The Role of Symmetry in the Aesthetics of Residential Building Façades Using Cognitive Science Methods

Round 1

Reviewer 1 Report

The paper is much more clear and concise.  Well done picking up the edits and follow-thru in presentation and logical order.  

Author Response

Dear referee,

Thank you very much for your nice report.

M Ghorbani

Reviewer 2 Report

The premise of this paper is good: it would be useful to know how symmetry is used to assess the aesthetics of building facades. The study discussed in the paper describes an experiment with 21 expert subjects (those with architectural experience) and 17 non-expert subjects. Comparisons are made between the results from the two sets of subjects.

However, there are substantial problems with how the paper presents the experiment, the results, and the conclusions. The experiment is either poorly constructed or poorly described. The results given are partial so the reader cannot determine whether the conclusions are valid. The conclusions that are drawn go well beyond the data that is presented but also, I feel, that there could be better analysis of the results that would lead to a much better paper.

The following are detailed comments on the paper:

1. The abstract needs to describe the results and key conclusions of the paper. The abstract should be a summary of the entire paper. At present it reads like the blurb for a novel, trying to tempt the reader into reading without giving away the ending.
2. The hypotheses in lines 53-56 are not presented as hypotheses that can be disproved. They need to be rewritten. The hypotheses then need to be revisited later in the paper to explain whether they have been supported or disproven - there is nothing in the current paper that does that.
3. Section 2, "Research literature", has a number of oddities. The sentences at lines 62-65 make no sense. The symmetry illustrated in Figure 1 is rotational but the paper deals largely with mirror symmetries. Figure 2 is an example of mirror symmetry but this is not explained in the caption and it bears only a distant relationship to the sort of facades that are handled in the study. Bilateral symmetry (line 105) is the same as mirror symmetry (line 95) - so why is it split out as a separate entry. Line 114 "people lament that this style is boring or monotonous" - citation needed. Lines 114-115 "This type of symmetry means global transmission" is a meaningless statement. Line 125 "It is well-known that beauty is a result of the existence of symmetry [30-32]" directly contradicts line 81 "recent studies cast doubts regarding the assumption that symmetry could be a universal standard of beauty [20]" - please get your story consistent when writing a literature review!
4. When I got to the "Results" section, I felt that I had not yet had a clear description of the experiment that was carried out and what was to be tested. This is likely owing to the hypotheses being unclear (see point 2 above). Lines 323-324 say "The important point in the preferred facades of non-experts is [that] reflection/mirror symmetry [is] significant", but at no point before this was it clear that you had prepared stimuli or an experiment that was explicitly testing for this. So what you have here is a new hypothesis that has arisen from your experimental work, but your experiment was not designed to test this hypothesis so you cannot claim that it is a result of the present experiment.
5. Section 4, "Results", does not present the results well. For example, lines 302-303 imply that the null hypothesis cannot be rejected ("the H0 is true") while line 305 says that it can be rejected. The results in Figure 7 are taken to show that "the existing facades are more in line with the tastes of non-experts", whereas I would take the results to mean that experts have more variability in their preferences than non-experts. Line 325 "Non-experts like this facade because the symmetry can cause fluently processed [67]" - you have offered no evidence that this is the reason for the non-experts choice. You would need to design and run a different experiment to test that.Lines 325-335: these are all interesting conjectures but you were not testing for any of these and you would need to design other experiments to see if your conjectures are true. Lines 307 onwards (and Figures 7 and 8): why do you concentrate on the results for the top five preferred facades rather than presenting results for the full set of 110 facades? Line 370 "balanced facades" - this is the first use of this term; if it is part of the experiment to test for balanced facades then that should have been explained in the hypotheses and the materials and methods section. "top-down" and "bottom-up" were used Section 1, but are not mentioned anywhere in Section 4.
6. Figure 9 makes no sense: The total reaction time must lie between the reaction time for symmetrical and asymmetrical facades because every tested facade lies in one of those sets and the average of the two numbers will lie between them. Please check your data.
7. Lines 401, Figures 10, 11 and 12 - I can see no great difference between the heat maps for experts and non-experts. You say that the scan paths are different. You have given evidence for a single expert and a single non-expert in figures 11 and 12 - does that evidence generalise to all experts and all non-experts?
8. Section 5, "Conclusion", is not a conclusion. It would be better placed as part of the introduction, giving motivation for why such a study should be conducted. The paper needs a clear summary of what has been demonstrated in the experiments.
9. The written English needs some work. It occasionally strays into the realms of being incomprehensible, for example, lines 187-188 "The fact that a participant how and to which point of an image looks at, can be examined through fixations."; or lines 256-258 "The sample method in this experiment is voluntary with random placement, which is controlled be the pretest to eliminate the error due to sampling." Occasionally the wrong word is used, for example, "mutations" in line 203.

