pith. machine review for the scientific record. sign in

arxiv: 2604.18768 · v1 · submitted 2026-04-20 · 💻 cs.HC

Recognition: unknown

AffectCity: An Empirical Investigation of Complexity, Transparency, and Materiality in Shaping Affective Perception of Building Facades

Chenxi Wang , Haining Ding , Michal Gath-Morad
Authors on Pith no claims yet

Pith reviewed 2026-05-10 03:27 UTC · model grok-4.3

classification 💻 cs.HC
keywords building facadesaffective perceptioncomplexitytransparencymaterialitymediation analysisurban designvalence and arousal
0
0 comments X

The pith

Perceived complexity of building facades is the dominant driver of emotional arousal and valence, with human perception acting as the key link to measurable features.

A machine-rendered reading of the paper's core claim, the machinery that carries it, and where it could break.

The paper introduces a new dataset of facade images rated for arousal and valence to test how complexity, transparency, and materiality shape affective responses. It finds that perceived complexity strongly and positively associates with both arousal and valence, showing stronger effects at higher levels, while machine metrics have weak direct links and require human ratings as mediators. Transparency follows an inverted-U curve with valence, and artificial materials lower pleasantness in line with biophilic ideas. This matters for moving urban design toward models that predict and improve how buildings make people feel daily.

Core claim

Buildings shape how people feel, yet the mechanisms through which specific facade properties drive affective states remain empirically underspecified. Focusing on complexity, transparency, and materiality, perceived complexity emerges as the dominant affective predictor with significant positive associations for arousal and valence and curvilinear amplification at higher levels. Machine-derived metrics show limited direct predictive power; human perceptual evaluation functions as a necessary intermediate layer, with perceived materiality mediating the machine-valence relationship. Transparency exhibits an inverted-U with valence, and artificiality suppresses arousal consistent with biophilic

What carries the argument

Human perceptual evaluation as a necessary intermediate layer between machine-vision-derived surface metrics and affective outcomes, with perceived complexity as the strongest predictor.

If this is right

  • Facade design guidelines can prioritize optimal complexity levels to enhance positive affective responses in urban settings.
  • Machine vision tools for facade analysis must incorporate human perceptual ratings to accurately predict emotional impact.
  • Transparency and materiality can be adjusted based on their specific relationships to valence and arousal for better biophilic outcomes.
  • Cross-context studies can build on the moderate stability of complexity and materiality ratings while accounting for valence's context dependence.

Where Pith is reading between the lines

These are editorial extensions of the paper, not claims the author makes directly.

  • These results imply that cities could use perceptual data to design facades that reduce negative emotions or increase engagement in public spaces.
  • The approach opens paths to integrate affective modeling into architectural software for real-time design feedback.
  • Extending the dataset to more global contexts might reveal cultural differences in how complexity affects valence.

Load-bearing premise

The sample of 86 facade images and 85 participants represents typical affective responses to building facades sufficiently to support the mediation and prediction claims.

What would settle it

Finding no significant mediation effect or no association between perceived complexity and arousal in a replication study using a larger and more diverse set of participants and real-world facade observations would falsify the central claims.

Figures

Figures reproduced from arXiv: 2604.18768 by Chenxi Wang, Haining Ding, Michal Gath-Morad.

