Assessing landscape aesthetic values: Do clouds in photographs influence people’s preferences?

Erich Tasser; Alexandros A Lavdas; Uta Schirpke

doi:10.1371/journal.pone.0288424

. 2023 Jul 28;18(7):e0288424. doi: 10.1371/journal.pone.0288424

Assessing landscape aesthetic values: Do clouds in photographs influence people’s preferences?

Erich Tasser ¹, Alexandros A Lavdas ^2,³, Uta Schirpke ^1,^4,^5,^*

Editor: Chaohai Shen⁶

PMCID: PMC10381034 PMID: 37506121

Abstract

Photo-based surveys are widely applied to elicit landscape preferences and to assess cultural ecosystem services. Variations in weather and light conditions can potentially alter people’s preferences, as sunny landscapes are more positively perceived than those under inclement weather conditions. To assure comparability across pictures, studies usually include photographs taken at sunny days (i.e., blue sky). However, the influence of clouds in sunny landscapes on people’s preferences has been rarely considered, although color contrasts between clouds and the blue sky may attract people’s attention. This study therefore aimed to assess the effects of clouds in landscape photographs on people’s preferences by (1) examining differences in preference between pairs of landscape photographs (i.e., with clouds and without clouds), and (2) explaining variations through variables from eye-tracking simulation, photo content analysis, and Geographic Information System (GIS)-based analysis. Our results indicate no significant differences in preferences between pictures with and without clouds when the pictures with clouds contained a proportion of sky around 22% and a cloud cover of about 39%. However, a higher proportion of sky positively influenced landscape preferences, while a higher proportion of clouds, especially in combination with a lower proportion of sky, had negative effects. These findings suggest that landscape preference studies should pay attention not only to the appearance of the sky in terms of cloudiness, but they also should control the proportion of sky across different pictures to obtain comparable results. Future research should address limitations regarding the transferability of our findings to other types of landscapes and regarding potential differences in perceptions between respondents with different socio-cultural characteristics. Moreover, landscape preferences under changing weather conditions or different cloud types as well as diurnal and seasonal changes should be further explored.

Introduction

Landscapes provide a wide range of cultural ecosystem services such as recreational, aesthetic, educational, and spiritual values [1]. Aesthetic values are one of the most appreciated cultural ecosystem services [2–4], associated with a wide range of subjective benefits to people such as spirituality, knowledge, identity, and health [5]. Also, many aspects of objective physical and psychological health benefits of exposure to natural landscapes have been well documented, from stress reduction to improved recovery from surgery [6–8]. Appealing landscapes also can provide economic benefits [9], as they are strongly linked to outdoor recreation [10, 11]. Furthermore, high aesthetic quality of landscapes can support biodiversity conservation efforts [12, 13]. Due to rapid landscape transformations caused by global change pressures, decision-makers need reliable information on people’s landscape preferences to develop sustainable management strategies and, at the same time, to preserve or restore attractive landscapes [14, 15].

Aesthetic landscape values arise from the interaction of people with the biophysical characteristics of the natural environment [16], and are usually assessed applying two distinct approaches. While expert-based approaches analyze visual landscape properties through quantitative information such as indicators describing landscape patterns [17, 18], perception-based approaches consider subjective values associated to the landscape [19]. Despite the subjectivity of people’s perceptions, perception-based approaches are considered to provide reliable results for the general public [20]. Nevertheless, landscape preferences of individuals may be strongly heterogeneous [21] and are often influenced by socio-cultural characteristics [17, 22–27]. In perception-based approaches, questionnaires eliciting landscape preferences on the basis of photographs are the most frequently used methodology [22–25, 28, 29]. These studies usually ask the participants to evaluate the photographs using rating scales, e.g., ranging from 1–5 [17, 27], 1–6 [30], or 1–10 [31, 32], or applying discrete choice experiments [23, 33, 34]. Photographs are considered reliable stimuli to gather people’s preferences, comparable to on-site surveys [35]. A big advantage of using photographs is that it allows researchers to involve a high number of respondents, which enables the generation of a large database and a strong data analysis [36]. Moreover, landscape pictures can be manipulated or digitally designed to control the image content for assessing specific aspects [37–39].

To guarantee comparability across different photographs, studies usually use pictures taken at sunny days, i.e., blue sky [24, 25, 30]. Weather conditions in the photographs can alter people’s preferences, as the degree of cloudiness and the proportion of cloud cover in the sky affect the illumination of the landscape, i.e., the landscape appears darker with less contrast and details and colors are less discernable compared to clear weather conditions [40]. Thus, inclement weather conditions can moderate preferences for waterscapes [41] and associated benefits such as happiness [42]. In contrast, sunrise, rainbows, and sunset can significantly increase preferences compared to blue sky [43]. In arts and photography, which have been intensively concerned with the aesthetics and emotional effects of appearance of the sky, clouds are often used to convey certain moods and messages [44, 45]. For example, since the sky symbolized the supernatural instance, light and rays were considered symbols of the divine, thunderclouds and lightning as symbols of God’s punishment [46]. The sky and cloud patterns were also used by artists to depict their emotions and inner states, which gave rise to "mood painting", where the weather in combination with the landscape was used as a symbol or allegory for feelings [44, 47]. The importance of clouds, also as mood-setters, was underpinned by the introduction of the term cloudscapes [44]. The appearance of different cloud types ranges from small to big puffy shapes (e.g., low-altitude cumulus or nimbostratus clouds), to wavy shape (e.g., mid-level altitude altocumulus clouds) and patchy or wispy shapes (e.g., high altitude cirrus, cirrostratus) [48]. Landscape photography also took up the subject of clouds to influence the aesthetics and emotional impact of photographs [49]. Accordingly, photographs posted in social media often reflect weather phenomena capturing the atmosphere (mood) of a place and aesthetic experiences of landscapes [50, 51].

For these reasons, perception-based surveys, using photographs to gather landscape preferences, usually include only pictures taken at sunny days. However, it remains unclear how the presence of clouds in those photographs also alters preferences. We hypothesize that clouds in landscape pictures may influence the stated preferences due to color contrasts between clouds and the blue sky, as color contrasts, hue variations, and edges attract people’s attention when looking at landscape pictures [29, 52, 53]. Moreover, the preferences may be influenced by emotions and moods that clouds can bring to a landscape [45, 49, 51]. Thus, this study aimed to assess the effect of clouds in landscape photographs on people’s preferences. The analysis is guided by two research questions: (R1) Is there a difference in preference between a landscape photograph with clouds and without clouds? (R2) If differences in preference occur, how can these variations be explained? Our findings provide valuable information for designing and evaluating photo-based assessments in landscape research.

Materials and methods

Study design

To answer the two guiding research questions R1 and R2, we applied different analysis steps based on landscape photographs (Fig 1, see following sections). The original photographs (A) had different degrees of cloud cover, and a cloud-free version (B) was created for each photograph. People’s preferences were collected in an online survey. Predictor variables were derived from three methodological approaches, including eye-tracking simulation, photo content analysis, and Geographic Information System (GIS)-based analysis [53]. Finally, statistical analyses were carried out to test for differences between picture pairs (R1) and to explain variations in preferences scores (R2).

Photo-based survey

To evaluate the influence of clouds on landscape preferences, we used landscape photographs from two previous studies [25, 54]. The photographs were taken in various locations of the Central European Alps and depicted different mountain landscapes that were typical for this area, including rural settlement areas, agricultural areas, alpine pastures, and high mountain landscapes. All photographs showed a 360° panorama and were taken at normal eye level (approx. 1.5–1.7 m) during sunny summer days. When taking the pictures, attention was paid that the horizon did not extend over the center of the picture [55]. Despite a blue sky, the photographs contained variations in cloud patterns, ranging from almost no clouds to a sky largely covered by clouds, but no threatening cloud atmosphere.

From this pool of 49 landscape photographs [25, 54], we selected 29 photographs with varying cloud cover (S1 Fig) to not overload the respondents with a too high number of pictures. The sky of all original photographs (A) was manipulated by removing the clouds and recoloring these areas to derive a version of the photograph without clouds (B), using Adobe Photoshop image processing software (Adobe Photoshop CS6, Adobe Systems Inc, CA, USA). The manipulation only altered the appearance of the sky, and no landscape features were changed (S1 Fig).

All original (A) and manipulated (B) photographs (n = 58) were arranged in an online questionnaire using LimeSurvey (LimeSurvey GmbH, Hamburg, Germany; www.limesurvey.org/). The questionnaire was prepared in German and contained three sections. In the first section, participants were informed about the scope of the survey and that the participation in the survey was voluntary and anonymous (see ethics statement below). In the second section, we asked the respondents to express their preference for each photograph on a 10-point Likert scale (1 = ‘I don’t like it at all’ to 10 = ‘I like it very much’). There is no consensus among landscape preference studies on the most appropriate rating scale. We used an even scale, because we wanted the respondents to decide whether their preference for an image is more positive or negative. Moreover, response scales from 7 to 10 are most reliable, and respondents prefer the 10-point scale the most [56]. All participants assessed all 58 photographs, but the order of the pictures was randomly arranged for each respondent to avoid priming effects [57]. In the third section, participants could also indicate socio-demographic information such as gender, age group, language group, place of residence (city, country), school education and field of work.

The online survey was carried out between March and July 2022, using a convenience sampling approach. We invited students in ecology of the University of Innsbruck (Austria) and students in physical geography of the Ludwig Maximilian University in Munich (Germany) during lectures on applied ecology and on environmental management to participate in the survey by providing a link to the questionnaire. We also asked them to forward the link to friends and relatives (particularly parents/grandparents, but no minors). The participants received neither a compensation nor other incentives. They completed the questionnaire in an unsupervised setting using a digital device and the responses were saved in a database. We estimated that the time to fill it out was about 10 minutes. There were no time restrictions in viewing and rating the photographs. The participants could also pause the survey by saving their responses and continue at a later time.

To ensure a minimum sample size for statistical analysis, i.e., to set up a significant regression model, we determined the minimum sample size through statistical power analysis [58]. With a determination coefficient of R² = 0.7, a statistical power of 0.9 and a significance level of α = 0.01, we needed a sample size of at least 68 valid responses. In total, 176 people opened the questionnaire and 124 participants completed it. The participants were mostly female (70.3%), under 25 years old (66.4%), and residents in North Tyrol (69.2%). Most of them had a highschool degree (69.7%) and were not (yet) employed/still in eduction (63.3%). For full demographic details, see S1 Table.

Due to the convenience sampling approach, we first evaluated the representativeness of the sample, i.e., whether our results were specific for this group of respondents or whether they matched the preferences of the general public. For all pictures, we first calculated mean preference scores and then compared those of the original photographs in our study to the results of the two former surveys, from which we selected the photographs [25, 54]. Both studies gathered landscape preferences in a comparable way between 2016 and 2019 in five different study areas in the Central Alps. Both studies used a stratified sampling approach based on the demographic distribution in the study areas, i.e., considering key socio-demographic characteristics such as gender, age, origin (residents, tourists), place of living (village, city), and native language (German, Italian). The two surveys, comprising 967 respondents [25] and 384 respondents [54], can therefore be considered to be representative for the general public of the Central Alps. In both surveys, differences between socio-demographic groups in preference scores were generally very small, although some significant differences occurred, mostly between people of different language, age, and origin [25]. After aligning the preference scores among the three surveys using the linear stretch method [59], we calculated the degree of correlation applying linear regression analysis. Since we obtained a high correlation (R² = 0.842; p<0.001; S2 Fig), we can assume that our sample also represents well enough general preferences.

Ethics statement

The participants were informed before filling out the questionnaire that participation in the survey was voluntary and anonymous and that no personal data would be requested that would allow the person to be identified. Access to the questionnaire was only granted if the person actively requested it and gave his/her the consent by checking the box before starting the questionnaire. The information for this study was collected in a manner, in which the institutions did not collect personal data that would allow identification of individual human participants. According to the institutional and national rules, this study did not need approval of an ethics committee, as it was not a clinical study treating sensitive personal data nor health-related information. Only adults were invited to participate. The study was in line with the principles established by national and international regulations, including the Declaration of Helsinki (64th WMA General Assembly, Fortaleza, Brazil, Oct 2013) and the Code of Ethics.

Collection of variables

To explain variations in preferences, we prepared a set of predictor variables for each picture (Tables 1 and S2–S4), using three different methodological approaches [53] as described in the following.

Table 1. Variables derived for all pictures using three different analysis approaches [53].

Analysis approach	Variables	Unit	Description
Eye-tracking simulation	Hotspot area	%	Total estimated area of the hotspots in the photo
	Top-hotspot area	%	Estimated area of the top-hotspot
	Lowest hotspot probability	%	Probability of initial eye-tracking movement of the hotspot with the lowest probability
	Sky within hotspots	%	Estimated area of sky horizon within the hotspots
	Intensity of visual elements	%	Estimated mean contribution of intensity of visual elements (i.e., luminance contrast, brightness, black/white contrast) within the hotspots to the overall probability of the hotspot
	Red-green color contrast	%	Estimated mean contribution of red-green color contrast of visual elements within the hotspots to the overall probability of the hotspot
	Blue-yellow color contrast	%	Estimated mean contribution of blue-yellow color contrast of visual elements within the hotspots to the overall probability of the hotspot
Photo content analysis	Sky	%	Estimated area of sky within the photo
	Clouds	%	Estimated area of clouds within the sky
	Artificial elements	%	Estimated area of clearly recognizable artificial elements within the photo (e.g., street, street signs, cars, fences)
	Percentage near zone	%	Estimated near zone (0–60 m) within the photo
	Percentage middle zone	%	Estimated middle zone (0.06–1.5 km) within the photo
	Percentage far zone	%	Estimated far zone (>1.5 km) within the photo
GIS-based analysis	Patch density	n ha^-1	Patch density
	Largest patch index	index	Largest patch index
	Area-weighted mean of patch area	ha	Area-weighted mean patch area distribution
	Area near zone	ha	Total visible area of near zone (<1.5 km)
	Area middle zone	ha	Total visible area of middle zone (1.5–10 km)

Open in a new tab

(1) Eye-tracking simulation

Eye-tracking simulation software 3M-VAS (3M™ Visual Attention Software, St. Paul/Minnesota, USA; https://vas.3m.com) was used to identify hotspots of initial eye-tracking movements on the pictures. 3M-VAS uses an artificial intelligence (AI) algorithm that is based on 30 years of lab-based eye-tracking research to simulate human pre-attentive processing in vision [60]. Based on five visual elements that are recognized to attract human attention (edges, intensity, red-green color contrast, blue-yellow color contrast, and faces), the algorithm predicts what people would be focusing their eyes on before the conscious brain can react, i.e., in the initial 3–5 seconds of looking at an image [60]. 3M-VAS claims to be as reliable and effective as eye tracking hardware, predicting pre-attentive vision with around 92% accuracy [60]. The software also overcomes issues related to repeatability of real eye tracking experiments and allows direct comparison of the outputs [61].