Author Response

Dear Referee,

Thank you very much for your nice reports. A point by point answer to the reviewer’s comments and references to corresponding changes in the manuscript follow:

Reviewer #2

The premise of this paper is good: it would be useful to know how symmetry is used to assess the aesthetics of building facades. The study discussed in the paper describes an experiment with 21 expert subjects (those with architectural experience) and 17 non-expert subjects. Comparisons are made between the results from the two sets of subjects.

However, there are substantial problems with how the paper presents the experiment, the results, and the conclusions. The experiment is either poorly constructed or poorly described. The results given are partial so the reader cannot determine whether the conclusions are valid. The conclusions that are drawn go well beyond the data that is presented but also, I feel, that there could be a better analysis of the results that would lead to a much better paper.

The following are detailed comments on the paper:

The abstract needs to describe the results and key conclusions of the paper. The abstract should be a summary of the entire paper. At present it reads like the blurb for a novel, trying to tempt the reader into reading without giving away the ending.

• Answer: Thank you for your attention. Your comment was applied in the abstract- lines 15-27.
• The hypotheses in lines 53-56 are not presented as hypotheses that can be disproved. They need to be rewritten. The hypotheses then need to be revisited later in the paper to explain whether they have been supported or disproven - there is nothing in the current paper that does that.
• Answer: Thanks for your helpful comments, the hypotheses were corrected- Lines 59-63.
• Section 2, "Research literature", has a number of oddities.
•  
• The sentences at lines 62-65 make no sense.
• Answer: The manuscript has been revised and these sentences were removed in the changes made to the article.

 

• The symmetry illustrated in Figure 1 is rotational but the paper deals largely with mirror symmetries.
• Answer: The figure 1 is replaced with three new figures. It should be noted that the first figure only shows that the other types of symmetry can be appeared in the nature.

 

• Figure 2 is an example of mirror symmetry but this is not explained in the caption and it bears only a distant relationship to the sort of facades that are handled in the study.

 

• Answer: Thank you. In the corrections made in the article, we combined several figures in Figure 2 and the description was added in the caption- Line 108-110.

 

• Bilateral symmetry (line 105) is the same as mirror symmetry (line 95) - so why is it split out as a separate entry.

 

• Answer: Added to explain mirror symmetry and this section was removed- Lines 73-76.

 

• Line 114 "people lament that this style is boring or monotonous" - citation needed.

 

• Answer: We edited the sentence and added a related reference in the main text – Lines 87-88.

 

• Lines 114-115 "This type of symmetry means global transmission" is a meaningless statement.

 

• Answer: This sentence was removed in the changes made to the article.

 

• Line 125 "It is well-known that beauty is a result of the existence of symmetry [30-32]" directly contradicts line 81 "recent studies cast doubts regarding the assumption that symmetry could be a universal standard of beauty [20]" - please get your story consistent when writing a literature review!

 

• Answer: Dear referee's comment was very helpful. Carefully applied in the main text- Lines 116-126.

“It is well-known that beauty is a result of the existence of symmetry [30-32]. There is a theory stating that symmetry conveys the image of health and those good genes are present. Symmetry is an indicator of fitness that reflects a healthy nervous and immune system. This is true for infants and adults as well as people from different cultures. Beauty seems to have constants that are widespread among all ages and racial boundaries [33] because the principles of attractiveness in faces have a biological and evolutionary basis [34]. However, the truth is that recent studies cast doubts regarding the assumption that symmetry could be a universal standard of beauty [35] and it seems more likely that aesthetic preferences related to symmetry, as effective factors in the "top-down" process of aesthetic judgment, might be shaped by our own personal and educational experiences [35, 36]. This difference in opinions is a topic that is also examined in this article between two different groups in terms of expertise in the field of architecture.“

 

• When I got to the "Results" section, I felt that I had not yet had a clear description of the experiment that was carried out and what was to be tested. This is likely owing to the hypotheses being unclear (see point 2 above).