Figure 1
Figure 1. Figure 1: Pipeline linking façade features and affective responses. This diagram illustrates the dual pathway from architectural façades to affective experience via formal properties (complexity, trans￾parency, materiality: C, T, M) and human perception towards emotional outcomes (arousal and va￾lence: A, V). Two critical gaps (Gap 1, Gap 2) highlight underexplored transitions between physical form, subjective appra… view at source ↗
Figure 2
Figure 2. Figure 2: Research design of the Cambridge Façade Affect Dataset (CFAD). This schematic outlines the multi-stage construction of the CFAD, integrating machine-derived façade features (C/T/M: complexity, transparency, materiality) with human-rated perceptual and affective responses (A/V: arousal, valence). The dataset comprises an 86-façade corpus processed via computer vision (Step 1), rated by 85 participants in an… view at source ↗
Figure 3
Figure 3. Figure 3: Russell’s Circumplex Model of Affect and modified affect grid. Russell’s [9] Circumplex Model illustrates core affect along two orthogonal dimensions: valence (pleasure–displeasure) and arousal (activation–deactivation). A modified affect grid based on this model was adapted for study purposes to measure emotional responses to architectural façades. Source: https://psu.pb.unizin. org/psych425/chapter/circu… view at source ↗
Figure 4
Figure 4. Figure 4: Dataset protocol for the Cambridge Façade Affect Dataset (CFAD). Two-step protocol for constructing the CFAD image dataset (N = 86), including data collection (Step 1) and standardisation (Step 2). Photographs were taken using a fixed DSLR setup under uniform daylight and minimal urban interference. Images underwent orthographic correction and cropping to ensure consistent alignment, aspect ratio (3:1), an… view at source ↗
Figure 5
Figure 5. Figure 5: Cambridge Façade Affect Dataset (CFAD). Source: Author. 2.1. Computational Measurement of Façade Attributes Three façade attributes, complexity, transparency, and materiality, were extracted computationally from the 86 orthogonally rectified images comprising the Cambridge Façade Affect Dataset (CFAD). All metrics were normalised to a continuous range of [0, 1] to enable systematic cross-façade comparison … view at source ↗
Figure 6
Figure 6. Figure 6: Computational quantification of façade features: complexity, transparency, and materiality. Three architectural variables were quantified from façade imagery using a combination of computer vision tech￾niques. Complexity was derived via OpenCV edge density and fractal dimension (D) methods. Trans￾parency was estimated by calculating the window-to-wall ratio (WWR) using YOLO and semantic segmen￾tation (e.g.… view at source ↗
Figure 7
Figure 7. Figure 7: Sample pipeline of machine-derived façade feature extraction (CFAD ID: 11). Example showcasing automated processing of a single façade image to extract three visual metrics. Complexity is computed via edge density, transparency via window detection and area ratio, and materiality via class-based segmentation of natural versus artificial materials. Final scores (0–1) are displayed for each variable. This im… view at source ↗
Figure 8
Figure 8. Figure 8: Survey workflow for online subjective evaluation. Participants (Np = 85) completed an online survey comprising two sections: background profiling (demographics, cognitive traits) and façade evaluation. Each participant was randomly assigned 10 façade images from the CFAD dataset and asked to rate them along three perceptual dimensions(complexity, transparency, and materiality) as well as two affective dime… view at source ↗
Figure 9
Figure 9. Figure 9: Self-Assessment Manikin (SAM) for affect measurement. The Self-Assessment Manikin (SAM) is a nonverbal pictorial assessment technique that directly measures the dimensions of affec￾tive response. The top row depicts the valence scale (from unpleasant to pleasant), and the bottom row shows the arousal scale (from calm to excited). Source: Author. The SAM scale [56] is a non-verbal, pictorial instrument that… view at source ↗
Figure 10
Figure 10. Figure 10: Four emotional categories are defined by the combination of valence and arousal dimensions: high valence and high arousal (pleasant and activating); high valence but low arousal (pleasant and calm); low valence and low arousal (unpleasant and calm); low valence but high arousal (unpleasant and activating). Colour coding corresponds to affective tone used throughout the figures. Source: Author. https://doi… view at source ↗
Figure 11
Figure 11. Figure 11: Mapping of architectural façades in a two-dimensional affective space. Each circle represents one CFAD façade image (N = 86), positioned by its mean valence (x-axis) and arousal (y-axis) ratings collected from 85 participants on a 5-point Likert scale (1 = low, 5 = high). The space is divided into four quadrants reflecting classic emotional categories. Circle colour indicates the quadrant’s affective tone… view at source ↗
Figure 12
Figure 12. Figure 12: Mixed-effects regression models and non-linear effects linking perceived façade features to affective responses. Scatterplots show the relationship between three human-rated façade attributes—complexity (blue), transparency (green), and materiality (red)— and affective responses: valence (top row) and arousal (bottom row). Data are drawn from the CFAD (N = 86 façades), each rated by ≥12 participants (Np =… view at source ↗
Figure 13
Figure 13. Figure 13: Standardised regression coefficients for the effects of perceived façade features on arousal and valence. Complexity emerged as the strongest positive predictor for both affective dimensions; transparency showed weaker yet significant positive effects; materiality had negative coefficients for both arousal and valence, indicating that more natural materials (lower materiality score) were associated with g… view at source ↗
Figure 14
Figure 14. Figure 14: Boxplots showing the effects of perceived complexity, transparency, and materiality on participants’ affective responses: (a) arousal (excitement) and (b) valence (pleasantness). Each dot represents an individual Likert rating (1–5), jittered for clarity. Boxes indicate interquartile ranges (IQR) with medians. Only groups with significantly higher median scores compared to others (via Kruskal–Wallis test … view at source ↗
Figure 15
Figure 15. Figure 15: Scatter plots showing the relationship between machine-derived façade features and affective responses. Each subplot il￾lustrates the predictive relationship between a machine-derived façade feature—complexity, transparency, or materiality—and one of two affective dimensions: valence (top row) or arousal (bottom row). Machine ratings were computed via computer vision pipelines and normalised between 0 and… view at source ↗
Figure 16
Figure 16. Figure 16: Machine–human agreement across three façade dimensions. Scatter plots show the cor￾relation between machine-derived and human-perceived scores for (a) complexity, (b) transparency, and (c) materiality, all normalised to a 0–1 scale. Red lines represent linear fits; grey dashed lines indicate identity lines. R 2 and Spearman’s ρ are shown for each dimension. Source: Author [PITH_FULL_IMAGE:figures/full_fi… view at source ↗
Figure 17
Figure 17. Figure 17: Correlation between online and on-site affective ratings. Scatter plots with regression lines and identity lines (dashed) illustrate consistency levels. Valence scores show significant corre￾lation (ρ = 0.58, p = 0.015), while arousal scores exhibit weaker alignment (ρ = 0.41, p = 0.107). Source: Author. Computational–perceptual alignment. Among the three façade attributes, material￾ity showed the stronge… view at source ↗
Figure 18
Figure 18. Figure 18: Comparison between online and on-site affective ratings. Paired plots for valence and arousal showing individual changes across 15 façades, with corresponding boxplots displaying cen￾tral tendency and dispersion of scores. Valence ratings were significantly higher in the on-site con￾dition relative to online evaluations, while arousal ratings showed no significant difference between conditions. Source: Au… view at source ↗
read the original abstract