To obtain the simulated eye-tracking results, we uploaded the pictures to the 3M-VAS website, which were then processed under a minute. The pictures were standardized size (2300x360 pixel) and resolution (300 dpi), as used in the questionnaires. We scanned all pictures using the category “Other,” representing the most general and unbiased modality [62]. The analysis report of 3M-VAS included (1) a heatmap, (2) the hotspots that were derived from the heatmap, specifying the probability that a person will look somewhere within the hotspot areas within the first 3–5 seconds, (3) a gaze sequence indicating the most probable viewing order of the 4 most-likely seen hotspots, and (4) a report of visual elements (i.e., edges, intensity, red-green color contrast, blue-yellow color contrast, faces) indicating their contribution to the overall probability, from which the heatmap was derived [60]. All output images had a size of 1024x160 pixel with a resolution of 96 dpi. As 3M-VAS analyzes each picture individually, the results are independent and comparable [61]. To estimate different quantitative information for each pictures, we used the hotspots as well as the report on the visual elements (Tables 1 and S2). From the hotpspots (S3 Fig), we derived, for example, the area of the hotspots, probabilities of initially viewing the hotspots, and the area of sky within the hotspots. To systematically analyze all hotspots, we overlaid a grid with 3x9 cells over the 3M-VAS output (S4 Fig). In each grid cell, we counted the coverage (%) of the hotspots and summarized the percentages across all cells. Furthermore, the contribution of the visual elements (edges, intensity, red-green color contrast, blue-yellow color contrast, faces) was estimated for the hotspots. For this purpose, we estimated the spatial coverage of the edges within each hotspot from 0 to 100% (S5 Fig). For the other elements, we estimated the level of the color intensity, ranging from no value (black = 0%) to high values (white = 100%) (S5 Fig). Finally, the values for each visual element were averaged across all hotspots to obtain their contribution to the probability of the hotspots in each photograph. All estimations were carried out by one researcher to assure comparability of the estimations. For further details, see Schirpke et al. [53].

(2) Photo content analysis

In the photo content analysis, we estimated the structural composition of the pictures (S3 Table) by visual analysis, including the percentage of sky and cloud cover. Using the same grid with 3x9 cells (S4 Fig), the area of landscape features was estimated by one researcher counting the coverage (%) of the features in each cell and summarizing the percentages across all cells. The same person also estimated the composition of the landscape in terms of distance zones (i.e., near zone, middle zone, far zone; S4 Fig). In the near zone (0–60 m), individual landscape features (e.g., trees, buildings) could be clearly identified [63]. The middle zone extended to about 1.5 km from the photo location and individual landscape features were still clearly discernible. In the far zone, landscape features were not visible and only different land cover types such as grassland, forest, settlement areas were distinguishable.

(3) GIS-based analysis

GIS-based analysis was applied to quantify the composition and configuration of the landscape that was depicted on the photographs through landscape metrics [64]. In a first step, viewsheds up to 50 km [65] were calculated from the photo locations based on a digital surface model (DSM). For each cell of the DSM, viewshed analysis determines whether the cell is within the observer’s line-of-sight or not [66], i.e., which areas on the map are visible and which areas are hidden by vegetation, buildings, or mountains. In a second step, the visible area was overlaid with land-use/cover maps, while non-visible areas were excluded from further analysis. In this way, we aligned the content of the map with the content of the photographs. To account for changes in discernibility of landscape features (see above), we used different habitat and land cover maps as well as different spatial resolution for different distance zones (S4 Table). Finally, we composed the resulting maps of the different zones into one single map. Spatial analysis and map preparation was done in ArcGIS 10.4^TM (ESRI, Redlands, CA, USA). The final map was used to calculate 12 landscape metrics (S5 Table) with FRAGSTATS Version 4.2 [64]. For full details, see Schirpke et al. [67].

Cloud types and spatial arrangement

In addition to the predictor variables, we determined the cloud type of each original photograph. Furthermore, the photo composition in terms of spatial arrangement, e.g., position of horizon, can influence people’s preferences [55, 68]. Based on the Rule of Thirds, as a simplification of the golden section or golden ratio (Phi, ϕ), we divided the picture into thirds both horizontally and vertically [69, 70]. This grid with 3x3 cells (S6 Fig) was overlaid over the hotspots to determine a spatial shift between pictures for which the preference scores significantly differed between the original picture with clouds and the manipulated picture without clouds.

Statistical analyses

To assess differences in the preference scores between the original landscape photograph with clouds (A) and the manipulated picture without clouds (B), we performed statistical comparisons using a bootstrap paired samples t-tests with 10,000 iterations. Significant differences were set as p < 0.05. We calculated mean values of the variables sky and clouds across picture pairs with decreased (A>B), not changed (A = B), and increased (A<B) preference score. Statistically significant differences were tested using the least significant difference (LSD) post hoc test.

To explain variations in landscape preferences between the picture pairs, we applied backward stepwise linear regression. The backward stepwise approach starts from a full model with all predictors and successively reduces them to obtain a model that best explains the data, while solving problems of multicollinearity and overfitting [71]. We used the difference in preference score (B-A) between the manipulated picture without clouds (B) and the original landscape photograph with clouds (A) as dependent variable and all variables from the eye-tracking simulation, photo content analysis, and GIS-based analysis as independent predictor variables (S6 Table). We removed collinear variables to avoid overfitting and to support the interpretability of the model by first screening for multicollinearity among the predictor variables. We calculated the variance inflation factor (VIF) values for each variable and removed variables with a VIF <10. Then, we applied a backward stepwise linear regression routine to further diminish the number of variables [72] and iteratively excluded variables with low predicting performance until no further improvement was possible (Table 1). To assess the level of variability explained by the model, we used the difference between R² and the adjusted R², as the adjusted R² is less influenced by overfitting than R² [73]. Through analysis of variance (ANOVA) or t-tests, validity, quality, and significance of the coefficients were evaluated. The reliability of the regression was determined using the Shapiro-Wilk test (residual normality), the Durbin-Watson test (autocorrelation of the variables), the Breusch-Pagan test (homoscedasticity) and the VIF statistics (multicollinearity effect). All preconditions for the reliability of the regression using were fulfilled (Shapiro-Wilk test for residual normality: p = 0.503; Durbin-Watson test for autocorrelation of the variables: 2.143), Breusch-Pagan test for homoscedasticity: p = 0.326) and the VIF statistics (multicollinearity effect, see Table 2). All statistical analyses were performed in SPSS Statistics (IBM SPSS 27, NY, USA).

Table 2. Result of the multiple linear regression, using the difference in preference score (B-A) between the manipulated picture without clouds (B) and the original landscape photograph with clouds (A) as dependent variable.

Positive differences indicate that the photo without clouds is preferred over the photo with clouds, negative values indicate the opposite result. Accordingly, a negative regression coefficient B indicates that this variable has a positive effect on the preference score of photos with clouds. Only predictors with variance inflation factor (VIF) <10 during collinearity diagnostics are included.

Variables	Non standardized coefficient		Standardized coefficient Beta	T	Sig.	VIF
Variables	Regression coefficient B	SD	Standardized coefficient Beta	T	Sig.	VIF
(Constant)	1.755	0.298		50.893	<0.001
Hotspot area (%)	-0.026	0.005	-0.583	-40.864	<0.001	2.940
Top-hotspot area (%)	0.085	0.015	0.848	50.713	<0.001	4.514
Lowest hotspot probability (%)	-0.011	0.004	-0.462	-30.046	0.012	4.713
Sky within hotspots (%)	-0.026	0.005	-0.911	-50.259	<0.001	6.152
Intensity of visual elements (%)	0.005	0.003	0.274	20.015	0.072	3.800
Red-green color contrast (%)	-0.013	0.002	-0.909	-50.421	<0.001	5.757
Blue-yellow color contrast (%)	0.008	0.003	0.424	30.075	0.012	3.902
Sky (%)	-0.029	0.003	-10.593	-100.572	<0.001	4.652
Clouds (%)	0.003	0.001	0.353	30.051	0.012	2.750
Artificial elements (%)	-0.011	0.002	-0.602	-50.177	<0.001	2.771
Percentage near zone (%)	0.007	0.003	0.299	20.308	0.044	3.438
Percentage middle zone (%)	0.010	0.003	0.763	30.918	0.003	7.763
Percentage far zone (%)	-0.015	0.005	-0.376	-20.970	0.014	3.277
Patch density (n ha^-1)	-0.093	0.016	-0.910	-50.856	<0.001	4.950
Largest patch index (index)	-0.004	0.002	-0.191	-10.860	0.093	2.164
Area-weighted mean of patch area (ha)	0.008	0.001	0.838	60.078	<0.001	3.893
Area near zone (ha)	-0.001	0.000	-0.293	-20.767	0.020	2.295
Area middle zone (ha)	0.000	0.000	-0.958	-50.162	<0.001	7.061

Open in a new tab

Results

Comparing the preference scores between the original landscape photograph with clouds and the manipulated picture without clouds, landscape preferences did not significantly differ in about half of the cases (n = 15). For fourteen picture pairs, differences in preference scores were statistically significant. In nine cases, preference scores were lower with clouds, while the removal of clouds increased the preference scores in five cases (Fig 2 and S7 Table).

Comparing the mean proportion of sky and clouds of the picture pairs with decreased (A>B), not changed (A = B) and increased (A<B) preference score revealed opposing effects of cloud removal on preference scores (Fig 3 and S8 Table). The removal of clouds in photographs with a high proportion of sky (34%) and few clouds (18%) led to a decrease in preference score (A>B). In contrast, if the original photograph contained a low proportion of sky (16%) and a high degree of cloud cover (44%), preference scores were higher for the manipulated picture (A<B). For pictures with a proportion of sky covering around 22% of the photograph and a low level of cloud cover (about 39%), the removal of the clouds had no effect on the preference score (A = B). In addition, the hotspots were located more towards the sides than in the center in the more attractive image (S9 Table). In 10 of 14 cases, there was a significant shift of hotspots from the center to the top or bottom corners of the photos with cloud removal (pictures 3, 5, 6, 9, 11, 14, 22, 26, 28, 29). In 3 cases, hotspots did not shift (pictures 2, 13, 24), and in only one case, hotspots moved from the sides to the center (picture 20). Pictures with single, smaller clouds (mostly cumulus or cirrostratus clouds) obtained higher preference scores, while in pictures with lower preference due to a high cloud cover, nimbostratus clouds occurred more often (S10 Table).

Fig 3 — Values represent mean values of the variables sky and clouds across picture pairs with decreased (A>B), not changed (A = B), and increased (A<B) preference score. A: Original picture with clouds, B: manipulated picture without clouds. Different colored letters (a, b) near the symbols indicate statistically significant differences (LSD post hoc test, p < 0.05).

The difference in preferences between picture pairs could be explained by seven hotspot characteristics, six photo content variables, and six landscape metrics (Table 2). The stepwise linear regression analysis led to a model explanation of 95.1% of the differences in preferences (ANOVA p<0.001; R² = 0.951; adjust. R² = 0.863). The variables derived from eye-tracking simulation indicated that higher rated pictures with clouds comprised on average a higher percentage of hotspot areas, also containing a part of the horizon or sky (Table 2) than the modified picture without clouds. The hotspots were often characterized by high red-green contrast (see Example 1, Fig 4). In contrast, preferences for pictures without clouds were often related to the presence of a large top hotspot, a high intensity of visual elements, and high blue-yellow contrasts (Table 2).

Fig 4 — For each example, above are shown the original picture with clouds (A; photographs by Eurac research) and the manipulated picture without clouds (B), the heatmap indicating the probability that areas are seen within the first 3–5 seconds (middle), and the hotspots derived from the heatmap with the probability that a person will look somewhere within the hotspot areas within the first 3–5 seconds (below). Hotspots are delimited with a red line a probability > = 80% and with a yellow line for a probability < 80%.

In terms of photo content, landscape preferences for pictures with clouds were related to a higher proportion of sky, provided that the degree of cloud cover was not too high (Table 2). In these pictures, a high proportion of background area had a positive influence on landscape preferences, while artificial elements generally negatively influenced preferences. Contrastingly, high proportions of foreground and middle-ground reduced preferences for non-cloudy landscapes (see Example 2, Fig 4).

Landscape patterns also had different influence on landscape preferences for pictures with and without clouds (Table 2). For pictures with clouds, a high patch density with a large central land cover type (Largest patch index) and a high proportion of the foreground was positively related to landscape preferences. In contrast, a more homogeneous landscape composition (Area-weighted mean of patch area) and a higher proportion of the middle ground increased landscape preferences for pictures without clouds.