 

• Answer: Thank you for your attention. By receiving your useful comment, we have tried to increase the quality of this section and correct the presentation.

 

• Lines 323-324 say "The important point in the preferred facades of non-experts is [that] reflection/mirror symmetry [is] significant", but at no point before this was it clear that you had prepared stimuli or an experiment that was explicitly testing for this. So what you have here is a new hypothesis that has arisen from your experimental work, but your experiment was not designed to test this hypothesis so you cannot claim that it is a result of the present experiment.

 

• Answer: We have completed the description of the test stimuli in the materials and methods section. This study is a part of comprehensive study. There are three factors that we sould study here, they are color, material, and symmetry of facade. In the present paper, only the symmetry factor was presented, and two other factors are discussed in another paper by the author in Persian language. Factors influencing the aesthetic judgment of the facade were also asked through an online questionnaire of participants in the test. A total of 463 items were questioned. In the category of items related to visual rules, which included 12 items, symmetry was ranked 10th among experts and 2th among non-experts.

 

• Section 4, "Results", does not present the results well.
• For example, lines 302-303 imply that the null hypothesis cannot be rejected ("the H0 is true") while line 305 says that it can be rejected.

 

• Answer: According to the text below, the first null hypothesis is related to the levene's test, which is tested before the main test to select the significance level (sig) based on variances- Line 275

To compare the mean response of the two groups to the pleasantness of the images, the T-test of two independent samples was used. In Table 1, levene's test is a test for identifying equality and non-equality of variances, and its statistical assumption is as follows:

H0: σ1² - σ2² = 0 ("the population variances of group 1 and 2 are equal")
H1: σ1² - σ2² ≠ 0 ("the population variances of group 1 and 2 are not equal")

As can be seen, according to levene's test results, the significance level (sig) is higher than 0.05. This means that the H1 is false and the H0 is true. This tells us that we should look at the "Equal variances assumed" row for the t-test results. As a result, the statistical assumption is set as follows:

H0: µ1 = µ2 ("the two-population means are equal")

H1: µ1 ≠ µ2 ("the two-population means are not equal")

Based on the information in Table 1, since p < 0.05 is less than our chosen significance level α = 0.05, we can reject the null hypothesis. This means that the degree of the pleasantness of the façades has a significant difference between experts and non-experts in all façades.

Table 1. Independent Samples Test for all façades

	Levene's Test for Equality of Variances		t-test for Equality of Means
	F	Sig.	t	df	Sig. (2-tailed)	Mean Difference	Std. Error Difference	95% Confidence Interval of the Difference
								Lower	Upper
Equal variances assumed	3.764	.056	-4.640	74	.000	-21.74790	4.68681	-31.08657	-12.40923
Equal variances not assumed			-4.759	73.946	.000	-21.74790	4.56990	-30.85374	-12.64206

 

 

• The results in Figure 7 are taken to show that "the existing facades are more in line with the tastes of non-experts", whereas I would take the results to mean that experts have more variability in their preferences than non-experts.

 

• Answer: Figure 6 shows the preferred façades of each group with the agreement of what percentage of the total participants in that group has been selected, and the results show more non-experts agreement.

 

• Line 325 "Non-experts like this facade because the symmetry can cause fluently processed [67]" - you have offered no evidence that this is the reason for the non-expert’s choice. You would need to design and run a different experiment to test that.

 

• Answer: This sentence was corrected. And based on your interesting comment, the intensity of its impact between the two groups can be examined in another specific study- Lines 290-293.

 

 

• Lines 325-335: these are all interesting conjectures but you were not testing for any of these and you would need to design other experiments to see if your conjectures are true.

 

• Answer: We assume that our results somewhat confirm the results of similar studies in other fields. But according to your useful comment, we will provide a more detailed plan for this review in future studies.

