Post-Conflict Urban Landscape Storytelling: Two Approaches to Contemporary Virtual Visualisation of Oral Narratives

Round 1

Reviewer 1 Report

Comments and Suggestions for Authors

This article examines the place of landscape representations in the context of internally displaced persons (due to armed conflict), and the consequences of the temporal evolution of these representations for 3 population groups (residents, IDPs and refuges). The originality of this work lies in the use of 2 methods, the storytelling approach for narrative construction and the AI-powered visualization approach to recreate the memory of lost landscapes. The article is well written. However, at this stage, the authors lack the pedagogy to better demonstrate, support and argue the conclusions of their work. The argumentation would benefit from being accompanied by better quality figures to illustrate the authors' statements. The points of improvement presented below are essential before allowing the publication of this work. I hope that the advices below will help to improve this work.

 

Major comments

Here are some recommendations to try to improve the article:

 ·         Further details are provided in section 2 "Materials and Methods":

o   It would be interesting to provide some quantitative data on the total number of people interviewed for this study, together with brief statistics on the sampling (residency status, gender, age, etc.), in order to better validate quantitatively the desire for diversity in the sampling of the population interviewed.

o   The authors insist on respecting a work ethic in their approach. Prior to conducting their survey, did the authors seek validation of their protocol from an external scientific ethics committee? This is now compulsory for work financed by European funds, sometimes required by the academic institutions to which scientists belong, and made mandatory in the codes of ethics of certain disciplines such as psychology.

o   Authors can specify how many images have been generated and tested using the AI-powered visualization approach.

o   More generally, it is difficult for the reader to have an overall vision of all the stages of the methodology deployed and to follow the successive steps. I suggest that the authors add an overall diagram summarizing all the stages of the methodology, from the narrative interviews, then the generation of virtual landscapes based on 2 AI models based on the discourse of the interviews, to the qualitative evaluation of the images by the respondents.

 ·         Separate the results from the discussion, as mixing them up creates confusion in the authors' arguments.

o   Create a part 3 "Results and interpretation" integrating parts 3.1, 3.2, 3.3 and 3.4.

o   Create a separate "4. Discussion" section, incorporating chapters 3.5, 3.6 and 4 "Research limitations".

 ·         The presentation of the results requires clarification:

o   The 4 layers of the landscape narrative are discussed very briefly (physical, social, cultural and sense of place), but never defined; it is essential that the authors define these 4 components at the beginning of the presentation of the results.

o   About the construction of virtual landscapes, we understand that they were largely built from discourse elements relating to the physical landscape narrative. However, it is not clear whether the other 3 components (social, cultural and sense of place) are also exploited in the modeling of these virtual landscapes.

o   Figures 7 and 8 are of very poor quality and illegible, making it impossible for the reader to correlate the authors' words with the information in these tables.

 ·         Some elements need to be further developed in the discussion and conclusion:

o   The use of 2 AI text-to image models (DALL.E and Bing Image Creator) is never discussed in the article. Why were these 2 models chosen? What were the authors trying to study? It would be interesting to have a critical analysis of the images produced by the 2 models, to know which images were produced with which model, and to know which of the 2 models generates images that elicit greater validation from respondents.

o   There is a lack of analysis of the contribution of the AI-based virtual landscape generation technique in the conclusion, and in the recommendations. This raises the question of the value of coupling the 2 methods 1/ of storytelling approach for narrative construction and 2/ of AI-powered visualization approach.

 

Minor comments

-          Line 806: the reference Soini et al. is missing.

-          Figure B1 is illegible.

Author Response

Please see the attachment.

Author Response File: Author Response.docx

Reviewer 2 Report

Comments and Suggestions for Authors

Your study delves into intriguing aspects concerning innovative strategies for landscape narrative construction and visualization of urban historical and social challenges. Notably, it employs generative AI to gather genuine public perspectives and creatively depict the post-conflict landscape as an alternative analytical instrument. This pioneering approach holds promise for catalyzing diverse avenues of inquiry within the landscape research domain. Additionally, it opens avenues for utilizing creative methodologies in various urban policy formulation and management endeavors. However, enhancing the legibility of the current figures (Figure 1-8) and diagrams by upgrading them to high-resolution images is advisable. Moreover, it would be beneficial to expand upon the delineation of research constraints, particularly given the experimental nature of the methodology employed in this study.

Author Response

Please see the attachment.

Author Response File: Author Response.docx

Reviewer 3 Report

Comments and Suggestions for Authors

This article proposes a storytelling approach for narrative construction and an AI-powered visualisation approach to revive the image of the elusive landscapes.  It indicates that narratives, supported by AI visualisation, are reliable for comprehending landscape transformation and changes in the sense of place. Here are some detailed comments and suggestions for improvement:

1. In the section of methods, it is mentioned that “the representation of each sent image was validated on a 5-point Likert scale: very low, low, medium, high, and very high.”, Is the heat map the result of these? If so, it needs to be mentioned in the Methods section.

2. The white header in the heat map is very blurry, and the resolution needs to be adjusted.

3. Suggest further refining the conclusion section. 

Author Response

Please see the attachment.

Author Response File: Author Response.docx

Round 2

Reviewer 1 Report

Comments and Suggestions for Authors

A major revision work has been conducted  by the authors on the basis of the comments made in my previous review. They were able to provide detailed, well-argued responses to each of the comments. I congratulate them on this work and recommend publication of this article.

Author Response

Dear Reviewer,

Thank you very much for your thorough review of our manuscript. We are grateful for the opportunity to address your comments. Your comments were valuable in improving the quality of our study, and we appreciate your recommendation for publication.