Discussion

Perception of sky and clouds

In this study, the hypothesis that clouds in landscape photographs can influence the stated preferences due to color contrasts between clouds and the blue sky was partially confirmed. Significant differences in preferences were dependent on the percentage of cloud cover and proportion of the sky. Our findings indicate that a high proportion of clouds leads to lower preference scores compared to a sky without clouds. In these pictures, the percentage of clouds in the sky was on average 44% and the sky sometimes appeared darker, for example, due to nimbostratus clouds. This aligns with studies showing that inclement weather conditions or ephemeral phenomena such as thunderstorms lead to lower landscape preferences [41, 43]. In contrast, our results also indicate that single, smaller clouds (e.g., cumulus clouds) positively influenced preference scores. The visibility of the fractal outline of clouds [74] in these cases, helped by its clear contrast to the sky, might be part of the explanation for this. Such a finding fits well with our current understanding of the preferential perception of geometrical structures based on a hierarchy of scales, as in the nested symmetries of a fern leaf, for example, see Taylor et al. [75] as a review. Functional magnetic resonance imaging (fMRI) work has revealed the potential neural correlates of this phenomenon, as the processing of recursive forms has been shown to recruit different resources to the processing of non-recursive forms. Remarkably, it has been shown to recruit the default mode network, a functional brain network known to be involved in the processing of internal information [76, 77], thus implying a less computation-intensive processing of such shapes. In the pictures with higher preference, the eye-tracking hotspots were also often located on such small but high-contrast clouds or smaller cloud groups. Similarly, the study by Smalley and White [43] demonstrates that the appearance of the sky during sunrise or sunset increases perceived beauty of landscapes.

Regarding the composition of the pictures, painting and photography indicate clear rules, which can be related to our findings. The golden section or golden ratio (Phi, ϕ), referring to the positioning of the main subject in the picture or photograph, is considered the ideal ratio value determining aesthetic preferences, although controversially discussed [55, 78]. It is based on the fact that the human brain perceives images as harmonious if this rule is respected. As a simplification of the golden ratio, the Rule of Thirds is also central in design, films, and paintings, as well as in photography [69, 70]. Visual preferences have been found for photographs that depict about one third sky and two thirds foreground, if the foreground is interesting and striking [55]. In contrast, if the foreground is inconspicuous and/or there is a beautiful sky with clouds, then the rule of thirds is reversed, i.e., about one third foreground and two thirds sky [55]. Regarding our results, this means that landscape preferences may be higher if the proportion of sky reaches up to one third of the picture compared to another photograph that has considerably less proportion of sky. The vertical alignment was less important in our study, as the landscape did not change between the picture pairs. Nevertheless, the hotspots in the more attractive pair were located more towards the sides due to the lack or presence of clouds.

Implications for landscape research

Our findings also provide insights for landscape research. Many landscape preference studies using photographs as stimuli included pictures with blue or clear sky [24, 25, 30, 79], which is still recommendable in the light of our findings. There are also no constraints to include pictures with a proportion of sky around 22% and a cloud cover of about 39%. Nevertheless, for specific applications aiming at supporting real-world decisions, for example, evaluating aesthetic impacts of wind power farms, different weather conditions are often included [39, 80]. For this purpose, eye-tracking simulation may be a useful tool to detect changes in contrasts and to test different appearance of the infrastructures to reduce their visibility and thus, their impact on landscape aesthetic values. More generally, eye-tracking simulation may support visual impact assessment in an efficient way. For example, it could improve saliency mapping, which has been recently applied to evaluate visual impacts for wind and solar infrastructure in vineyard landscapes [81]. However, while there is a high correlation between eye-tracking data and saliency maps, it still needs be tested whether eye-tracking simulation provides similar results.

With regard to the composition of the picture (i.e., percentage of sky), our findings corroborate those of Svobodova et al. [55]. This aspect has been considered less often in landscape research, although other studies describe the influence of the characteristics of the horizon line on people’s perceptions, e.g., referring to the presence of mountains [29], in urban settings [82], and in relation with fractal characteristics [83, 84]. To reduce the influence of the sky on landscape preferences, our findings suggest that studies need to pay attention to the percentage of sky. Avoiding variations in percentage of sky will therefore produce a better comparability across different landscapes. For photographs taken in landscapes with complex horizon lines (e.g., mountains, differing vegetation height, and tall buildings) this may be more difficult but can be estimated, as done in this study. Using manipulated or computer-generated pictures, the percentage of sky can and should be rigorously controlled.

Limitations and future research

In general, our findings contribute to a deeper understanding of the influence of clouds in photographs used for eliciting landscape preferences in mountain regions. However, these results may not be transferrable to other landscapes. For example, the importance of sky and clouds may be differently evaluated in urban environments [43, 83, 85], in flat or hilly landscapes [86, 87]. Similarly, our results suggest that the influence of clouds was less relevant in pictures with diverse landscape patterns. It is therefore important to include different types of landscapes in future research. Moreover, the manipulations of the sky in some pictures did not completely reproduce natural conditions, as the sky sometimes had little texture and the lighting of the foreground landscape did not always match the lighting provided by a blue sky. It should therefore paid attention in future studies that the manipulations are not noticeable by the participants to avoid a potential negative influence on the preference scores of the manipulated pictures.

While we can assume that our results represented well enough general preferences, as suggested by the high correlation of our results with those of two former representative surveys [25, 54], future studies should use a stratified sampling approach, requiring also a higher number of respondents, to assess potential differences across socio-cultural groups. Regarding the composition of the landscape, differences in preferences were found, for example, between lay people and experts [17, 26, 27], and between residents and tourists [25, 26]. Preferences may also depend on the educational level [27], environmental values [30], or the connection of respondents with the landscape [88]. However, it still needs to be verified whether such differences also occur in terms of clouds.

While the above indicated recommendations apply for research focusing on general differences in perceptions across different landscapes, a less studied research field opens up in examining temporary changes, such as seasonal shifts [22, 89, 90] and diurnal changes [43, 85]. For example, different weather and light conditions can moderate preferences for a specific landscape [43]. The influence of such temporary states of landscapes on cultural ecosystem services is still less explored, despite studies addressing specific ‘events,’ such as the cherry blossoms [91], wildflower blooms [92], or the occurrence of rainbows [43, 93]. It may therefore not enough to use ‘blue-sky’ pictures to evaluate the overall aesthetic value of a location but to account for typical weather conditions and seasonal changes. A deeper understanding of the relationships between landscapes and human well-being may also allow decision-making and planning to better scope with global change pressures and integrate cultural ecosystem services into management plans and policies [15].

Conclusions

Our findings suggest that the presence of clouds in sunny landscape photographs can alter landscape preferences. A high proportion of clouds, in particular in combination with a low proportion of sky, can lead to lower preference scores compared to a sky without clouds. In contrast, a high proportion of sky with single, small clouds can increase preference scores. Our findings therefore have important implications for landscape preference studies using photographs as stimuli, suggesting that such studies need to pay attention to variations in the percentage of sky and cloud patterns.

However, as we focused on mountain landscapes, further research is needed to verify our findings in other landscapes, such as flat and hilly landscapes or in urban environments. It would be also necessary to apply a stratified sampling approach to assess potential differences in perceptions between respondents with different socio-cultural characteristics. We focused on the appearance of the sky in photographs taken at sunny days, as recommended to ensure comparability of the stated preferences across pictures. Yet, examining variations in appearance of the same landscape, for example, under different weather conditions, different times of the day, or different seasonal stages, may improve the understanding of the relationships between landscapes and human well-being.

Supporting information

S1 Fig

Original photographs with clouds (A) and manipulated pictures without clouds (B). Own photographs.

(PDF)

Click here for additional data file.^{(8.5MB, pdf)}

S2 Fig. Correlation between the results from this study (preference of sample represents) and the general preferences from the studies by Schirpke et al. (2016) and Forer (2020).

(TIF)

Click here for additional data file.^{(145.2KB, tif)}

S3 Fig

Hotspots in original photographs (A) and manipulated pictures (B) derived from eye-tracking simulation using 3M-VAS. Own photographs.

(PDF)

Click here for additional data file.^{(3.2MB, pdf)}

S4 Fig. Estimation of landscape composition in terms of distance zones.

Blue = near zone (< 60 m), red = middle zone (> 60 m– 1.5 km), yellow = far zone (>1.5 km), and auxiliary grids to determine the area fraction of distance zones and landscape features. Own photograph.

(TIF)

Click here for additional data file.^{(1.9MB, tif)}

S5 Fig. Contribution of visual elements to the overall probability that areas are seen within the first 3–5 seconds.

The yellow circles indicate the hotspots identified in 3M-VAS. Within each hotspot, the importance of each visual element was estimated on a scale of from 0 to 100%. For example, for hotspot no. 3, edges take up 50% of the area, intensity has mostly medium to high values (grey to light grey areas), red-green contrast values are low (dark grey patterns) and blue-yellow contrasts as well as no values for faces (black) are missing. Estimated contributions are 50% for edges, 40% for intensity, 5% for red-green color contrast and 0% for blue-yellow color contrast and faces. Own photograph.

(TIF)

Click here for additional data file.^{(587.4KB, tif)}

S6 Fig. Grid with 3x3 cells according to the Rule of Thirds.

This grid was used to determine a spatial shift between pictures, for which the preference scores significantly differed between the original picture with clouds and the manipulated picture without clouds. Own photograph.

(TIF)

Click here for additional data file.^{(774.1KB, tif)}

S1 Table. Demographic information of the survey participants (n = 124).

This information was not required to complete the questionnaire, and some participants did not at all or only partly indicate demographic details.

(DOCX)

Click here for additional data file.^{(98.9KB, docx)}

S2 Table. List of all variables derived from eye-tracking simulation as proposed by Schirpke et al. [1].

(1) Primary hotspot characteristics (i.e., quantitative and spatial information generated by 3M-VAS) and (2) secondary hotspot characteristics (i.e., variables in relation to the different distance zones as well as natural and artificial features within the hotspots).

(DOCX)

Click here for additional data file.^{(101.5KB, docx)}

S3 Table. List of all variables derived from photo content analysis as proposed by Schirpke et al. [1].

(DOCX)

Click here for additional data file.^{(99.1KB, docx)}

S4 Table. Datasets used to calculate landscape metrics.

(DOCX)

Click here for additional data file.^{(98.9KB, docx)}

S5 Table. List of all variables derived from Geographic Information System (GIS)-based analysis as proposed by Schirpke et al. [1].

(DOCX)

Click here for additional data file.^{(98.8KB, docx)}

S6 Table. Predictor variable values used as input for the regression analysis.

(XLSX)

Click here for additional data file.^{(25.4KB, xlsx)}

S7 Table. Differences in perceptions for 29 photo pairs with clouds (A) and without clouds (B).

Significant differences are formatted in bold (sample size: 124 participants).

(DOCX)

Click here for additional data file.^{(106.2KB, docx)}

S8 Table. Mean values and standard deviation (SD) of the variables sky and clouds across picture pairs with decreased (A>B), not changed (A = B), and increased (A<B) preference score.

A: Original picture with clouds, B: manipulated picture without clouds.

(DOCX)

Click here for additional data file.^{(96.7KB, docx)}

S9 Table. Spatial shift hotspots due to cloud removal.

Only pictures with significant differences between preference scores of the original and the manipulated picture are included. The spatial shift was determined based on a grid with 9 cells that was overlaid over the pictures (see S6 Fig).

(DOCX)

Click here for additional data file.^{(98KB, docx)}

S10 Table. Main cloud types in the original pictures.

(DOCX)

Click here for additional data file.^{(99KB, docx)}

S1 File

(PDF)

Click here for additional data file.^{(248.5KB, pdf)}

Acknowledgments

We thank all respondents for participating in the survey. The authors are grateful to Kelly Canavan, Global Marketing Development Manager for VAS at 3M Company for allowing the use of 3M-VAS software. We would also like to thank Mr. Wolfgang Gurgiser (meteorologist) for his help in determining the cloud types.

Data Availability

All relevant data are within the paper and its Supporting Information files.

Funding Statement

This work was supported by the Department of Innovation, Research, University and Museums of the Autonomous Province of Bozen/Bolzano. The authors thank the Department of Innovation, Research, University and Museums of the Autonomous Province of Bozen/Bolzano for covering the Open Access publication costs. The funders had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript.

