes		−0.08	0.03	−0.06	−0.01	−0.01	0.93	−0.09	0.02	−0.01	−0.15

*P < 0.005, with Bonferroni correction applied to account for the 10 comparisons.

Table 10. Overview of the c′-path effects of the mediation and moderation analyses: influence of independent measure electric light condition with and without the impact of the moderator daylight on the dependent measures coziness (C), liveliness (L), tenseness (T), and detachment (D). Regression coefficients are given for all combinations of electric light condition (that is, 1 versus 2, et cetera). An asterisk indicates a significant effect.

				Regression coefficients (comparison of two electric light conditions)
Path	Independent measure	Moderated by daylight	Dependent measure	1–2	1–3	1–4	1–5	2–3	2–4	2–5	3–4	3–5	4–5
c′	Electric light condition		C	−0.26	−1.07	−2.88*	−1.46*	−1.05	−1.93*	−0.89	−0.59	−0.22	1.10
w7		yes		−0.07	−0.03	1.83*	0.86	0.43	−1.19	0.56	0.53	0.80	−0.52
c′			L	0.15	−1.23	−2.75*	−2.51*	−1.16*	−2.20*	−2.26	−1.27*	−0.60	0.98
w7		yes		−0.64	0.14	1.67	1.34	0.39	1.10	1.31	1.20*	0.39	−1.12
c′			T	0.15	2.22*	1.79	1.34	1.49*	1.55	2.44	−0.11	−0.34	−0.26
w7		yes		0.33	−0.78	−1.30	−0.82	−1.00	−1.71	−2.36	−0.49	−0.04	1.15
c′			D	1.11	0.83	1.03	2.02*	−0.61	−0.07	0.64	0.88	0.48	−0.19
w7		yes		−1.43	−1.44	−0.89	−0.26	0.71	0.99	1.06	−0.06	1.22	0.83

*P < 0.005, with Bonferroni correction applied to account for the 10 comparisons.

Stokkermans MGM, Chen Y, Murdoch MJ, Vogels IMLC, Heynderickx IEJ. 2015. Effect of daylight on atmosphere perception: comparison of a real space and visualizations. Proceedings Volume 9394, Human Vision and Electronic Imaging XX, SPIE/IS&T Electronic Imaging, 8–12 February 2015, San Francisco (CA).

 Google Scholar