 

 

• Lines 307 onwards (and Figures 7 and 8): why do you concentrate on the results for the top five preferred facades rather than presenting results for the full set of 110 facades?

 

• Answer: Presenting 110 images took up a lot of space in the article. Also, the present study is part of a comprehensive study that for a more detailed study, analyzes were performed on the first five preferred façades of the two groups of experts and non-experts.
• Line 370 "balanced facades" - this is the first use of this term; if it is part of the experiment to test for balanced facades then that should have been explained in the hypotheses and the materials and methods section.

 

• Answer: We meant asymmetric façades. This sentence was removed in the changes made to the article.

 

• "top-down" and "bottom-up" were used Section 1, but are not mentioned anywhere in Section 4.

 

• Answer: We have added the relevant description in the text of the article- Lines 269 & 330-332.

 

• Figure 9 makes no sense: The total reaction time must lie between the reaction time for symmetrical and asymmetrical facades because every tested facade lies in one of those sets and the average of the two numbers will lie between them. Please check your data.

 

• Answer: Thank you very much. Yes, your comment is correct. The results of symmetric and asymmetric façades were presented for only 5 preferred façades of each group, which were wrongly compared to the total façades. The figure was corrected and the results were presented based on 55 symmetrical façades and 55 asymmetric façades- Lines 319-328.

 

• Lines 401, Figures 10, 11 and 12 - I can see no great difference between the heat maps for experts and non-experts. You say that the scan paths are different. You have given evidence for a single expert and a single non-expert in figures 11 and 12 - does that evidence generalise to all experts and all non-experts?

 

• Answer: In the corrections made in the article, a heat map (Figure 9) was provided for all 4 symmetrical facades. In this picture, we can see the differences and similarities between the two groups. In Figure 10, we show the path of the eye movement of an expert and non-expert as an example. This claim has been made for future studies and provides more detailed research capability that may be of interest.

 

• Section 5, "Conclusion", is not a conclusion. It would be better placed as part of the introduction, giving motivation for why such a study should be conducted. The paper needs a clear summary of what has been demonstrated in the experiments.

 

• Answer: Thanks for your very helpful comments. The Conclusion has been revised- Lines 363- 386 . And based on your suggestion, part of the conclusion was added to the introduction- Lines 49-51.

 

• The written English needs some work. It occasionally strays into the realms of being incomprehensible, for example:

 

- lines 187-188 "The fact that a participant how and to which point of an image looks at, can be examined through fixations."

 

• Answer: This sentence was removed in the changes made to the article.

 

• or lines 256-258 "The sample method in this experiment is voluntary with random placement, which is controlled be the pretest to eliminate the error due to sampling."

 

• Answer: The manuscript has been revised. We edit this sentence- Lines 230-231.

 

• Occasionally the wrong word is used, for example, "mutations" in line 203.

 

• Answer: The manuscript has been revised. This word was deleted.

 

Yours sincerely,

Modjtaba Ghorbani

 

Reviewer 3 Report

This is an interesting paper on the role of symmetry in the aesthetics of residential building façades, the analysis includes a description of the pattern of eye movements. There are many aspects that can be improved (on focus and clarity) and I provide a list of suggestions below.

Abstract: it needs at least a sentence on the findings.

Intro. This is long, and it can be made shorter and more focussed. Some figures can be combined. At least one paper on the aesthetics of façades does exist:

            Ruta, N., Mastandrea, S., Penacchio, O., Lamaddalena, S., & Bove, G. (2019). A comparison between preference judgments of curvature and sharpness in architectural façades. Architectural Science Review, 62(2), 171–181.

Perhaps more importantly, there are also two papers on symmetry and eye movements that are relevant, since some of the findings are similar to those reported here (movements along the axis).

            Meso, A. I., Montagnini, A., Bell, J., & Masson, G. S. (2016). Looking for symmetry: Fixational eye movements are biased by image mirror symmetry. Journal of Neurophysiology, 116(3), 1250–1260.

            Makin, A.D.J., Poliakoff, E. Rampone, G. & Bertamini, M. (2020). Spontaneous ocular scanning of visual symmetry is similar during classification and evaluation tasks. i-Perception. DOI: 10.1177/2041669520946356

Methods: some details are not necessary, for example we can assume the EyeLink was calibrated for each participant. In other words this part could be more concise.