References

1.Cheng X, Van Damme S, Li L, Uyttenhove P. Evaluation of cultural ecosystem services: A review of methods. Ecosystem Services. 2019;37: 100925. doi: 10.1016/j.ecoser.2019.100925 [DOI] [Google Scholar]
2.Müller SM, Peisker J, Bieling C, Linnemann K, Reidl K, Schmieder K. The importance of cultural ecosystem services and biodiversity for landscape visitors in the biosphere reserve Swabian Alb (Germany). Sustainability (Switzerland). 2019. doi: 10.3390/su11092650 [DOI] [Google Scholar]
3.Zoderer BM, Tasser E, Erb KH, Lupo Stanghellini PS, Tappeiner U. Identifying and mapping the tourists’ perception of cultural ecosystem services: A case study from an Alpine region. Land Use Policy. 2016;56: 251–261. doi: 10.1016/j.landusepol.2016.05.004 [DOI] [Google Scholar]
4.Soy-Massoni E, Langemeyer J, Varga D, Sáez M, Pintó J. The importance of ecosystem services in coastal agricultural landscapes: Case study from the Costa Brava, Catalonia. Ecosystem Services. 2016;17: 43–52. doi: 10.1016/j.ecoser.2015.11.004 [DOI] [Google Scholar]
5.Nowak-Olejnik A, Schirpke U, Tappeiner U. A systematic review on subjective well-being benefits associated with cultural ecosystem services. Ecosystem Services. 2022;57: 101467. doi: 10.1016/j.ecoser.2022.101467 [DOI] [Google Scholar]
6.Bratman GN, Anderson CB, Berman MG, Cochran B, de Vries S, Flanders J, et al. Nature and mental health: An ecosystem service perspective. Science Advances. 2019;5: eaax0903. doi: 10.1126/sciadv.aax0903 [DOI] [PMC free article] [PubMed] [Google Scholar]
7.Barton J, Pretty J. What is the best dose of nature and green exercise for improving mental health- A multi-study analysis. Environmental Science and Technology. 2010;44: 3947–3955. doi: 10.1021/es903183r [DOI] [PubMed] [Google Scholar]
8.Berman MG, Jonides J, Kaplan S. The cognitive benefits of interacting with nature. Psychological Science. 2008;19: 1207–1212. doi: 10.1111/j.1467-9280.2008.02225.x [DOI] [PubMed] [Google Scholar]
9.Kirillova K, Lehto X. Destination Aesthetics and Aesthetic Distance in Tourism Experience. Journal of Travel & Tourism Marketing. 2015;32: 1051–1068. doi: 10.1080/10548408.2014.958608 [DOI] [Google Scholar]
10.van Zanten BT, Zasada I, Koetse MJ, Ungaro F, Häfner K, Verburg PH. A comparative approach to assess the contribution of landscape features to aesthetic and recreational values in agricultural landscapes. Ecosystem Services. 2016;17: 87–98. doi: 10.1016/j.ecoser.2015.11.011 [DOI] [Google Scholar]
11.Casado-Arzuaga I, Onaindia M, Madariaga I, Verburg PH. Mapping recreation and aesthetic value of ecosystems in the Bilbao Metropolitan Greenbelt (northern Spain) to support landscape planning. Landscape Ecology. 2014;29: 1393–1405. doi: 10.1007/s10980-013-9945-2 [DOI] [Google Scholar]
12.Assandri G, Bogliani G, Pedrini P, Brambilla M. Beautiful agricultural landscapes promote cultural ecosystem services and biodiversity conservation. Agriculture, Ecosystems and Environment. 2018;256: 200–210. doi: 10.1016/j.agee.2018.01.012 [DOI] [Google Scholar]
13.Brady E, Prior J. Environmental aesthetics: A synthetic review. People and Nature. 2020;2: 254–266. doi: 10.1002/pan3.10089 [DOI] [Google Scholar]
14.Dou Y, Zhen L, Yu X, Bakker M, Carsjens GJ, Xue Z. Assessing the influences of ecological restoration on perceptions of cultural ecosystem services by residents of agricultural landscapes of western China. Science of the Total Environment. 2019;646: 685–695. doi: 10.1016/j.scitotenv.2018.07.205 [DOI] [PubMed] [Google Scholar]
15.Plieninger T, Bieling C, Fagerholm N, Byg A, Hartel T, Hurley P, et al. The role of cultural ecosystem services in landscape management and planning. Current Opinion in Environmental Sustainability. 2015. pp. 28–33. doi: 10.1016/j.cosust.2015.02.006 [DOI] [Google Scholar]
16.Daniel TC, Muhar A, Arnberger A, Aznar O, Boyd JW, Chan KMA, et al. Contributions of cultural services to the ecosystem services agenda. Proceedings of the National Academy of Sciences of the United States of America. 2012;109: 8812–8819. doi: 10.1073/pnas.1114773109 [DOI] [PMC free article] [PubMed] [Google Scholar]
17.Tveit MS. Indicators of visual scale as predictors of landscape preference; a comparison between groups. Journal of Environmental Management. 2009;90: 2882–2888. doi: 10.1016/j.jenvman.2007.12.021 [DOI] [PubMed] [Google Scholar]
18.Ode Sang Å, Tveit MS, Pihel J, Hägerhäll CM. Identifying cues for monitoring stewardship in Swedish pasture landscapes. Land Use Policy. 2016;53: 20–26. doi: 10.1016/j.landusepol.2015.09.020 [DOI] [Google Scholar]
19.Lothian A. and the philosophy of aesthetics: Is landscape quality inherent in the landscape or in the eye of the beholder? Landscape and Urban Planning. 1999;44: 177–198. doi: 10.1016/S0169-2046(99)00019-5 [DOI] [Google Scholar]
20.Daniel TC. Whither scenic beauty? Visual landscape quality assessment in the 21st century. Landscape and Urban Planning. 2001;54: 267–281. doi: 10.1016/S0169-2046(01)00141-4 [DOI] [Google Scholar]
21.Schirpke U, Tappeiner G, Tasser E, Tappeiner U. Using conjoint analysis to gain deeper insights into aesthetic landscape preferences. Ecological Indicators. 2019;96: 202–212. doi: 10.1016/j.ecolind.2018.09.001 [DOI] [Google Scholar]
22.Junge X, Schüpbach B, Walter T, Schmid B, Lindemann-Matthies P. Aesthetic quality of agricultural landscape elements in different seasonal stages in Switzerland. Landscape and Urban Planning. 2015;133: 67–77. doi: 10.1016/j.landurbplan.2014.09.010 [DOI] [Google Scholar]
23.Sahraoui Y, Clauzel C, Foltête JC. Spatial modelling of landscape aesthetic potential in urban-rural fringes. Journal of Environmental Management. 2016;181: 623–636. doi: 10.1016/j.jenvman.2016.06.031 [DOI] [PubMed] [Google Scholar]
24.Soliva R, Bolliger J, Hunziker M. Differences in preferences towards potential future landscapes in the Swiss Alps. Landscape Research. 2010;35: 671–696. doi: 10.1080/01426397.2010.519436 [DOI] [Google Scholar]
25.Schirpke U, Timmermann F, Tappeiner U, Tasser E. Cultural ecosystem services of mountain regions: Modelling the aesthetic value. Ecological Indicators. 2016;69: 78–90. doi: 10.1016/j.ecolind.2016.04.001 [DOI] [PMC free article] [PubMed] [Google Scholar]
26.Hunziker M, Felber P, Gehring K, Buchecker M, Bauer N, Kienast F. Evaluation of Landscape Change by Different Social Groups. Mountain Research and Development. 2008;28: 140–147. doi: 10.1659/mrd.0952 [DOI] [Google Scholar]
27.Dramstad WE, Tveit MS, Fjellstad WJ, Fry GLA. Relationships between visual landscape preferences and map-based indicators of landscape structure. Landscape and Urban Planning. 2006;78: 465–474. doi: 10.1016/j.landurbplan.2005.12.006 [DOI] [Google Scholar]
28.Hedblom M, Hedenås H, Blicharska M, Adler S, Knez I, Mikusiński G, et al. Landscape perception: linking physical monitoring data to perceived landscape properties. Landscape Research. 2020;45: 179–192. doi: 10.1080/01426397.2019.1611751 [DOI] [Google Scholar]
29.Arriaza M, Cañas-Ortega JF, Cañas-Madueño JA, Ruiz-Aviles P. Assessing the visual quality of rural landscapes. Landscape and Urban Planning. 2004;69: 115–125. doi: 10.1016/j.landurbplan.2003.10.029 [DOI] [Google Scholar]
30.Howley P. Landscape aesthetics: Assessing the general publics’ preferences towards rural landscapes. Ecological Economics. 2011;72: 161–169. doi: 10.1016/j.ecolecon.2011.09.026 [DOI] [Google Scholar]
31.Sung DG, Lim SH, Ko JW, Cho GS. Scenic evaluation of landscape for urban design purposes using GIS and ANN. Landscape and Urban Planning. 2001;56: 75–85. doi: 10.1016/S0169-2046(01)00174-8 [DOI] [Google Scholar]
32.López-Martínez F. Visual landscape preferences in Mediterranean areas and their socio-demographic influences. Ecological Engineering. 2017;104: 205–215. doi: 10.1016/j.ecoleng.2017.04.036 [DOI] [Google Scholar]
33.Schaak H, Musshoff O. Public preferences for pasture landscapes in Germany—A latent class analysis of a nationwide discrete choice experiment. Land Use Policy. 2020;91: 104371. doi: 10.1016/j.landusepol.2019.104371 [DOI] [Google Scholar]
34.Ebenberger M, Arnberger A. Exploring visual preferences for structural attributes of urban forest stands for restoration and heat relief. Urban Forestry and Urban Greening. 2019;41: 272–282. doi: 10.1016/j.ufug.2019.04.011 [DOI] [Google Scholar]
35.Silvennoinen H, Koivula M, Tikkanen J, Tyrväinen L. Consistency of photo and field surveys in measuring the attractiveness of forest stands. Journal of Outdoor Recreation and Tourism. 2022;40: 100586. doi: 10.1016/j.jort.2022.100586 [DOI] [Google Scholar]
36.Sevenant M, Antrop M. Landscape Representation Validity: A Comparison between On-site Observations and Photographs with Different Angles of View. Landscape Research. 2011;36: 363–385. doi: 10.1080/01426397.2011.564858 [DOI] [Google Scholar]
37.Kuper R. Evaluations of landscape preference, complexity, and coherence for designed digital landscape models. Landscape and Urban Planning. 2017;157: 407–421. doi: 10.1016/j.landurbplan.2016.09.002 [DOI] [Google Scholar]
38.Häfner K, Zasada I, van Zanten BT, Ungaro F, Koetse M, Piorr A. Assessing landscape preferences: a visual choice experiment in the agricultural region of Märkische Schweiz, Germany. Landscape Research. 2018;43: 846–861. doi: 10.1080/01426397.2017.1386289 [DOI] [Google Scholar]
39.Spielhofer R, Hunziker M, Kienast F, Wissen Hayek U, Grêt-Regamey A. Does rated visual landscape quality match visual features? An analysis for renewable energy landscapes. Landscape and Urban Planning. 2021;209: 104000. doi: 10.1016/j.landurbplan.2020.104000 [DOI] [Google Scholar]
40.Pólrolniczak M, Potocka I, Kolendowicz L, Rogowski M, Kupiński S, Bykowski A, et al. The impact of biometeorological conditions on the perception of landscape. Atmosphere. 2019. doi: 10.3390/atmos10050264 [DOI] [Google Scholar]
41.White MP, Cracknell D, Corcoran A, Jenkinson G, Depledge MH. Do Preferences for Waterscapes Persist in Inclement Weather and Extend to Sub-aquatic Scenes? Landscape Research. 2014;39: 339–358. doi: 10.1080/01426397.2012.759919 [DOI] [Google Scholar]
42.Seresinhe CI, Preis T, MacKerron G, Moat HS. Happiness is Greater in More Scenic Locations. Scientific Reports. 2019;9: 4498. doi: 10.1038/s41598-019-40854-6 [DOI] [PMC free article] [PubMed] [Google Scholar]
43.Smalley AJ, White MP. Beyond blue-sky thinking: Diurnal patterns and ephemeral meteorological phenomena impact appraisals of beauty, awe, and value in urban and natural landscapes. Journal of Environmental Psychology. 2023;86: 101955. doi: 10.1016/j.jenvp.2023.101955 [DOI] [Google Scholar]
44.Konopka E. The Multi-Layered Symbolism of Clouds. University of Warsaw. 2018. [Google Scholar]
45.Dietrich S. The sky as mirror of the mind: exploring the role of light and weather for emotional expression in northern European landscape painting. University of Aberdeen. 2019. [Google Scholar]
46.Damisch H. A theory of /cloud: toward a history of painting. Stanford: Stanford University Press; 2002. [Google Scholar]
47.Nova A, Michalsky T. Wind und Wetter: die Ikonologie der Atmosphäre. Marsilio; 2009. [Google Scholar]
48.Stewart AE, Blau JJC. Weather as Ecological Events. Ecological Psychology. 2019;31: 107–126. doi: 10.1080/10407413.2018.1552496 [DOI] [Google Scholar]
49.Starl T. Eine kleine Geschichte der Wolkenfotografie BT—stark bewölkt: flüchtige Erscheinungen des Himmels. In: Siegrun A, Ona B, Hildegund B, Gerhard B, Astrid B, Fritz B, et al., editors. Vienna: Springer Vienna; 2009. pp. 22–41. doi: 10.1007/978-3-211-89114-8_2 [DOI] [Google Scholar]
50.Schirpke U, Tasser E, Ebner M, Tappeiner U. What can geotagged photographs tell us about cultural ecosystem services of lakes? Ecosystem Services. 2021;51: 101354. doi: 10.1016/j.ecoser.2021.101354 [DOI] [Google Scholar]
51.Tieskens KF, Van Zanten BT, Schulp CJE, Verburg PH. Aesthetic appreciation of the cultural landscape through social media: An analysis of revealed preference in the Dutch river landscape. Landscape and Urban Planning. 2018;177: 128–137. doi: 10.1016/j.landurbplan.2018.05.002 [DOI] [Google Scholar]
52.Huang AS-H, Lin Y-J. The effect of landscape colour, complexity and preference on viewing behaviour. Landscape Research. 2020;45: 214–227. doi: 10.1080/01426397.2019.1593336 [DOI] [Google Scholar]
53.Schirpke U, Tasser E, Lavdas AA. Potential of eye-tracking simulation software for analyzing landscape preferences. PLoS ONE. 2022;17: e0273519. doi: 10.1371/journal.pone.0273519 [DOI] [PMC free article] [PubMed] [Google Scholar]
54.Forer M. Ästhetisches Erleben von Landschaft–eine quantitative Analyse der Landschaftswahrnehmung in Südtirol. University of Innsbruck. 2020. [Google Scholar]
55.Svobodova K, Sklenicka P, Molnarova K, Vojar J. Does the composition of landscape photographs affect visual preferences? The rule of the Golden Section and the position of the horizon. Journal of Environmental Psychology. 2014;38: 143–152. doi: 10.1016/j.jenvp.2014.01.005 [DOI] [Google Scholar]
56.Preston CC, Colman AM. Optimal number of response categories in rating scales: Reliability, validity, discriminating power, and respondent preferences. Acta Psychologica. 2000;104: 1–15. doi: 10.1016/s0001-6918(99)00050-5 [DOI] [PubMed] [Google Scholar]
57.Couper MP, Tourangeau R, Conrad FG, Crawford SD. What They See Is What We Get: Response Options for Web Surveys. Social Science Computer Review. 2004;22: 111–127. doi: 10.1177/0894439303256555 [DOI] [Google Scholar]
58.Cohen J. Statistical Power Analysis for the Behavioral Sciences. Statistical Power Analysis for the Behavioral Sciences. Elsevier Science; 2013. doi: 10.4324/9780203771587 [DOI] [Google Scholar]
59.de Jonge T, Veenhoven R, Arends L. Homogenizing Responses to Different Survey Questions on the Same Topic: Proposal of a Scale Homogenization Method Using a Reference Distribution. Social Indicators Research. 2014;117: 275–300. doi: 10.1007/s11205-013-0335-6 [DOI] [PMC free article] [PubMed] [Google Scholar]
60.3M. 3M Visual Attention Serv: Validation Study. 2010 [cited 29 Apr 2023] p. 7. Available: http://solutions.3m.com/3MContentRetrievalAPI/BlobServlet?locale=en_WW&lmd=1272291661000&assetId=1258566024761&assetType=MMM_Image&blobAttribute=ImageFile
61.Lee JH, Ostwald MJ. The ‘visual attractiveness’ of architectural facades: measuring visual complexity and attractive strength in architecture. Architectural Science Review. 2023;66: 42–52. doi: 10.1080/00038628.2022.2137458 [DOI] [Google Scholar]
62.Lavdas AA, Salingaros NA, Sussman A. Visual Attention Software: A New Tool for Understanding the “Subliminal” Experience of the Built Environment. Applied Sciences. 2021. doi: 10.3390/app11136197 [DOI] [Google Scholar]
63.Fábrega-Álvarez P, Parcero-Oubiña C. Now you see me. An assessment of the visual recognition and control of individuals in archaeological landscapes. Journal of Archaeological Science. 2019;104: 56–74. doi: 10.1016/j.jas.2019.02.002 [DOI] [Google Scholar]
64.McGarigal K, Cushman SA, Ene E. FRAGSTATS v4: Spatial Pattern Analysis Program for Categorical and Continuous Maps. Computer software program produced by the authors at the University of Massachusetts, Amherst. Fragstats. 2012. Available: https://www.umass.edu/landeco/research/fragstats/fragstats.html [Google Scholar]
65.Horvath H. Estimation of the average visibility in central Europe. Atmospheric Environment. 1995;29: 241–246. doi: 10.1016/1352-2310(94)00236-E [DOI] [Google Scholar]
66.Kim Y-H, Rana S, Wise S. Exploring multiple viewshed analysis using terrain features and optimisation techniques. Computers & Geosciences. 2004;30: 1019–1032. 10.1016/j.cageo.2004.07.008 [DOI] [Google Scholar]
67.Schirpke U, Zoderer BM, Tappeiner U, Tasser E. Effects of past landscape changes on aesthetic landscape values in the European Alps. Landscape and Urban Planning. 2021;212: 104109. doi: 10.1016/j.landurbplan.2021.104109 [DOI] [Google Scholar]
68.Barroso FL, Pinto-Correia T, Ramos IL, Surová D, Menezes H. Dealing with landscape fuzziness in user preference studies: Photo-based questionnaires in the Mediterranean context. Landscape and Urban Planning. 2012;104: 329–342. doi: 10.1016/j.landurbplan.2011.11.005 [DOI] [Google Scholar]
69.Meech S. Contemporary quilts: design, surface and stitch. Anova Books; 2005. [Google Scholar]
70.Wang Y, Ke Y, Wang K, Guo J, Qin F. A composition-oriented aesthetic view recommendation network supervised by the simplified golden ratio theory. Expert Systems with Applications. 2022;195: 116500. doi: 10.1016/j.eswa.2022.116500 [DOI] [Google Scholar]
71.John Neter D, Kutner MH, Nachtsheim CJ. MP Applied Linear Regression Models-Revised Edition with Student CD. McGraw-Hill Education; 2004. [Google Scholar]
72.Hao X, Yujun S, Xinjie W, Jin W, Yao F. Linear Mixed-Effects Models to Describe Individual Tree Crown Width for China-Fir in Fujian Province, Southeast China. PLOS ONE. 2015;10: e0122257. doi: 10.1371/journal.pone.0122257 [DOI] [PMC free article] [PubMed] [Google Scholar]
73.Frost J. Regression Analysis: An Intuitive Guide for Using and Interpreting Linear Models. Statistics By Jim Publishing; 2020. [Google Scholar]
74.Lovejoy S. Area-perimeter relation for rain and cloud areas. Science. 1982;216: 185–187. doi: 10.1126/science.216.4542.185 [DOI] [PubMed] [Google Scholar]
75.Taylor RP, Spehar B, van Donkelaar P, Hagerhall CM. Perceptual and physiological responses to Jackson Pollock’s fractals. Frontiers in Human Neuroscience. 2011;5. doi: 10.3389/fnhum.2011.00060 [DOI] [PMC free article] [PubMed] [Google Scholar]
76.Fischmeister FP, Martins MJD, Beisteiner R, Fitch WT. Self-similarity and recursion as default modes in human cognition. Cortex. 2017;97: 183–201. doi: 10.1016/j.cortex.2016.08.016 [DOI] [PubMed] [Google Scholar]
77.Martins MJ, Fischmeister FP, Puig-Waldmüller E, Oh J, Geißler A, Robinson S, et al. Fractal image perception provides novel insights into hierarchical cognition. NeuroImage. 2014;96: 300–308. doi: 10.1016/j.neuroimage.2014.03.064 [DOI] [PubMed] [Google Scholar]
78.De Bartolo D, De Luca M, Antonucci G, Schuster S, Morone G, Paolucci S, et al. The golden ratio as an ecological affordance leading to aesthetic attractiveness. PsyCh Journal. 2022;11: 729–740. doi: 10.1002/pchj.505 [DOI] [PMC free article] [PubMed] [Google Scholar]
79.Ode Å, Fry G, Tveit MS, Messager P, Miller D. Indicators of perceived naturalness as drivers of landscape preference. Journal of Environmental Management. 2009;90: 375–383. doi: 10.1016/j.jenvman.2007.10.013 [DOI] [PubMed] [Google Scholar]
80.Torres Sibille A del C, Cloquell-Ballester VA, Cloquell-Ballester VA, Darton R. Development and validation of a multicriteria indicator for the assessment of objective aesthetic impact of wind farms. Renewable and Sustainable Energy Reviews. 2009;13: 40–66. doi:10.1016/j.rser.2007.05.002
81.Mohammadi M, Chen Y, Rahman HMT, Sherren K. A saliency mapping approach to understanding the visual impact of wind and solar infrastructure in amenity landscapes. Impact Assessment and Project Appraisal. 2023;41: 154–161. doi: 10.1080/14615517.2023.2169460 [DOI] [Google Scholar]
82.Lin L, Homma R, Iki K. Preferences for a lake landscape: Effects of building height and lake width. Environmental Impact Assessment Review. 2018;70: 22–33. doi: 10.1016/j.eiar.2018.03.001 [DOI] [Google Scholar]
83.Stamps AE. Fractals, skylines, nature and beauty. Landscape and Urban Planning. 2002;60: 163–184. doi: 10.1016/S0169-2046(02)00054-3 [DOI] [Google Scholar]
84.Hagerhall CM, Purcell T, Taylor R. Fractal dimension of landscape silhouette outlines as a predictor of landscape preference. Journal of Environmental Psychology. 2004;24: 247–255. doi: 10.1016/j.jenvp.2003.12.004 [DOI] [Google Scholar]
85.Tang J, Long Y. Measuring visual quality of street space and its temporal variation: Methodology and its application in the Hutong area in Beijing. Landscape and Urban Planning. 2019;191: 103436. doi: 10.1016/j.landurbplan.2018.09.015 [DOI] [Google Scholar]
86.Petrova EG, Mironov Y V., Aoki Y, Matsushima H, Ebine S, Furuya K, et al. Comparing the visual perception and aesthetic evaluation of natural landscapes in Russia and Japan: cultural and environmental factors. Progress in Earth and Planetary Science. 2015;2: 6. doi: 10.1186/s40645-015-0033-x [DOI] [Google Scholar]
87.Gosal AS, Geijzendorffer IR, Václavík T, Poulin B, Ziv G. Using social media, machine learning and natural language processing to map multiple recreational beneficiaries. Ecosystem Services. 2019;38: 100958. doi: 10.1016/j.ecoser.2019.100958 [DOI] [Google Scholar]
88.Sayadi S, González-Roa MC, Calatrava-Requena J. Public preferences for landscape features: The case of agricultural landscape in mountainous Mediterranean areas. Land Use Policy. 2009;26: 334–344. doi: 10.1016/j.landusepol.2008.04.003 [DOI] [Google Scholar]
89.Wang R, Zhao J. Effects of evergreen trees on landscape preference and perceived restorativeness across seasons. Landscape Research. 2020;45: 649–661. doi: 10.1080/01426397.2019.1699507 [DOI] [Google Scholar]
90.Kuper R. Here and Gone. Landscape Journal. 2013;32: 65 LP– 78. doi: 10.3368/lj.32.1.65 [DOI] [Google Scholar]
91.Nagai S, Saitoh TM, Yoshitake S. Cultural ecosystem services provided by flowering of cherry trees under climate change: a case study of the relationship between the periods of flowering and festivals. International Journal of Biometeorology. 2019;63: 1051–1058. doi: 10.1007/s00484-019-01719-9 [DOI] [PubMed] [Google Scholar]
92.Graves RA, Pearson SM, Turner MG. Landscape dynamics of floral resources affect the supply of a biodiversity-dependent cultural ecosystem service. Landscape Ecology. 2017;32: 415–428. doi: 10.1007/s10980-016-0452-0 [DOI] [Google Scholar]
93.Carlson KM, Mora C, Xu J, Setter RO, Harangody M, Franklin EC, et al. Global rainbow distribution under current and future climates. Global Environmental Change. 2022;77. doi: 10.1016/j.gloenvcha.2022.102604 [DOI] [Google Scholar]