Buildings shape how people feel, yet the mechanisms through which specific facade properties drive affective states remain empirically underspecified. Here we introduce the Cambridge Facade Affect Dataset (CFAD), 86 orthogonally rectified facade images annotated with continuous arousal and valence ratings from 85 participants, and establish a validated pipeline linking machine-vision-derived surface metrics to human affective responses. Focusing on three quantifiable attributes, complexity, transparency (window-to-wall ratio), and materiality (proportion of natural versus artificial surface composition), we show that perceived complexity is the dominant affective predictor, with significant positive associations for both arousal (beta = 0.507, p < 0.001) and valence (beta = 0.376, p < 0.001) and a curvilinear amplification at higher complexity levels. Transparency exhibits an inverted-U relationship with valence, while increasing surface artificiality suppresses arousal and reduces pleasantness consistent with biophilic response theory. Critically, machine-derived metrics show limited direct predictive power over affective outcomes; mediation analyses reveal that human perceptual evaluation functions as a necessary intermediate layer, with perceived materiality significantly mediating the machine-valence relationship (indirect effect = -0.205, p = 0.003). Cross-context validation demonstrates moderate stability of complexity and materiality ratings across image-based and in-situ conditions, while affective responses, particularly valence, exhibit significant context-dependence (ICC = 0.332). These findings advance facade research from descriptive morphological analysis toward predictive, perception-grounded modelling, and provide an empirically validated basis for affect-informed design of the urban environment.

Editorial analysis

A structured set of objections, weighed in public.

Desk editor's note, referee report, simulated authors' rebuttal, and a circularity audit. Tearing a paper down is the easy half of reading it; the pith above is the substance, this is the friction.

Referee Report

2 major / 3 minor

Summary. The paper introduces the Cambridge Facade Affect Dataset (CFAD) with 86 orthogonally rectified facade images and continuous arousal/valence ratings from 85 participants. It tests three facade attributes—complexity, transparency (window-to-wall ratio), and materiality (natural vs. artificial surface proportion)—and reports that perceived complexity is the dominant predictor (arousal beta=0.507, p<0.001; valence beta=0.376, p<0.001) with curvilinear amplification at high levels. Transparency shows an inverted-U with valence, artificiality suppresses arousal, machine metrics have weak direct effects, and human perceptual ratings mediate the machine-to-affect link (e.g., perceived materiality indirect effect=-0.205, p=0.003). Cross-context validation shows moderate stability for complexity/materiality but context-dependence for valence (ICC=0.332).

Significance. If the statistical results and mediation models hold, the work supplies a reusable dataset and validated pipeline that moves facade research from purely morphological description to perception-grounded, predictive modeling. The explicit mediation tests, quadratic terms, and in-situ vs. image comparison provide falsifiable links between machine-vision metrics and affective outcomes, with direct relevance to biophilic design and urban HCI.

major comments (2)
  1. [§4.3] §4.3 (Mediation models): the reported indirect effect for perceived materiality on machine-valence is load-bearing for the central claim that human perception is a necessary intermediate layer, yet the manuscript does not report the full set of path coefficients, standard errors for the indirect effect, or bootstrap confidence intervals; without these the mediation conclusion cannot be fully evaluated.
  2. [§3.2] §3.2 (Image selection): the claim that the 86 CFAD images support general inferences about facade affect rests on the orthogonal rectification and diversity criteria, but the exact stratification by complexity/transparency/materiality bins and any exclusion rules for atypical facades are not quantified, weakening the representativeness argument for the mediation and cross-context results.
minor comments (3)
  1. [Table 2] Table 2: the quadratic term for complexity-arousal is significant but the manuscript does not state whether the model includes all lower-order terms and interactions; add the full regression table with all predictors.
  2. [Figure 4] Figure 4: the curvilinear plots lack participant-level scatter or 95% confidence bands around the fitted lines, reducing visual assessment of effect size and outlier influence.
  3. [§5.1] §5.1: the ICC=0.332 for valence context-dependence is reported without the corresponding ICCs for arousal or the exact formula (one-way vs. two-way random effects); clarify the ICC type and interpretation.

Simulated Author's Rebuttal

2 responses · 0 unresolved

We thank the referee for the constructive and detailed feedback. The comments highlight opportunities to increase transparency in our statistical reporting and dataset documentation, which we will address in the revision.

read point-by-point responses

  1. Referee: [§4.3] §4.3 (Mediation models): the reported indirect effect for perceived materiality on machine-valence is load-bearing for the central claim that human perception is a necessary intermediate layer, yet the manuscript does not report the full set of path coefficients, standard errors for the indirect effect, or bootstrap confidence intervals; without these the mediation conclusion cannot be fully evaluated.

    Authors: We agree that the full mediation statistics are required to allow readers to evaluate the central claim. The manuscript currently reports only the indirect effect and its p-value. In the revised version we will expand §4.3 to include the complete path coefficients (a, b, c′), their standard errors, and 95% bootstrap confidence intervals for the indirect effects, presented in a new table. revision: yes

  2. Referee: [§3.2] §3.2 (Image selection): the claim that the 86 CFAD images support general inferences about facade affect rests on the orthogonal rectification and diversity criteria, but the exact stratification by complexity/transparency/materiality bins and any exclusion rules for atypical facades are not quantified, weakening the representativeness argument for the mediation and cross-context results.