Conclusions: The content is too general, please refer more directly to the specific findings here.

 

Author Response

Dear referee,

Thank you very much for your nice reports. We improved the abstract, some figures were combined and we removed some unnecessary sentences. We also added some new references to the paper as you requested. Finally, we referred more directly to the specific findings.

 

Round 2

Reviewer 2 Report

This is a substantial improvement over the first version. Thank you to the authors for the considerable effort that they have invested in improving the presentation. I believe that this is ready for publication, with some small changes to formatting and phrasing.

The new abstract is excellent. It clearly explains what the paper is about and the results that can be found in the paper, which will help future readers to make a clear decision to read the paper.

Recommended corrections

Do not use "the authors" to refer to the authors of other papers, a reader can misunderstand it to mean the authors of the current paper and will only find that they have misunderstood if they look to the reference list. This occurs in seven places:

Line 90: "In [23] the authors examined..." → "Villarroel and Merino [23] examined..."

Line 175: "In the experiment of [58], the authors showed..." → Ruta et al. [58] showed..."

Line 188: "In [61], the authors provided..." → "Yousefi et al. [61] provided..."

Line 193: "Authors in [63] asked questions..." → "Sussmann and Ward [63] asked questions..."

Line 199: "In [65] the authors examined..." → "Makin et al. [65] examined..."

Line 354: "In [4] the authors stated..." → "Locher states [4]..." - note that this is singular rather than plural as there is only one author

Line 364: "In a visual art study [83, 35, 36, 84], the authors concluded..." → "A number of visual arts studies [35,36,83,84] have concluded..."

 

Figure 3: this is an attractive way to present the information but consider whether it is appropriate for a scientific paper.

 

Line 163: "Researches have shown..." → "Research has shown..."

Line 219:"10 to 9 Likert scale" - should probably be "0 to 9 Likert scale"

Line 276: "the H₁ is false and the H₀ is true" → "the null hypothesis cannot be rejected"

Line 288: "existing" - is this the correct word?

Figure 7: please exchange (a) and (b) so that experts are on the top and non-experts on the bottom to make this figure match Figures 6 and 9, otherwise it is very confusing to try to compare Figures 6 and 7.

Line 299: add some text to explicitly say that experts take longer to make a judgement than non-experts - this is an interesting result.

Line 341: "they" → "humans"

Line 369: "in architecture era" - is this the right phrase?

 

Author Response

We corrected all of them in the manuscript.

Reviewer 3 Report

This revision is a clear improvement on the previous. The changes go in the right direction and make the paper clearer. I feel a bit more in this direction could be done still. For example on the language, there are still problematic sentences. Here is one example:

"the authors stated that how much understand and process an object during aesthetic evaluation is easier, it will be more desirable"

In a revised version these language issues could be resolved. I recommend that this process is not rushed, as the final format is important.

Author Response

We polished the paper grammatically and we colored the new sentences by red.

This manuscript is a resubmission of an earlier submission. The following is a list of the peer review reports and author responses from that submission.

Round 1

Reviewer 1 Report

This manuscript evaluates aesthetic and eye-tracing responses of "experts" and "nonexperts" to building facades. The authors raise and address an interesting question with a reasonable experiment, but there are major problems with the presentation of methods, data, interpretation, and review of literature that need to be addressed before publication. 

Top down processes related to culture should be presented at line 37, as well, and at least one reference included. 

At 52-55, the experts and non-experts are ill-defined and should be described, and the hypotheses should be embedded within a discussion of theory instead of simply presenting what was found later in the paper. 

At 64-65 the sentence about symmetry asymmetry is unnecessary. 

At 69 dihedral should be defined. 

90 - does this refer to architecture specifically? please clarify.

107 - "Avoidance of the dualism" is confusing - please clarify.

123-126 - provide references and citations for each sentence, and consider adding counter example or clarification to the symmetry of the face statement. 