PLoS One. doi: 10.1371/journal.pone.0288424.r001

Decision Letter 0

Chaohai Shen

19 Apr 2023

PONE-D-23-07973Do clouds in landscape photographs influence people’s preferences?PLOS ONE

Dear Dr. Schirpke,

Thank you for submitting your manuscript to PLOS ONE. After careful consideration, we feel that it has merit but does not fully meet PLOS ONE’s publication criteria as it currently stands. Therefore, we invite you to submit a revised version of the manuscript that addresses the points raised during the review process.

Please submit your revised manuscript by Jun 03 2023 11:59PM. If you will need more time than this to complete your revisions, please reply to this message or contact the journal office at plosone@plos.org. When you're ready to submit your revision, log on to https://www.editorialmanager.com/pone/ and select the 'Submissions Needing Revision' folder to locate your manuscript file.

Please include the following items when submitting your revised manuscript:

A rebuttal letter that responds to each point raised by the academic editor and reviewer(s). You should upload this letter as a separate file labeled 'Response to Reviewers'.
A marked-up copy of your manuscript that highlights changes made to the original version. You should upload this as a separate file labeled 'Revised Manuscript with Track Changes'.
An unmarked version of your revised paper without tracked changes. You should upload this as a separate file labeled 'Manuscript'.

If you would like to make changes to your financial disclosure, please include your updated statement in your cover letter. Guidelines for resubmitting your figure files are available below the reviewer comments at the end of this letter.

If applicable, we recommend that you deposit your laboratory protocols in protocols.io to enhance the reproducibility of your results. Protocols.io assigns your protocol its own identifier (DOI) so that it can be cited independently in the future. For instructions see: https://journals.plos.org/plosone/s/submission-guidelines#loc-laboratory-protocols. Additionally, PLOS ONE offers an option for publishing peer-reviewed Lab Protocol articles, which describe protocols hosted on protocols.io. Read more information on sharing protocols at https://plos.org/protocols?utm_medium=editorial-email&utm_source=authorletters&utm_campaign=protocols.

We look forward to receiving your revised manuscript.

Kind regards,

Chaohai Shen

Academic Editor

PLOS ONE

Journal Requirements:

When submitting your revision, we need you to address these additional requirements.

1. Please ensure that your manuscript meets PLOS ONE's style requirements, including those for file naming. The PLOS ONE style templates can be found at https://journals.plos.org/plosone/s/file?id=wjVg/PLOSOne_formatting_sample_main_body.pdf and

https://journals.plos.org/plosone/s/file?id=ba62/PLOSOne_formatting_sample_title_authors_affiliations.pdf

2. You indicated that ethical approval was not necessary for your study. We understand that the framework for ethical oversight requirements for studies of this type may differ depending on the setting and we would appreciate some further clarification regarding your research. Could you please provide further details on why your study is exempt from the need for approval and confirmation from your institutional review board or research ethics committee (e.g., in the form of a letter or email correspondence) that ethics review was not necessary for this study? Please include a copy of the correspondence as an ""Other"" file

3. Please amend your current ethics statement to address the following concerns:

a) Did participants provide their written or verbal informed consent to participate in this study?

b) If consent was verbal, please explain i) why written consent was not obtained, ii) how you documented participant consent, and iii) whether the ethics committees/IRB approved this consent procedure.