    Authors: Section 3.2 describes the orthogonal rectification procedure and the intent to sample across the three attributes, but does not provide numerical bin counts or explicit exclusion criteria. We will revise §3.2 to add a quantitative summary of the stratification (number of images per complexity, transparency, and materiality bin) together with the precise exclusion rules, either in the main text or as a supplementary table. revision: yes

Circularity Check

0 steps flagged

No significant circularity in empirical derivation chain

full rationale

The manuscript reports an empirical investigation based on a new dataset (CFAD: 86 images rated by 85 participants), machine-vision surface metrics, and standard statistical procedures (linear and quadratic regressions yielding reported betas, plus mediation models with indirect effects). No equations or derivations are present that reduce any claimed result to its inputs by construction. Self-citations of prior facade work, if present, are not load-bearing for the central claims, which rest on the fresh data collection, within-subjects ratings, ICC values, and mediation estimates. The work is self-contained against external benchmarks and does not exhibit any of the enumerated circularity patterns.

Axiom & Free-Parameter Ledger

0 free parameters · 2 axioms · 0 invented entities

Central claims rest on standard statistical assumptions for linear and mediation models applied to survey ratings; no free parameters are introduced beyond those estimated from the 85-participant data, and no new entities are postulated.

axioms (2)
  • domain assumption Participant arousal and valence ratings are reliable and valid measures of affective response to static facade images.
    Invoked throughout the results section when interpreting beta coefficients and mediation effects.
  • domain assumption The 86 selected images adequately sample the range of complexity, transparency, and materiality in real building facades.
    Required for generalizing the dominance of complexity and the mediation findings.

pith-pipeline@v0.9.0 · 5596 in / 1396 out tokens · 40975 ms · 2026-05-10T03:27:41.467859+00:00 · methodology

discussion (0)

Sign in with ORCID, Apple, or X to comment. Anyone can read and Pith papers without signing in.

Reference graph

Works this paper leans on

65 extracted references · 52 canonical work pages · 1 internal anchor

  1. [1]

    City living and urban upbringing affect neural social stress processing in humans

    Lederbogen, F.; Kirsch, P .; Haddad, L.; Streit, F.; Tost, H.; Schuch, P .; Wüst, S.; Pruessner, J.C.; Rietschel, M.; Deuschle, M.; et al. City living and urban upbringing affect neural social stress processing in humans. Nature 2011, 474, 498–501. https://doi.org/10.1038/nature10190

    work page doi:10.1038/nature10190 2011

  2. [2]

    Neural correlates of individual differences in affective benefit of real-life urban green space exposure

    Tost, H.; Reichert, M.; Braun, U.; Reinhard, I.; Peters, R.; Lautenbach, S.; Meyer-Lindenberg, A. Neural correlates of individual differences in affective benefit of real-life urban green space exposure. Nature Neuroscience 2015, 18, 799–805. https://doi.org/10.1038/nn.4035

    work page doi:10.1038/nn.4035 2015

  3. [3]

    Green infrastructure and health

    Nieuwenhuijsen, M.J. Green infrastructure and health. Annual Review of Public Health 2021, 42, 317–328. https://doi.org/10.1146/annurev-publhealth-090419-102511

    work page doi:10.1146/annurev-publhealth-090419-102511 2021

  4. [4]

    Neuroaesthetics

    Chatterjee, A.; Vartanian, O. Neuroaesthetics. T rends in Cognitive Sciences 2014, 18, 370–375. https://doi.org/10.1016/j.tics.2014.03.003

    work page doi:10.1016/j.tics.2014.03.003 2014

  5. [5]

    Impact of contour on aesthetic judgments and approach-avoidance decisions in architecture

    Vartanian, O.; Navarrete, G.; Chatterjee, A.; Fich, L.B.; Leder, H.; Modroño, C.; Nadal, M.; Rostrup, N.; Skov , M. Impact of contour on aesthetic judgments and approach-avoidance decisions in architecture. Proceedings of the National Academy of Sciences 2013, 110, 10446–10453. https://doi.org/10.1073/pnas.1301227110

    work page doi:10.1073/pnas.1301227110 2013

  6. [6]

    Buildings, beauty , and the brain: A neuroscience of architectural experience

    Coburn, A.; Vartanian, O.; Chatterjee, A. Buildings, beauty , and the brain: A neuroscience of architectural experience. Journal of Cognitive Neuroscience 2017, 29, 1521–1531. https://doi.org/ 10.1162/jocn_a_01146. https://doi.org/10.3390/buildings1010000 V ersion April 22, 2026 submitted to Buildings 30 of 32

    work page doi:10.1162/jocn_a_01146 2017

  7. [8]

    Humanise: A maker’s guide to designing our cities ; Penguin, 2023

    Heatherwick, T. Humanise: A maker’s guide to designing our cities ; Penguin, 2023

    2023

  8. [9]

    A Circumplex Model of Affect

    Russell, J.A. A circumplex model of affect. Journal of Personality and Social Psychology 1980, 39, 1161–1178. https://doi.org/10.1037/h0077714

    work page doi:10.1037/h0077714 1980

  9. [10]

