Research Article

The Impact of Prompt Modalities on Architectural Representation in AI-Generated Contemporary House Imagery

Volume: 7 Number: 2 September 22, 2026
TR EN

The Impact of Prompt Modalities on Architectural Representation in AI-Generated Contemporary House Imagery

Abstract

This study investigates how functional, atmospheric, material, and environmental-performance-oriented prompt modalities shape architectural representation in AI-generated images based on a fixed contemporary house reference. It aims to evaluate the capacity of these prompt modalities to make the intended representational layers visible and preserve the reference geometry. The exploratory qualitative design combines a controlled prompt experiment supported by a reference image with qualitative visual content analysis. In the Leonardo AI environment, four exterior facade visuals were generated for each prompt modality type, totaling 16 visuals, using the same reference, model, style, and production parameters. No selection, elimination, or post-editing was applied to the outputs. The images were coded at low, medium, or high levels across seven categories: formal preservation, functional legibility, atmospheric expression, material and surface character, visual representation of environmental performance, architectural–visual consistency, and responsiveness to the prompt modality. The findings show that each prompt modality accentuated its intended representational layer and that responsiveness was high in 14 of the 16 outputs. Formal preservation was low in 15 outputs, whereas architectural–visual consistency was high in 13. Material and surface character extended across all prompt groups, while environmental performance indicators were concentrated in the performance-oriented group. Taken together, the seven coding categories support an interpretive distinction among semantic guidance, geometric continuity, and visual plausibility. Within the tested conditions, prompting serves as an effective linguistic-conceptual tool for determining the focal point of the architectural representation, while providing limited control over the preservation of the reference geometry.

Keywords

Architectural prompt, Architectural representation, Generative artificial intelligence, Prompt modalities, Visual content analysis

Supporting Institution

The author received no financial support for the research, authorship, and/or publication of this article.

Ethical Statement

Ethics committee approval was not obtained because this study does not involve direct human participation or any data requiring ethics committee approval.

References

  1. Albaghajati, Z. M., Bettaieb, D. M., & Malek, R. B. (2023). Exploring text-to-image application in architectural design: Insights and implications. Architecture, Structures and Construction, 3, 475–497. https://doi.org/10.1007/s44150-023-00103-x
  2. Aydin, S., Koçak, S., Yıldız, M., Baydere, G., & Erdoğan, N. (2025). Grand Budapeşte Oteli Filmindeki Otelin İki Döneme ait Tasarım Anlayışı Üzerinden Yapay Zekâ Desteği ile İncelenmesi (Analysis of the Hotel in the Grand Budapest Hotel Movie Based on Design Concepts of Two Periods with Artificial Intelligence Support). EKSEN Dokuz Eylül Üniversitesi Mimarlık Fakültesi Dergisi, 6(2), 72-87. https://doi.org/10.58317/eksen.1642008
  3. Bolojan, D., Vermisso, E., & Yousif, S. (2022). Is language all we need? A query into architectural semantics using a multimodal generative workflow. In Post-carbon: Proceedings of the 27th International Conference of the Association for Computer-Aided Architectural Design Research in Asia (CAADRIA 2022) (Vol. 1, pp. 353–362). Association for Computer-Aided Architectural Design Research in Asia. https://doi.org/10.52842/conf.caadria.2022.1.353
  4. Ching, F. D. K. (2023). Architecture: Form, space, and order (5th ed.). Wiley.
  5. Çınar Kalenderoğlu, S., & Demiröz, M. (2024). Integrating text-to-image AI in architectural design education: Analytical perspectives from a studio experience. Journal of Design Studio, 6(2), 247–258. https://doi.org/10.46474/jds.1526771
  6. DeepL SE. (2026). DeepL Translator. https://www.deepl.com/en/translator
  7. Düzenli, K., & Perdahçı, Z. N. (2023). Metaverse ve NFT’nin mimarlığa etkileri: Geleceğin yapıları nasıl şekillenecek? Eksen Dokuz Eylül Üniversitesi Mimarlık Fakültesi Dergisi, 4(2), 165–182. https://doi.org/10.58317/eksen.1286569
  8. Evans, R. (1997). Translations from drawing to building and other essays. MIT Press.
  9. Ewenstein, B., & Whyte, J. (2009). Knowledge practices in design: the role of visual representations asepistemic objects'. Organization studies, 30(1), 07-30. https://doi.org/10.1177/0170840608083014
  10. Frampton, K. (1995). Studies in tectonic culture: The poetics of construction in nineteenth and twentieth century architecture (J. Cava, Ed.). MIT Press.
APA
Şen, E. (2026). The Impact of Prompt Modalities on Architectural Representation in AI-Generated Contemporary House Imagery. EKSEN Dokuz Eylül Üniversitesi Mimarlık Fakültesi Dergisi, 7(2). https://doi.org/10.58317/eksen.2022584
AMA
1.Şen E. The Impact of Prompt Modalities on Architectural Representation in AI-Generated Contemporary House Imagery. EKSEN Dokuz Eylül Üniversitesi Mimarlık Fakültesi Dergisi. 2026;7(2). doi:10.58317/eksen.2022584
Chicago
Şen, Elif. 2026. “The Impact of Prompt Modalities on Architectural Representation in AI-Generated Contemporary House Imagery”. EKSEN Dokuz Eylül Üniversitesi Mimarlık Fakültesi Dergisi 7 (2). https://doi.org/10.58317/eksen.2022584.
EndNote
Şen E (September 1, 2026) The Impact of Prompt Modalities on Architectural Representation in AI-Generated Contemporary House Imagery. EKSEN Dokuz Eylül Üniversitesi Mimarlık Fakültesi Dergisi 7 2
IEEE
[1]E. Şen, “The Impact of Prompt Modalities on Architectural Representation in AI-Generated Contemporary House Imagery”, EKSEN Dokuz Eylül Üniversitesi Mimarlık Fakültesi Dergisi, vol. 7, no. 2, Sept. 2026, doi: 10.58317/eksen.2022584.
ISNAD
Şen, Elif. “The Impact of Prompt Modalities on Architectural Representation in AI-Generated Contemporary House Imagery”. EKSEN Dokuz Eylül Üniversitesi Mimarlık Fakültesi Dergisi 7/2 (September 1, 2026). https://doi.org/10.58317/eksen.2022584.
JAMA
1.Şen E. The Impact of Prompt Modalities on Architectural Representation in AI-Generated Contemporary House Imagery. EKSEN Dokuz Eylül Üniversitesi Mimarlık Fakültesi Dergisi. 2026;7. doi:10.58317/eksen.2022584.
MLA
Şen, Elif. “The Impact of Prompt Modalities on Architectural Representation in AI-Generated Contemporary House Imagery”. EKSEN Dokuz Eylül Üniversitesi Mimarlık Fakültesi Dergisi, vol. 7, no. 2, Sept. 2026, doi:10.58317/eksen.2022584.
Vancouver
1.Elif Şen. The Impact of Prompt Modalities on Architectural Representation in AI-Generated Contemporary House Imagery. EKSEN Dokuz Eylül Üniversitesi Mimarlık Fakültesi Dergisi. 2026 Sep. 1;7(2). doi:10.58317/eksen.2022584