Additional Editor Comments (if provided):

Dear Authors,

I have received all the required reviewer reports. I agree with the comments from the reviewers and I would like to invite you to revise the manuscript accordingly.

Thank you!

Sincerely,

[Note: HTML markup is below. Please do not edit.]

Reviewers' comments:

Reviewer's Responses to Questions

Comments to the Author

1. Is the manuscript technically sound, and do the data support the conclusions?

The manuscript must describe a technically sound piece of scientific research with data that supports the conclusions. Experiments must have been conducted rigorously, with appropriate controls, replication, and sample sizes. The conclusions must be drawn appropriately based on the data presented.

Reviewer #1: Yes

Reviewer #2: Partly

Reviewer #3: Partly

**********

2. Has the statistical analysis been performed appropriately and rigorously?

Reviewer #1: I Don't Know

Reviewer #2: Yes

Reviewer #3: I Don't Know

**********

3. Have the authors made all data underlying the findings in their manuscript fully available?

The PLOS Data policy requires authors to make all data underlying the findings described in their manuscript fully available without restriction, with rare exception (please refer to the Data Availability Statement in the manuscript PDF file). The data should be provided as part of the manuscript or its supporting information, or deposited to a public repository. For example, in addition to summary statistics, the data points behind means, medians and variance measures should be available. If there are restrictions on publicly sharing data—e.g. participant privacy or use of data from a third party—those must be specified.

Reviewer #1: No

Reviewer #2: No

Reviewer #3: No

**********

4. Is the manuscript presented in an intelligible fashion and written in standard English?

PLOS ONE does not copyedit accepted manuscripts, so the language in submitted articles must be clear, correct, and unambiguous. Any typographical or grammatical errors should be corrected at revision, so please note any specific errors here.

Reviewer #1: Yes

Reviewer #2: No

Reviewer #3: Yes

**********

5. Review Comments to the Author

Please use the space provided to explain your answers to the questions above. You may also include additional comments for the author, including concerns about dual publication, research ethics, or publication ethics. (Please upload your review as an attachment if it exceeds 20,000 characters)

Reviewer #1: I'm not able to see where they mention a repository and I don't see their data attached as supplementary, so I answered no to that question above. I may simply have just missed something.

For the authors:

This well-crafted and explained study explores the impacts of clouds in landscape photographs in mountainous settings, using paired original and manipulated photos. The combination of simulated eye-tracking, photo structural analysis and landscape metrics provide a rich assessment of when clouds matter and when they do not. I do not have the statistics skills to assess their quantitative methods, so hope another reviewer is doing that work. But as a landscape researcher I can take important methodological advice from this work.

A few details of the survey are unclear. Could the authors confirm if survey respondents received all 58 images, or a subset of them? That is a large set of photos to review for any one respondent. The time that the survey took suggests the respondents would spend only 5-10 seconds per photo, which seems to be a short time to carry out such cognitive tasks over such a number of them. I also wonder if the survey respondents were incentivized at all? I can’t imagine getting so many people to do a survey of this type without offering academic credit, or a chance to win something. Were the students in any particular program, and how were they invited? Demographics were clearly collected, but are not reported. Such details are generally disclosed, even though the demographics are not the key interest in this survey.

I also enjoyed learning about the 3M-VAS software, but I think it is important to provide more about how the software works. Looking at the webpage it clearly has an AI engine, but can the authors provide a brief explanation of this software (as far as is known, given its commercial status) so that we can assess replicability? Is the program using a similar process as is described elsewhere as ‘salience mapping’? If so, it may be appropriate to connect those literatures. For instance, in the discussion the authors mention the potential value of such methods in understanding the visual impact of wind turbines, and this has been done with salience maps. (e.g., Mohammadi, Mehrnoosh, et al. "A saliency mapping approach to understanding the visual impact of wind and solar infrastructure in amenity landscapes." Impact Assessment and Project Appraisal 41.2 (2023): 154-161.)

A little more detail would also be useful in the section discussing the regression analysis. Can the authors describe exactly what the dependent variable is? Is it A-B?

The graphics are largely effective and well-designed, but Figure 2 needs a little more work. The A and B becomes confusing because there are two uses of these in the figure, for the two parts of the figure and for the cloud and no-cloud image versions. In black and white, the colours for A>B and A=B are identical, and the link to the border of the images is lost. I would suggest bars instead of a pie, extending beside each of the photos, with labels alongside, to completely disambiguate. This would also remove the need for this figure to be seen as two things, as the elements would be entirely integrated.

Language is clear. Only one “und”, and minor issues like “ration” instead of “ratio” in one spot, and “contribute” instead of “contribution” through Table 1.

There is an indication in the cover materials that the authors have made their data fully available, but I’m not able to see where it is attached or pointed to in a repository. My apologies if I simply missed this.

Reviewer #2: Dear Colleagues,

I have read and reviewed your article titled, Do clouds in landscape photographs influence people’s preferences? The research is nicely done and I myself has learned a lot as well. As this is one of my favourite research areas and a discipline, I am familiar with, I would like to make some comments about the manuscript and other aspects. I am grateful for the authors for their attempt in this study which would contribute highly to the landscape and perception related research in the domain.

Some of these comments might be not 100% relevant but I would like to suggest those to improve the structure and the overall quality of your paper.

The title of the manuscript seems a bit too simple. There is only one question as the title. If we incorporate some extra phrase to provide more insights to the manuscript following the question, then I think it would be better. EX:- , Do clouds in landscape photographs influence people’s preferences?; A study in XXXX, or else.

The structure needs a bit more arranging. The titles, Collection of variables, Photo-based survey, Study design can be all made into one topic of "materials and methods". The content in the statistical analysis can be shifted under methods/ results appropriately,

Irrespective of the suggested rearrangement, I will comment based on the existing topics (by the author)

Abstract

Try to be more emphasized on the study and be more focussed. This abstract focusses more on the results which is a good point. But the concluding sentence can be more made useful. Think if some one wants to read your whole article, the abstract is what gets their attention. So try to attract them with the concluding sentence.

Line 14 – 16 – Rewrite the sentence better.

Write the purpose of the study more focussed in the abstract.

Introduction

This is okay, but I would like to have some more insights to the clouds in the intro. Shapes of the clouds and some other information you have discussed in the discussion part seems to be fitting more in the introduction. The clouds and its visible aspects in the landscapes is an interesting part related to the perception research. If you could include those information, then it would be more interesting for the readers.

Line 77 – Sentence should start with a new line.

Photo based survey

Line 94 – Give credits to photoshop

Line 102 – it took about 5-10 minutes ........ – this part is too casual. May be the estimated time to fill the questionnaire?

Line 103 – What is the process of selecting the respondents? A proper sampling technique?

Why 10 point likert scale? Why not an odd number (with a mid point - neutral response)? Why 10 points? Please explain why these were selected for the study.

What is the basis for selecting the photographs? Please explain why?

The general preference of the sample is justified using previous studies. How can you be sure the previous samples reflect general conditions? It is true that high numbers correspond to the normality. Is this comparison to previous studies necessary? Think again.

Ethics statement

This in the middle of the manuscript? Better to include in the body without having a separate section

Line 124 - Adult persons or Adults?

Collection of Variables

Give credits to 3M-VAS software. This information is missing in the manuscript. For the softwares used (manufacturer, country) should be mentioned in parentheses

line 136 – 3 -5 seconds?

photo content analysis – How the structural composition was estimated? Using which software or how? Please explain properly.

We estimated the composition of the landscape in terms of distance zones (i.e., near zone, middle zone, far zone). An illustration of how this was done would be better.

Statistical Analysis

Why backward stepwise linear regression was used? Explain

Results

Line 205 – R2

Line 236 – und?

Line 238 – seconds?

Line 238 – give credits right next to the photographs in the description.

The red and yellow markings in the photographs? What are those?

Each photograph has two analysis shown right? Explain out outcome is this? What these numbers represent. May be I am not quite familiar with the software, I find it difficult to understand. But better if we can include them to increase the readability and for greater exposure.

Where is the conclusion? A paper with no conclusion does not sound right.

The paper is interesting and a great attempt to contribute to the existing literature. Congratulations on the future as a scholar.

Regards!

Reviewer #3: Title

Do clouds in landscape photographs influence people’s preferences?

Summary

The authors present a study examining how clouds, and a host of other independent variables, might affect preference ratings for landscape views. Results indicate differences between images with and without clouds, and according to the proportion of sky depicted. The paper attempts to relate these findings to a cultural ecosystem services framework.

Overall comments

Whilst I endorse the authors’ focus on this overlooked component of landscape evaluation, in its current form the paper does not provide adequate information about the study design or its participants, and reporting of results requires substantial additional information.

The introduction, and materials and methods, should be improved to help the reader understand the visual stimuli being tested and the demographic characteristics of participants. The limited information presented here renders the study impossible to reproduce.

The results are difficult to follow and interpret, and seem to combine empirical findings with visually inferred observations. It is difficult to assess the validity of the statistical approach with the information provided. This section needs careful rewriting to present the results in a way that readers can both understand and suitably interrogate.

The authors must also consider publishing their image sets, analysis scripts, and raw data on the Open Science Framework (or similar platform) such that their findings can be reproduced.

Specific comments

#1 The intentions of the study could be clearer in the abstract. For example at line #13 “Usually, pictures with similar weather and light conditions (i.e., blue sky) are used to reduce the potential influence on landscape preferences” should be rephrased to explain what potential confounders ‘blue sky’ conditions are mitigating.

#2 The introduction would benefit from further information on the different forms clouds can take. The example photographs included in the paper depict various types of cumulus clouds, were these the only cloud formations included across the images used? For example, participant responses to cirrus stratus or altocumulus may have varied substantially compared to cumulus clouds. Moreover, why did the authors choose the cloudscapes included in this study over other formations?

#3 An overview of the methods typically used to assess landscape preferences in photographs would help the reader understand how aesthetic value is usually measured, and therefore how this study's methods might be applicable to other work in this field.

#4 The discussion on study design should include an explanation of how the authors arrived at a suitable sample size. Was a power calculation performed before recruitment? What sample sizes are common in other studies that assess landscape aesthetics?

#5 What were the characteristics of the participants? The authors note that the survey was distributed to undergraduates and their family members, but no information is provided on who actually took part and therefore who their findings relate to. Please include a breakdown of participant demographics, with n’s for each grouping specified.

#6 The authors note that the survey took 5-10 minutes to complete (line #102). This description would benefit from further details. Where did participants complete the study, at home or in a controlled laboratory setting? On what kind of device? How were the images displayed and for how long? Was completion time monitored, and any quality control in place to prevent people from ‘speeding’ through without paying attention? Were participants primed as to the purposes of the study? Virtual eye tracking was used but few details are provided on how and why this method is appropriate, how does it compare to actual eye tracking? On this point, the authors simply refer to “Schirpke et al.” but I suggest that the reader should be presented with these points without having to refer to another study – were the elements of that study followed exactly? How do the methods here deviate?

#7 At line #108 the authors refer to “two former surveys” which the reader assumes are those referenced previously. If this is the case, it should be clarified which studies these are with appropriate referencing and details where necessary. For example, why is it appropriate to compare preference scores from this study to those mentioned?

#8 At line 152 the authors note that “areas hidden by vegetation, buildings, or mountains, were excluded from further analysis.” The reader is curious why it is appropriate for these potential confounding elements to be omitted from the analysis. The authors should provide justification for this approach in the manuscript.

#9 Why it is acceptable to replace missing values with mean values (line #167)? Did this method apply to all IVs? Could the authors explain this in a way such that a reader who is unfamiliar with statistical methods can understand the steps taken.

#10 At line #176 the authors mention tests for normality but do not provide results from these tests until line #206. I suggest clarifying these outcomes here or at least acknowledging that tests were passed and reported later.

#11 To aid interpretation of the results, Figure 2 would be clearer if both A and B images were shown for each example. At present the comparisons being assessed are not clear.

#12 Tabular data underpinning Figure 3 should be displayed. Importantly, what manipulation to the data has been applied to create this figure? Are these mean preference scores across groupings? Are these differences across groups statistically significant? If so, what test has been applied to determine this relationship?

#13 Furthermore, it is not clear how the percentages reported in line #196 have been deduced. Further clarification is required to help these results be interpreted and reproduced.

#14 Table 2 is difficult to interpret in its current form since IVs have been described by their coded descriptors (which the reader can only interpret by cross referencing with Table S1 in supplementary materials). I suggest these are replaced with more easily interpretable labels, for example, “I_HP1..n (%)” could read “Sky within hotspots (%)”. Furthermore, significant coefficients in table 2 might be highlighted in bold to improve readability.

#15 Figure 4 shows some intriguing patterns, but I found it very difficult to follow the descriptions provided in lines #212 to #230. Are the authors referring to results from table 2? If so, it would be useful to report coefficients and p-values in these descriptions so the reader can understand the relationships being described. More importantly, these descriptions appear to refer to images grouped according to whether they do or do not contain clouds. Were two regression models run based on these categories? The IV in the table simply refers to % cloud cover, and clouds Yes/No as a binary variable is not included nor as an interaction term. Greater clarification is required on how the authors can qualify statements such as “Pictures with clouds that had higher preference scores comprised on average more hotspots” and “preferences for landscapes without clouds were often related to the presence of a large top hotspot”. Have these been empirically tested or visually inferred from the data?

#16 At line #220 the authors suggest that “a high proportion of background (EZ4_P1..n) also reduced the negative effects of anthropogenic structures in the foreground (Eart_P1..n)”. How was this dependency tested? No interaction effects between IVs are reported.

#17 Likewise, at line #299 the authors report that “the hotspots in the more attractive image pair are located more towards the sides than in the center.” Was this empirically tested or is this an observation based on one or more photos?

#18 I am unclear on how the 40% threshold for cloud cover noted in line #246 has been derived. The discussion section should not be the first place results are reported.