    Core affect, prototypical emotional episodes, and other things called emotion: Dissecting the elephant

    Russell, J.A.; Barrett, L.F. Core affect, prototypical emotional episodes, and other things called emotion: Dissecting the elephant. Journal of Personality and Social Psychology 1999, 76, 805–819. https://doi.org/10.1037/0022-3514.76.5.805

    work page doi:10.1037/0022-3514.76.5.805 1999

  10. [11]

    The theory of constructed emotion: An active inference account of interoception and categorization

    Barrett, L.F. The theory of constructed emotion: An active inference account of interoception and categorization. Social Cognitive and Affective Neuroscience 2017, 12, 1–23. https://doi.org/ 10.1093/scan/nsw154

    work page doi:10.1093/scan/nsw154 2017

  11. [12]

    Environmental neuroscience unravels the pathway from the physical environment to mental health

    Kühn, S.; Gallinat, J. Environmental neuroscience unravels the pathway from the physical environment to mental health. Nature Mental Health 2024, 2, 263–269. https://doi.org/10.103 8/s44220-023-00137-6

    2024

  12. [13]

    On the relationship between emotion and cognition

    Pessoa, L. On the relationship between emotion and cognition. Nature Reviews Neuroscience 2008, 9, 148–158. https://doi.org/10.1038/nrn2317

    work page doi:10.1038/nrn2317 2008

  13. [14]

    Urban green spaces and health: A review of evidence, 2016

    World Health Organization Regional Office for Europe. Urban green spaces and health: A review of evidence, 2016

    2016

  14. [15]

    Psychological and neural responses to architectural interiors

    Coburn, A.; Vartanian, O.; Kenett, Y .N.; Nadal, M.; Hartung, F.; Hayn-Leichsenring, G.; Navar- rete, G.; González Mora, J.L.; Chatterjee, A. Psychological and neural responses to architectural interiors. Cortex 2020, 126, 217–241. https://doi.org/10.1016/j.cortex.2020.01.009

    work page doi:10.1016/j.cortex.2020.01.009 2020

  15. [16]

    Measuring the unmeasurable: Urban design qualities related to walkabil- ity

    Ewing, R.; Handy , S. Measuring the unmeasurable: Urban design qualities related to walkabil- ity . Journal of Urban Design 2009, 14, 65–84. https://doi.org/10.1080/13574800802451155

    work page doi:10.1080/13574800802451155 2009

  16. [17]

    Deep learning the city: Quantifying urban perception at a global scale

    Dubey , A.; Naik, N.; Parikh, D.; Raskar, R.; Hidalgo, C.A. Deep learning the city: Quantifying urban perception at a global scale. In Proceedings of the Proceedings of the European Confer- ence on Computer Vision. Springer, 2016, pp. 196–212. https://doi.org/10.1007/978-3-319-46 448-0_12

    work page doi:10.1007/978-3-319-46 2016

  17. [18]

    Using deep learning to quantify the beauty of outdoor places

    Seresinhe, C.I.; Preis, T.; Moat, H.S. Using deep learning to quantify the beauty of outdoor places. Royal Society Open Science 2017, 4, 170170. https://doi.org/10.1098/rsos.170170

    work page doi:10.1098/rsos.170170 2017

  18. [19]

    Proceedings of the Royal Society of London

    Marr, D.; Hildreth, E. Theory of edge detection. Proceedings of the Royal Society of London. Series B: Biological Sciences 1980, 207, 187–217. https://doi.org/10.1098/rspb.1980.0020

    work page doi:10.1098/rspb.1980.0020 1980

  19. [20]

    What makes Paris look like Paris? ACM T ransactions on Graphics2012, 31, 1–9

    Doersch, C.; Singh, S.; Gupta, A.; Sivic, J.; Efros, A.A. What makes Paris look like Paris? ACM T ransactions on Graphics2012, 31, 1–9. https://doi.org/10.1145/2185520.2185597

    work page doi:10.1145/2185520.2185597

  20. [21]

    Fractals, skylines, nature and beauty

    Stamps, A.E. Fractals, skylines, nature and beauty . Landscape and Urban Planning 2002, 60, 163–

    2002

  21. [22]

    https://doi.org/10.1016/S0169-2046(02)00054-3

    work page doi:10.1016/s0169-2046(02)00054-3 2046

  22. [24]

    Adaptive façades for emotionally enriching indoor environments

    Beatini, V .; Pantilimonescu, F.; Djebbara, Z.; Drı¸ scu, M.C. Adaptive façades for emotionally enriching indoor environments. Journal of Building Engineering 2024, 98, 111472. https://doi. org/10.1016/j.jobe.2024.111472

    work page doi:10.1016/j.jobe.2024.111472 2024

  23. [25]

    Urban design aesthetics: The evaluative qualities of building exteriors

    Nasar, J.L. Urban design aesthetics: The evaluative qualities of building exteriors. Environment and Behavior 1994, 26, 377–401. https://doi.org/10.1177/001391659402600305

    work page doi:10.1177/001391659402600305 1994

  24. [26]

    Architecture and engineering students’ eval- uations of house façades: Preference, complexity and impressiveness