159-167 - citations and english editing necessary

cite at 171

204-205 - explain mutations in visual exploration or clarify

227-230 - experts and non-experts are poorly defined. please rephrase and clarify

237-238 explain criteria for dropping 300 photos

238-240 explain why experts were selected, what their qualifications were, how many there were, and their criteria/instructions for choosing the final 110 images. 

243 - how many of each type (pleasant, unpleasant, moderate) image were used? How were differences in expert opinion resolved? 

Were the experts in 243 the ones who later participated in the study? 

248 "single block" instead of "block"

248-249 - randomized or same trial order for each participant? 

254- full screen or partial screen for stimulus images? if partial, what color was background? 

270-294 - rewrite (APA style or equivalent)

Table 1 is unnecessary.

Figure 6 is confusing, please clarify

309-310 rewrite

311 "evaluating aesthetic value." instead of "evaluating aesthetic."

315 define "language of traditional forms"

319-320 describe hierarchy symmetry of the facades and the relevance of composition of the human face (archetype) - this is all confusing; please clarify. 

Figure 9 caption - please append to clarify heat map. 

327-340 - rewrite, add discussion of literature and references appropriate for discussion and conclusions sections.

Reviewer 2 Report

Although the experiment might be interesting the way it is presented and described in the article is not clear and far from good scientific standards. 

Line 240: “The results of the judging of the experts in each category had three levels of pleasant, unpleasant and moderate scores, and the final images of all the categories were selected according to the frequency of the response.”

We don’t know if the 110 out of 1000 images selected were the ones that got most “pleasant” votes or “unpleasant”. One can suspect, but it is not clear.

What is the reasoning behind using this 3 point scale? Why didn’t you use 5, 7 or 10 point (Likert) scale more typical in Visual Quality Assessment Studies.

Methods used for the first step of your research (hypothesis 1) are very much in-line with Visual Quality Assessment Methods such as  Scenic Beauty Estimation Method (SBE) but you don’t seem to refer or relate to any of them.

Line 293: “The results also show that non-experts in selected images have a shorter response time than experts”. This is not discussed. What does it prove? Is it important? Why?

In general the results are poorly described. There is no discussion of the results. I would expect a several solid discussion paragraphs that relate the results to other research but also to aesthetic aspects of architecture and design. What is the role of architectural facade? Where do designers and architects want to have the attention of the eye of the observer? Why would experts eyes focus on entrance of the building (Figure 9) - is this something that is desired and consciously designed by the architect? It is definitely one of the design goals to naturally focus viewer on the entrance of the building. Elements such as the central stairs and raised, central entrance help with that.

The whole “Conclusion” section is non-existent.
Paragraph 328-337 is just a copy of first paragraph 223-232.
Proper, clear and “Conclusion” section has to be added to the article.

The article lacks some sources. Examples - Figure. 1. Lacking source; Lines 109 -110 quotes lacking source.

And last but not least.

line: 229: quote: Sample groups include 38 participants (19 women and 21 experts).

I very much hope this is a typo, a mistake. This looks like the researchers assume that there are two groups of respondents - non experts (women) and experts (men???).
Why is the non-expert group only consisting of women? If the gender of non-experts is mentioned then why there is no mention of the gender of experts? Are the experts both women and man?
Did gender of the respondents play any role in the research?
If all non-experts respondents are female then this research (hypothesis 1) might be as much on the differences between the aesthetic judgement of the residential building façades between males and females.

Reviewer 3 Report

The subject is well covered historically, even though new technologies can now inform it in new ways.  Most research in this area has moved away from the object and the simple surface qualities of a facade (which also engage the multiplicities of 'subjective' notions that change per personal views including education levels).   There are also numerous grammatical errors as well as logical argumentation issues (specious dichotomies). The scientific and technological arguments are stronger than the ideological ones in this case, and are in fact countered by weak presumptions. The authors should raise the levels of inquiry and argumentation to foster new and stronger ways to view this topic within our current technologies and ways of seeing (as well as experiencing) architecture in urban settings.   Since perception is multi-faceted, the research could look into composite senses and kinaesthetic, spatial perceptions guided by the visual technologies.  Other avenues could even supply economic and value systems (reasons for understanding the perceptions of the facade by a general public may be significant).  The conclusions should inform us how we might as designers learn from this (to do design in better ways).