#19 More importantly, and in relation to comment #2, nimbostratus clouds are mentioned in line #247, and single cumulus clouds mentioned in line #249. This is the first time cloud type has been introduced and suggests substantial heterogeneity in the clouds depicted. More information is required on the composition of these cloud forms in the introduction and materials and methods, how has this been accounted for as a potential confounding factor?

#20 The authors present a quantitative analysis of how clouds might impact preferences for landscape scenes, and attempt to relate this to cultural ecosystem services. However, as hinted at by the phrase ‘cultural’, these values can be highly subjective. It would be valuable to the reader to understand the limitations of this approach, particularly with respect to the fertile literature on subjective vs objective aesthetic assessments.

**********

6. PLOS authors have the option to publish the peer review history of their article (what does this mean?). If published, this will include your full peer review and any attached files.

If you choose “no”, your identity will remain anonymous but your review may still be made public.

Do you want your identity to be public for this peer review? For information about this choice, including consent withdrawal, please see our Privacy Policy.

Reviewer #1: No

Reviewer #2: No

Reviewer #3: No

**********

[NOTE: If reviewer comments were submitted as an attachment file, they will be attached to this email and accessible via the submission site. Please log into your account, locate the manuscript record, and check for the action link "View Attachments". If this link does not appear, there are no attachment files.]

While revising your submission, please upload your figure files to the Preflight Analysis and Conversion Engine (PACE) digital diagnostic tool, https://pacev2.apexcovantage.com/. PACE helps ensure that figures meet PLOS requirements. To use PACE, you must first register as a user. Registration is free. Then, login and navigate to the UPLOAD tab, where you will find detailed instructions on how to use the tool. If you encounter any issues or have any questions when using PACE, please email PLOS at figures@plos.org. Please note that Supporting Information files do not need this step.

Attachment

Submitted filename: Peer reviewer feedback - PONE-D-23-07973.docx

Click here for additional data file.^{(15.6KB, docx)}

PLoS One. 2023 Jul 28;18(7):e0288424. doi: 10.1371/journal.pone.0288424.r002

Author response to Decision Letter 0

12 May 2023

Reviewer #1:

I'm not able to see where they mention a repository and I don't see their data attached as supplementary, so I answered no to that question above. I may simply have just missed something.

For the authors:

Response: Thank you for your time and effort to review our manuscript. We appreciate your encouraging evaluation and the valuable suggestions on how to improve the manuscript.

Response: The respondents had to evaluate all 58 images. We agree that 5 minutes were too short and now indicate about 10 minutes after recording the time again in a test run. The participants could also pause the survey by saving their responses and continue at a later time. We invited students in landscape ecology and physical geography during lectures on landscape perceptions by providing a link to the questionnaire. The participants received neither a compensation nor other incentives. We added these details and further information on the survey asked by the other reviewers to the methods. We added a short description of the main characteristics of the respondents in the manuscript. We also provide a table with full demographic details in the supplementary material (Table S1).

Response: We now added a detailed description of the 3M-VAS software. Thank you for your question about saliency mapping. Indeed, there is a high correlation between eye-tracking outputs obtained in experiments and salience maps, but it is still untested whether the simulation software would also be suitable. We added a short reflection on this in the discussion section.

A little more detail would also be useful in the section discussing the regression analysis. Can the authors describe exactly what the dependent variable is? Is it A-B?

Response: We agree that this was not clear. Indeed, we used the difference in preference score between the original landscape photograph with clouds (A) and the manipulated picture without clouds (B) as dependent variable. We now added this explanation to the manuscript.

Response: Thank you for the suggestion on how to improve figure 2. We revised the figure accordingly.

Language is clear. Only one “und”, and minor issues like “ration” instead of “ratio” in one spot, and “contribute” instead of “contribution” through Table 1.

Response: The manuscript has been checked again for grammar and spelling.

Response: Thank you for indicating this issue. Several data were indeed lacking in the supplementary material, which we now remedied.

Reviewer #2:

Dear Colleagues,

Some of these comments might be not 100% relevant but I would like to suggest those to improve the structure and the overall quality of your paper.

Response: Thank you for your time and effort to review our manuscript. We appreciate your encouraging evaluation and the valuable suggestions on how to improve the manuscript.

Response: We added a short extra phrase. The title now is: Assessing landscape aesthetic values: Do clouds in photographs influence people’s preferences?

Irrespective of the suggested rearrangement, I will comment based on the existing topics (by the author)

Response: The sections from ‘Study design’ to ‘Statistical analyses’ are all included in the ‘materials and methods’ section. Unfortunately, the journal does not use numbering, which would be clearer. We now changes text size to differentiate more clearly main sections and subsections.

Abstract

Response: We revised the abstract, particularly the concluding sentences, to better focus on the study.

Line 14 – 16 – Rewrite the sentence better.

Write the purpose of the study more focussed in the abstract.

Response: We rephrased the sentence to indicate more clearly the issues that are related to the weather and light conditions in the pictures and to better focus the purpose of the study.

Introduction

Response: We agree that the information on the clouds would be better placed in the introduction. We therefore moved a part of it from the discussion to the introduction section, which we also integrated with some further information.

Line 77 – Sentence should start with a new line.

Response: This is still part of the caption. It is the figure legend and was formatted according to the author guidelines. To clearer distinguish the main text from figure and table captions, we slightly reduced the text size of the captions.

Photo based survey

Line 94 – Give credits to photoshop

Response: Done.

Line 102 – it took about 5-10 minutes ........ – this part is too casual. May be the estimated time to fill the questionnaire?

Response: Thank you for your suggestion. We revised the sentence accordingly.

Line 103 – What is the process of selecting the respondents? A proper sampling technique?

Response: We used a convenience sampling approach. We are aware of the limited representativeness of the sample. Therefore, we compared our results to those of the two former surveys, which used a stratified sampling approach (see also below).

Why 10 point likert scale? Why not an odd number (with a mid point - neutral response)? Why 10 points? Please explain why these were selected for the study.

Response: There is no consensus among landscape preference studies on the most appropriate rating scale. We used an even scale, because we wanted the respondents to decide whether their preference for an image is more positive or negative. Moreover, response scales from 7 to 10 are most reliable, and respondents prefer the 10-point scale the most (Preston and Colman, 2000). We added this information in the manuscript.

What is the basis for selecting the photographs? Please explain why?

Response: We used landscape photographs that were used in two former surveys to gather landscape preferences in the Central Alps. The photographs depicted different mountain landscapes that were typical for this area. Despite a blue sky, the photographs contained variations in cloud patterns, ranging from almost no clouds to a sky largely covered by clouds, but no threatening cloud atmosphere. From this pool of 49 landscape photographs, we selected 29 photographs with varying cloud cover to not overload the respondents with a too high number of pictures. We revised the description to explain in greater detail the selection of the photographs for this study.

Response: We used the same pictures in our study as the two former survey, which gathered landscape preferences in a comparable way between 2016 and 2019 in five different study areas in the Central Alps. Both studies used a stratified sampling approach based on the demographic distribution in the study areas, i.e., considering key socio-demographic characteristics such as gender, age, origin (residents, tourists), place of living (village, city), and native language (German, Italian). The two surveys, comprising 967 respondents [30] and 384 respondents [35], can therefore be considered to be representative for the general public of the Central Alps. Moreover, in both surveys, differences between socio-demographic groups in preference scores were generally very small, although some significant differences occurred, mostly between people of different language, age, and origin [30]. We revised the description of the methods by adding these details.

Ethics statement

This in the middle of the manuscript? Better to include in the body without having a separate section

Response: During submission, the journal manager asked us to include it as a separate section.

Line 124 - Adult persons or Adults?

Response: We changed to ‘adults’.

Collection of Variables

Give credits to 3M-VAS software. This information is missing in the manuscript. For the softwares used (manufacturer, country) should be mentioned in parentheses

Response: Added.

line 136 – 3 -5 seconds?

Response: Corrected.

photo content analysis – How the structural composition was estimated? Using which software or how? Please explain properly.

Response: We estimated the structural composition of the pictures by visual analysis. To systematically analyze all pictures, we applied the Rule of Thirds and divided the pictures into thirds both horizontally and vertically [56,57]. We then further subdivided the cells to obtain a grid with 3x9 cells for a finer subdivision of the pictures. This grid was overlaid over each picture to determine the area of landscape features and the composition of the landscape in terms of distance zone s (i.e., near zone, middle zone, far zone). All variables were estimated by one researcher, counting the coverage (%) of the features in each cell and summarizing the percentages across all cells. We added this explanation to the methods section. In addition, we created a new figure to illustrate the approach.

We estimated the composition of the landscape in terms of distance zones (i.e., near zone, middle zone, far zone). An illustration of how this was done would be better.

Response: We revised the description to clearly indicate how this was done and created a new figure to illustrate the approach. See also previous comment.

Statistical Analysis

Why backward stepwise linear regression was used? Explain

Response: The backward stepwise approach starts from a full model with all predictors and successively reduces them to obtain a model that best explains the data, while solving problems of multicollinearity and overfitting. We added this explanation to the methods section.

Results

Line 205 – R2

Response: Corrected.

Line 236 – und?

Response: Corrected.

Line 238 – seconds?

Response: Yes, corrected.

Line 238 – give credits right next to the photographs in the description.

Response: Done.

The red and yellow markings in the photographs? What are those?

Response: The red and yellow markings indicate the hotspot area. Hotspots with a probability >=80% are indicated in red, while those with a probability <80 are indicated in yellow. We added an explanation to the figure caption. Moreover, we added a figure to the supplementary material (Figure Sx), explaining the outputs of the eye-tracking software and we refer to this figure in the methods section.

Where is the conclusion? A paper with no conclusion does not sound right.

Response: We added a short conclusion section.

The paper is interesting and a great attempt to contribute to the existing literature. Congratulations on the future as a scholar.

Regards!

Reviewer #3:

Title

Do clouds in landscape photographs influence people’s preferences?

Summary

Overall comments

Response: Thank you for your time and effort to review our manuscript. We appreciate your evaluation and the valuable suggestions on how to improve the manuscript. See also comments below.

Response: We now integrated further information into the introduction to better frame our study and to provide more background information. We also added more details in the methods section to assure replicability of the study. See also comments below.

Response: In addition to having added more details in the methods section, we also revised the description of the results to improve clarity and readability. See also comments below.

The authors must also consider publishing their image sets, analysis scripts, and raw data on the Open Science Framework (or similar platform) such that their findings can be reproduced.

Response: We added all images and raw data to the supplementary material.

Specific comments

Response: We rephrased several parts of the abstract to indicate more clearly the issues that are related to the weather and light conditions in the pictures.

Response: We agree that information on the clouds needs to be introduced in the introduction. We therefore moved the text reflecting on perceptions of clouds from the discussion section to the introduction section, also adding some further information. Furthermore, we revised the methods and results section to provide details on clouds for all photographs. As our aim was to assess the influence of clouds in photographs taken at sunny days (as recommended for studies using photographs to gather landscape preferences), only certain cloud types occurred.

Response: We added a short overview in the introduction session on how landscape preferences are usually assessed. We introduce expert-based and perception-based approaches. We explain how landscape preferences are usually assessed, i.e., photo-based questionnaires, which ask the participants to evaluate different photographs using rating scales or applying discrete choice experiments.

Response: To ensure a minimum sample size for statistical analysis , i.e., to set up a significant regression model, we determined the minimum sample size through statistical power analysis. With a determination coefficient of R² = 0.7, a statistical power of 0.9 and a significance level of α = 0.01, we needed a sample size of at least 68 valid responses. The sample size in other landscape preference studies is highly variable, depending whether they are also analysing differences among socio-demographic groups. Nevertheless, sample sizes between 100 and 200 respondents are very common. We now added an explanation regarding the sample size to the methods section.

Response: We added a short description of the main characteristics of the respondents in the manuscript. We also provide a table with full details in the supplementary material (Table S1).

Response: The respondents completed the questionnaire in an unsupervised setting, using a digital device. There were no time restrictions in viewing and rating the photographs. The participants could also pause the survey by saving their responses and continue at a later time. In the first section of the questionnaire, participants were informed about the scope of the survey and that the participation in the survey was voluntary and anonymous. The participants received neither a compensation nor other incentives. The completion time was not monitored, but participants who were not interested in rating the photographs did not complete the questionnaire. Indeed, 176 people opened the questionnaire, but only 124 also completed it. We added these details to section on the photo-based survey.

Virtual eye tracking was used but few details are provided on how and why this method is appropriate, how does it compare to actual eye tracking? On this point, the authors simply refer to “Schirpke et al.” but I suggest that the reader should be presented with these points without having to refer to another study – were the elements of that study followed exactly? How do the methods here deviate?

Response: 3M-WAS uses an artificial intelligence (AI) algorithm that is based on 30 years of lab-based eye-tracking research to simulate human pre-attentive processing in vision. Based on five visual elements that are recognized to attract human attention (edges, intensity, red-green color contrast, blue-yellow color contrast, and faces), the algorithm predicts what people would be focusing their eyes on in the initial 3-5 seconds of looking at an image, before the conscious brain can react. 3M-VAS claims to be as reliable and effective as eye tracking hardware, predicting pre-attentive vision with around 92% accuracy. The software also overcomes issues related to repeatability of real eye tracking experiments and allows direct comparison of the outputs.

We added this information to the manuscript. We also revised the description of the analysis by adding further details, e.g., the processing of the images, the analysis of the hotspots. This analysis did not deviate from the study done by Schirpke et al. (2022),

Response: To analyse the landscape seen on the photographs in terms of their composition and configuration through landscape metrics, the maps needed to match the photo content. Thus, we had to remove areas from the map that cannot be seen on the photograph. We revised to description of this paragraph to clearer describe our analysis approach.