    Akalın, A.; Yıldırım, K.; Wilson, C.; Kılıço ˘ glu, Ö. Architecture and engineering students’ eval- uations of house façades: Preference, complexity and impressiveness. Journal of Environmental Psychology 2009, 29, 124–132. https://doi.org/10.1016/j.jenvp.2008.05.005

    work page doi:10.1016/j.jenvp.2008.05.005 2009

  25. [27]

    Architectural experience: Clar- ifying its central components and their relation to core affect with a set of first-person-view videos

    Gregorians, L.; Fernandez V elasco, P .; Zisch, F.E.; Spiers, H.J. Architectural experience: Clar- ifying its central components and their relation to core affect with a set of first-person-view videos. bioRxiv 2022. Preprint, https://doi.org/10.1101/2022.04.05.487021

    work page doi:10.1101/2022.04.05.487021 2022

  26. [28]

    Architectural variation, building height, and the restorative quality of urban residential streetscapes

    Lindal, P .J.; Hartig, T. Architectural variation, building height, and the restorative quality of urban residential streetscapes. Journal of Environmental Psychology 2013, 33, 26–36. https: //doi.org/10.1016/j.jenvp.2012.09.003

    work page doi:10.1016/j.jenvp.2012.09.003 2013

  27. [29]

    Aesthetics and Psychobiology; Appleton-Century-Crofts, 1971

    Berlyne, D.E. Aesthetics and Psychobiology; Appleton-Century-Crofts, 1971. https://doi.org/10.3390/buildings1010000 V ersion April 22, 2026 submitted to Buildings 31 of 32

    work page doi:10.3390/buildings1010000 1971

  28. [30]

    Complexity , liking and familiarity: Architecture and non-architecture Turkish students’ assessments of traditional and modern house façades

    ˙Imamo ˘ glu, Ç. Complexity , liking and familiarity: Architecture and non-architecture Turkish students’ assessments of traditional and modern house façades. Journal of Environmental Psy- chology 2000, 20, 5–16. https://doi.org/10.1006/jevp.1999.0155

    work page doi:10.1006/jevp.1999.0155 2000

  29. [31]

    The Ecological Approach to Visual Perception ; Houghton Mifflin, 1979

    Gibson, J.J. The Ecological Approach to Visual Perception ; Houghton Mifflin, 1979

    1979

  30. [32]

    The restorative benefits of nature: Toward an integrative framework

    Kaplan, S. The restorative benefits of nature: Toward an integrative framework. Journal of Environmental Psychology 1995, 15, 169–182. https://doi.org/10.1016/0272-4944(95)90001-2

    work page doi:10.1016/0272-4944(95)90001-2 1995

  31. [33]

    Stress recovery during exposure to natural and urban environments

    Ulrich, R.S.; Simons, R.F.; Losito, B.D.; Fiorito, E.; Miles, M.A.; Zelson, M. Stress recovery during exposure to natural and urban environments. Journal of Environmental Psychology 1991, 11, 201–230. https://doi.org/10.1016/S0272-4944(05)80184-7

    work page doi:10.1016/s0272-4944(05)80184-7 1991

  32. [34]

    Cognitive Psychology; Appleton-Century-Crofts, 1967

    Neisser, U. Cognitive Psychology; Appleton-Century-Crofts, 1967

    1967

  33. [35]

    On the primacy of affect

    Zajonc, R.B. On the primacy of affect. American Psychologist 1984, 39, 117–123. https://doi. org/10.1037/0003-066X.39.2.117

    work page doi:10.1037/0003-066x.39.2.117 1984

  34. [36]

    Emotion and Adaptation; Oxford University Press, 1991

    Lazarus, R.S. Emotion and Adaptation; Oxford University Press, 1991

    1991

  35. [37]

    What are emotions? and how can they be measured? Social science information , 44(4): 695–729, 2005

    Scherer, K.R. What are emotions? And how can they be measured? Social Science Information 2005, 44, 695–729. https://doi.org/10.1177/0539018405058216

    work page doi:10.1177/0539018405058216 2005

  36. [38]

    The urban brain: Analysing outdoor physical activity with mobile EEG

    Aspinall, P .; Mavros, P .; Coyne, R.; Roe, J. The urban brain: Analysing outdoor physical activity with mobile EEG. British Journal of Sports Medicine 2013, 49, 272–276. https://doi.org/10.1136/ bjsports-2012-091877

    2013

  37. [39]

    Emotional responses to the visual patterns of urban streets: Evidence from physiological and subjective indicators

    Zhang, Z.; Zhuo, K.; Wei, W.; Li, F.; Yin, J.; Xu, L. Emotional responses to the visual patterns of urban streets: Evidence from physiological and subjective indicators. International Journal of Environmental Research and Public Health 2021, 18, 9677. https://doi.org/10.3390/ijerph18189 677

    work page doi:10.3390/ijerph18189 2021

  38. [40]

    The impact of architectural form on physiological stress: a systematic review

    Valentine, C. The impact of architectural form on physiological stress: a systematic review. Frontiers in Computer Science 2024, 5, 1237531. https://doi.org/10.3389/fcomp.2023.1237531

    work page doi:10.3389/fcomp.2023.1237531 2024

  39. [41]

    A novel mathematical model to measure individuals’ perception of the symmetry level of building façades

    Aydın, Y .C.; Mirzaei, P .A. A novel mathematical model to measure individuals’ perception of the symmetry level of building façades. Architectural Engineering and Design Management 2022, 18, 261–278. https://doi.org/10.1080/17452007.2020.1862042

    work page doi:10.1080/17452007.2020.1862042 2022

  40. [42]

    Role of physical attributes of preferred building façades on perceived visual complexity: A discrete choice experiment

    Hashemi Kashani, S.M.; Pazhouhanfar, M.; van Oel, C.J. Role of physical attributes of preferred building façades on perceived visual complexity: A discrete choice experiment. Environment, Development and Sustainability 2023, 25, 9458–9477. https://doi.org/10.1007/s10668-023-02980 -0

    work page doi:10.1007/s10668-023-02980 2023

  41. [43]

    Tall buildings and the urban skyline: The effect of visual com- plexity on preferences

    Heath, T.; Smith, S.G.; Lim, B. Tall buildings and the urban skyline: The effect of visual com- plexity on preferences. Environment and Behavior 2000, 32, 541–556. https://doi.org/10.1177/ 00139160021972658

    2000

  42. [44]

    The aesthetic advantage of greener cities: Measuring affective perception of urban streetscapes

    Hollander, J.B.; Anderson, E.C. The aesthetic advantage of greener cities: Measuring affective perception of urban streetscapes. Urban Forestry & Urban Greening 2020, 48, 126576. https: //doi.org/10.1016/j.ufug.2019.126576

    work page doi:10.1016/j.ufug.2019.126576 2020

  43. [45]

    The influence of emotional response and aesthetic perception of shopping mall façade color on entry decisions

    Zhu, Z.; Liu, Y .; Chen, Y . The influence of emotional response and aesthetic perception of shopping mall façade color on entry decisions. Buildings 2024, 14, 2302. https://doi.org/10.3 390/buildings14082302

    2024

  44. [46]

    Affective response to architecture: Investigating human reaction to spaces with different geometry

    Shemesh, A.; Talmon, R.; Karp, O.; Amir, I.; Bar, M.; Grobman, Y .J. Affective response to architecture: Investigating human reaction to spaces with different geometry . Architectural Science Review 2017, 60, 116–125. https://doi.org/10.1080/00038628.2016.1266597

    work page doi:10.1080/00038628.2016.1266597 2017

  45. [47]

    Research on the construction and application of a SVM-based quantification model for streetscape visual complexity

    Zhao, J.; Suo, W. Research on the construction and application of a SVM-based quantification model for streetscape visual complexity . Land 2024, 13, 1953. https://doi.org/10.3390/land1 3111953

    work page doi:10.3390/land1 2024

  46. [48]

    Berlyne revisited: Evidence for the multifaceted nature of hedonic tone in the appreciation of paintings and music

    Marin, M.M.; Leder, H. Berlyne revisited: Evidence for the multifaceted nature of hedonic tone in the appreciation of paintings and music. Frontiers in Human Neuroscience 2016, 10, 536. https://doi.org/10.3389/fnhum.2016.00536

    work page doi:10.3389/fnhum.2016.00536 2016

  47. [49]

    Processing fluency and aesthetic pleasure: Is beauty in the perceiver’s processing experience? Personality and Social Psychology Review 2004, 8, 364–382

    Reber, R.; Schwarz, N.; Winkielman, P . Processing fluency and aesthetic pleasure: Is beauty in the perceiver’s processing experience? Personality and Social Psychology Review 2004, 8, 364–382. https://doi.org/10.1207/s15327957pspr0804_3

    work page doi:10.1207/s15327957pspr0804_3 2004

  48. [50]

    Influence of complexity and Gestalt principles on aes- thetic preferences for building façades: An eye-tracking study

    Beder, D.; Pelowski, M.; ˙Imamo ˘ glu, Ç. Influence of complexity and Gestalt principles on aes- thetic preferences for building façades: An eye-tracking study . Journal of Eye Movement Research 2024, 17, 1–15. https://doi.org/10.16910/jemr.17.2.4. https://doi.org/10.3390/buildings1010000 V ersion April 22, 2026 submitted to Buildings 32 of 32

    work page doi:10.16910/jemr.17.2.4 2024

  49. [51]

    Greenery on residential buildings: Does it affect preferences and perceptions of beauty? Journal of Environmental Psychology 2011, 31, 89–98

    White, E.V .; Gatersleben, B. Greenery on residential buildings: Does it affect preferences and perceptions of beauty? Journal of Environmental Psychology 2011, 31, 89–98. https://doi.org/10 .1016/j.jenvp.2010.11.002

    2011

  50. [52]

    Windows, view, and office characteristics predict physical and psychological discomfort

    Ariës, M.B.C.; V eitch, J.A.; Newsham, G.R. Windows, view, and office characteristics predict physical and psychological discomfort. Journal of Environmental Psychology 2010, 30, 533–541. https://doi.org/10.1016/j.jenvp.2009.12.004

    work page doi:10.1016/j.jenvp.2009.12.004 2010

  51. [53]

    Simulated window views from different floors: How roof types of surrounding buildings associate with subjective restoration