Response: To include all cases (i.e., photographs) in the analysis, missing values are usually replaced by mean values. Mean values are neutral and do not alter the results. However, we noticed that our variables did not include missing values. We therefore removed the sentence to avoid misleading statements.

Response: We agree that the results from these tests are better placed together with the description of the methods. We now moved them from the results section to the end of the methods section.

#11 To aid interpretation of the results, Figure 2 would be clearer if both A and B images were shown for each example. At present the comparisons being assessed are not clear.

Response: We revised Figure 2 following the suggestion of Reviewer 1. As including both images would have reduced the image size and visibility of the differences, we did not include them here, but we all picture pairs as well as the results of the hotspot analysis have been included in the supplementary material.

Response: We apologize that an explanation was lacking. Yes, these values are mean preference scores across groupings. We now added the following explanation: The values represent mean values of the variables sky and clouds across all picture pairs with decreased (A>B), not changed (A=B), and increased preference score (A<B). We also added tabular data of this figure to the supplementary material (Table S7). Moreover, we tested statistically significant differences using the least significant difference (LSD) post hoc test. We added this in the methods section and in the Table caption.

#13 Furthermore, it is not clear how the percentages reported in line #196 have been deduced. Further clarification is required to help these results be interpreted and reproduced.

Response: The values represent mean values of the variables sky and clouds across all picture pairs without difference in preference score (A=B). We revised this paragraph by correcting the percentages as reported in Figure 3, adding also the percentages for the decreased and increased preference scores, as well as by adding explanations what these represent. See also previous comment.

Response: We agree that the codes are difficult to follow. We therefore replaced them with more easily interpretable labels. A description of the selected variables is also provided in Table 1 in the manuscript, while Table S1 contains also all variables that have been excluded during the regression analysis. We decided to not highlight significant coefficients in Table 2, as all significant at a significance level of p<0.1; and only two coefficients have a significance level < p<0.05.

Response: We revised the text to make the reference to table 2 more evident, but we decided to not include coefficients and p-values in the main text, as these would reduce readability and are redundant to the information provided in Table 2. In the text, we refer to differences between picture pairs (i.e., original picture with clouds, and modified picture without clouds). We therefore did not need to include a binary variable regarding the clouds. We ran only one regression model using the difference between picture pairs as dependent variable. We now made this clearer in the methods section as well as in the caption of Table 2.

Response: We did not test interactions between variables. We revised the text to avoid misleading statements.

Response: This has been empirically tested for all photos, analyzing the pictures according to the Rule of Thirds. We now describe this in the methods section, and we added the results to the supplementary material (Table S8).

#18 I am unclear on how the 40% threshold for cloud cover noted in line #246 has been derived. The discussion section should not be the first place results are reported.

Response: This has been derived from Figure 3. We now report the exact mean values also in results section in the text and revised the discussion section.

Response: We moved the text reflecting on perceptions of clouds from the discussion section to the introduction section, also adding some further information. Furthermore, we revised the methods and results section to provide details on clouds for all photographs. We also added the supplementary Table S9, specifying the main cloud types that are present in the original photos.

Response: People’s perceptions are of course highly subjective and landscape preferences can be very heterogeneous at the individual level. Nevertheless, perception-based approaches are considered to provide reliable results for the general public. We added more information in the introduction about the different approaches, with a focus on perception-based approaches. We also indicate potential differences between different socio-cultural groups. In the discussion section, we inserted a subheading ‘Limitations and future research’ to indicate more clearly the different subsections of the discussion. In this subsection, we also deepened the discussion about the subjectivity of landscape preferences.

Attachment

Submitted filename: Eyetracking-sky_responses.docx

Click here for additional data file.^{(40KB, docx)}

PLoS One. doi: 10.1371/journal.pone.0288424.r003

Decision Letter 1

Chaohai Shen

18 Jun 2023

PONE-D-23-07973R1Assessing landscape aesthetic values: Do clouds in photographs influence people’s preferences?PLOS ONE

Dear Dr. Schirpke,

Please submit your revised manuscript by Aug 02 2023 11:59PM. If you will need more time than this to complete your revisions, please reply to this message or contact the journal office at plosone@plos.org. When you're ready to submit your revision, log on to https://www.editorialmanager.com/pone/ and select the 'Submissions Needing Revision' folder to locate your manuscript file.

Please include the following items when submitting your revised manuscript:

A rebuttal letter that responds to each point raised by the academic editor and reviewer(s). You should upload this letter as a separate file labeled 'Response to Reviewers'.
A marked-up copy of your manuscript that highlights changes made to the original version. You should upload this as a separate file labeled 'Revised Manuscript with Track Changes'.
An unmarked version of your revised paper without tracked changes. You should upload this as a separate file labeled 'Manuscript'.

We look forward to receiving your revised manuscript.

Kind regards,

Chaohai Shen

Academic Editor

PLOS ONE

Journal Requirements:

Please review your reference list to ensure that it is complete and correct. If you have cited papers that have been retracted, please include the rationale for doing so in the manuscript text, or remove these references and replace them with relevant current references. Any changes to the reference list should be mentioned in the rebuttal letter that accompanies your revised manuscript. If you need to cite a retracted article, indicate the article’s retracted status in the References list and also include a citation and full reference for the retraction notice.

[Note: HTML markup is below. Please do not edit.]

Reviewers' comments:

Reviewer's Responses to Questions

Comments to the Author

1. If the authors have adequately addressed your comments raised in a previous round of review and you feel that this manuscript is now acceptable for publication, you may indicate that here to bypass the “Comments to the Author” section, enter your conflict of interest statement in the “Confidential to Editor” section, and submit your "Accept" recommendation.

Reviewer #1: (No Response)

Reviewer #2: All comments have been addressed

Reviewer #3: (No Response)

**********

2. Is the manuscript technically sound, and do the data support the conclusions?

Reviewer #1: Yes

Reviewer #2: Yes

Reviewer #3: Yes

**********

3. Has the statistical analysis been performed appropriately and rigorously?

Reviewer #1: I Don't Know

Reviewer #2: Yes

Reviewer #3: Yes

**********

4. Have the authors made all data underlying the findings in their manuscript fully available?

Reviewer #1: Yes

Reviewer #2: Yes

Reviewer #3: Yes

**********

5. Is the manuscript presented in an intelligible fashion and written in standard English?

Reviewer #1: Yes

Reviewer #2: Yes

Reviewer #3: Yes

**********

6. Review Comments to the Author

Reviewer #1: This paper has been improved with revision. I have only a few remaining comments to further strengthen or clarify the work.

139 I acknowledge that there is no set scale for assessing landscape preferences, and this may be an issue of translation, but I am finding least preferred and most preferred somewhat nonsensical for photos explored individually. Least and most are comparative terms. Could the authors check the best translation, if formal equivalence exists, as I’m sure this must have made more sense in German.

I’m finding the ‘intensity of visual elements’ variable a bit tricky to conceptualize, and indeed quite a few of the eye-tracking variables are a bit hard to conceptualize. Is it density or concentration or diversity? The Table 1 isn’t always helpful, given that the units are mostly described as percentages but the descriptions include things like areas and contributions to probability. 235-238 is not easy to understand and I think these steps should be understandable without looking at the Schirpke reference cited.

Some of the results could be clearer if some additional terms could be added, for instance in line 311 add “with clouds” after “lower”, or in line 323, add “in the” before “pair” (if I’m interpreting correctly). Additionally 325 add “with cloud removal after “photos” at the end of that line.

At the end of 345 you indicate a high intensity of visual elements was associated with pictures without clouds. I can’t seem to find the supplemental data reporting this. I see the difference score regression. Maybe include the relevant stats in the body and clearly note where reported where those models can be found.

Discussion starts with reporting on a hypothesis that I don’t remember being declared earlier.

Small things but still worth mentioning.

80-81 This sentence seems tautological, maybe drop the first part

89-93 Given the paragraph that comes before it seems unclear, but somewhat likely that clouds will alter preferences. This paragraph doesn’t seem to acknowledge the one before. It could also be done by noting that it is not just “attracting attention” that affects preferences but the emotional impact and mood that clouds can bring to an image.

143 use “assessed” rather than obtained

156 Is the estimated time to complete the researchers’ estimation or based on actual completion?

402 using “were” reads as if you found it rather than Svobodova. Maybe “have been”?

433 and elsewhere Using “percentage sky” reads overly technical for discussion. Percentage “of” sky

Reviewer #2: I appreciate the effort you have put in revising the manuscript answering all the comments from authors.

I only have one small doubt to clarify,

Starting from line 138 - You have mentioned that ....... to decide whether their preference is positive or negative

But you have included the likert scale as 1 - least preferred (which is the lowest level of preference and obviously not a negative response). Please explain this incase I misunderstood!

Line 209 - 3M-WAS

Line 213 - ...... their eyes on in the initial..... ???

Other than that all the comments are addressed. Great work.

Reviewer #3: I commend the authors on their revised manuscript. They have addressed all reviewer comments and the resulting paper is much improved and offers a valuable contribution to the field.

I have just one comment I think it would be worth addressing before recommending publication:

The manipulated images are, in some cases, very obviously edited. The blue sky has little texture, and the lighting of the foreground landscape does not match that provided by a typical blue sky. It would be valuable to mention this limitation, and its possible effects on ratings; could participant ratings have reflected (at least partly) negative reactions to Photoshopped scenes?

**********

7. PLOS authors have the option to publish the peer review history of their article (what does this mean?). If published, this will include your full peer review and any attached files.

If you choose “no”, your identity will remain anonymous but your review may still be made public.

Do you want your identity to be public for this peer review? For information about this choice, including consent withdrawal, please see our Privacy Policy.

Reviewer #1: No

Reviewer #2: No

Reviewer #3: No

**********

PLoS One. 2023 Jul 28;18(7):e0288424. doi: 10.1371/journal.pone.0288424.r004

Author response to Decision Letter 1

26 Jun 2023

Reviewer #1:

This paper has been improved with revision. I have only a few remaining comments to further strengthen or clarify the work.

Response: Thank you for your time and effort in re-reviewing our manuscript and your positive feedback. We have also now addressed all additional comments.

Response: Thank you for pointing this out. Indeed, there was no comparison among the pictures, but we asked the people to indicate how much they liked the pictures on a scale from 1 = ‘I don’t like it at all’ to 10 = ‘I like it very much’. We now corrected the description in the manuscript.

Response: 3M-VAS describes “intensity” with “luminance contrast, brightness, black/white contrast”. We added this explanation to the description in Table 1. The variables are expressed in %, as it indicates the overall contribution to the probability of looking at a hotspot. Accordingly, we refined the description of these variables in Table 1.

235-238 is not easy to understand and I think these steps should be understandable without looking at the Schirpke reference cited.

Response: We revised the description of the methodological steps by adding all details. We also added an additional figure to the supplementary material (new Fig. S5), which illustrates the estimation of the contribution of the visual elements.

Response: We added the suggested terms or rephrased the sentences to improve clarity. The sentence in line 345 still referred to Table 2, where the results of the regression analysis are reported. We now added the cross-reference to this table to clearly indicate the underlying data. Moreover, we added another explanation in the table caption to clearly indicate the meaning of the coefficients; i.e., a negative regression coefficient B indicates that this variable has a positive effect on the preference score of photos with clouds.

Discussion starts with reporting on a hypothesis that I don’t remember being declared earlier.

Response: We rephrased the sentence to match our hypothesis presented at the end of the introduction.

Small things but still worth mentioning.

80-81 This sentence seems tautological, maybe drop the first part

Response: We removed the first part of the sentence and slightly rephrased the second part.

Response: Thank you for this suggestion. We now also indicate that preferences are influence by emotions and moods.

143 use “assessed” rather than obtained

Response: Done.

156 Is the estimated time to complete the researchers’ estimation or based on actual completion?

Response: We could not record the time of the participants to complete the questionnaire. However, we measured how long it took us to fill it out. We rephrased the sentence to clarify that we estimated the time.

402 using “were” reads as if you found it rather than Svobodova. Maybe “have been”?

Response: Done.

433 and elsewhere Using “percentage sky” reads overly technical for discussion. Percentage “of” sky

Response: We changed to “percentage of sky” throughout the manuscript.

Reviewer #2:

I appreciate the effort you have put in revising the manuscript answering all the comments from authors.

I only have one small doubt to clarify,

Response: Thank you for your time and effort in re-reviewing our manuscript and your positive feedback. We have also now addressed all additional comments.

Starting from line 138 - You have mentioned that ....... to decide whether their preference is positive or negative. But you have included the likert scale as 1 - least preferred (which is the lowest level of preference and obviously not a negative response). Please explain this incase I misunderstood!

Response: Indeed, the description of the rating scale was wrong, as we asked the people to indicate how much they liked the pictures on a scale from 1 = ‘I don’t like it at all’ to 10 = ‘I like it very much’. In this case, values >= 5 can be interpreted as positive and below 5 as negative. We now corrected the description of the rating scale in the manuscript.

Line 209 - 3M-WAS

Response: Corrected.

Line 213 - ...... their eyes on in the initial..... ???

Response: We revised the sentence to clarify the meaning.

Other than that all the comments are addressed. Great work.

Response: Thank you.

Reviewer #3:

I commend the authors on their revised manuscript. They have addressed all reviewer comments and the resulting paper is much improved and offers a valuable contribution to the field.

I have just one comment I think it would be worth addressing before recommending publication:

Response: Thank you for your time and effort in re-reviewing our manuscript and your positive feedback. We have also now addressed all additional comments.

Response: We agree that this could have been partly influenced the preferences scores. We now added this limitation to the discussion section.

Attachment

Submitted filename: Eyetracking-sky_responses2.docx

Click here for additional data file.^{(19.4KB, docx)}