    Wang, F.; Munakata, J. Simulated window views from different floors: How roof types of surrounding buildings associate with subjective restoration. Urban Forestry & Urban Greening 2023, 80, 128096. https://doi.org/10.1016/j.ufug.2023.128096

    work page doi:10.1016/j.ufug.2023.128096 2023

  52. [54]

    Machine Learning to speed up Computational Fluid Dynamics engineering simulations for built environments: A review,

    Yin, J.; Zhu, S.; MacNaughton, P .; Allen, J.G.; Spengler, J.D. Physiological and cognitive performance of exposure to biophilic indoor environments. Building and Environment 2023, 228, 109810. https://doi.org/10.1016/j.buildenv .2018.01.006

    work page doi:10.1016/j.buildenv 2023

  53. [55]

    Mapping façade materi- als utilizing zero-shot segmentation for applications in urban microclimate research

    Tarkhan, N.; Klimenka, M.; Fang, K.; Duarte, F.; Ratti, C.; Reinhart, C. Mapping façade materi- als utilizing zero-shot segmentation for applications in urban microclimate research. Scientific Reports 2025, 15, 5492. https://doi.org/10.1038/s41598-025-86307-1

    work page doi:10.1038/s41598-025-86307-1 2025

  54. [56]

    Pyvision: Agentic vision with dynamic tooling.CoRR, abs/2507.07998, 2025

    Lüddecke, T.; Ecker, A.S. Image segmentation using text and image prompts. In Proceedings of the Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, 2022, pp. 7086–7096. https://doi.org/10.48550/arXiv .2112.10003

    work page internal anchor Pith review doi:10.48550/arxiv 2022

  55. [57]

    Measuring emotion: the self-assessment manikin and the semantic differential

    Bradley , M.M.; Lang, P .J. Measuring emotion: The Self-Assessment Manikin and the Semantic Differential. Journal of Behavior Therapy and Experimental Psychiatry 1994, 25, 49–59. https: //doi.org/10.1016/0005-7916(94)90063-9

    work page doi:10.1016/0005-7916(94)90063-9 1994

  56. [58]

    The cognitive-emotional design and study of architectural space: A scoping review of neuroarchitecture and its precursor approaches

    Higuera-Trujillo, J.L.; Llinares, C.; Macagno, E. The cognitive-emotional design and study of architectural space: A scoping review of neuroarchitecture and its precursor approaches. Sensors 2021, 21, 2193. https://doi.org/10.3390/s21062193

    work page doi:10.3390/s21062193 2021

  57. [59]

    Measuring arousal and valence generated by the dynamic experience of architectural forms in virtual environments

    Chiamulera, C.; Padovani, M.; Beltramello, A.; Sartori, G.; Broggio, E.; Perini, L. Measuring arousal and valence generated by the dynamic experience of architectural forms in virtual environments. Scientific Reports 2022, 12, 13303. https://doi.org/10.1038/s41598-022-17689-9

    work page doi:10.1038/s41598-022-17689-9 2022

  58. [60]

    Walking through architectural spaces: The impact of interior forms on human brain dynamics

    Banaei, M.; Hatami, J.; Yazdanfar, A.; Gramann, K. Walking through architectural spaces: The impact of interior forms on human brain dynamics. Frontiers in Human Neuroscience 2017, 11, 477. https://doi.org/10.3389/fnhum.2017.00477

    work page doi:10.3389/fnhum.2017.00477 2017

  59. [61]

    Statistical power and optimal design in experiments in which samples of participants respond to samples of stimuli

    Westfall, J.; Kenny , D.A.; Judd, C.M. Statistical power and optimal design in experiments in which samples of participants respond to samples of stimuli. Journal of Experimental Psychology: General 2014, 143, 2020–2045. https://doi.org/10.1037/xge0000014

    work page doi:10.1037/xge0000014 2014

  60. [62]

    Power analysis and effect size in mixed-effects models: A tutorial

    Brysbaert, M.; Stevens, M. Power analysis and effect size in mixed-effects models: A tutorial. Journal of Cognition 2018, 1, 1–20. https://doi.org/10.5334/joc.10

    work page doi:10.5334/joc.10 2018

  61. [63]

    Masked autoencoders are scalable vision learners

    Liu, Z.; Hu, H.; Lin, Y .; Yao, Z.; Xie, Z.; Wei, Y .; Ning, J.; Cao, Y .; Zhang, Z.; Dong, L.; et al. Swin Transformer V2: Scaling up capacity and resolution. In Proceedings of the Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, 2022, pp. 11999–12009. https://doi.org/10.1109/CVPR52688.2022.01170

    work page doi:10.1109/cvpr52688.2022.01170 2022

  62. [64]

    Seeing the city: Using eye-tracking technology to explore cognitive responses to the built environment

    Hollander, J.B.; Purdy , A.; Wiley , A.; Foster, V .; Jacob, R.J.K.; Taylor, H.A.; Brunié, T.T. Seeing the city: Using eye-tracking technology to explore cognitive responses to the built environment. Journal of Urbanism: International Research on Placemaking and Urban Sustainability 2019, 12, 156–

    2019

  63. [65]

    https://doi.org/10.1080/17549175.2018.1531908

    work page doi:10.1080/17549175.2018.1531908 2018

  64. [66]

    Context based emotion recognition using EMOTIC dataset

    Kosti, R.; Álvarez, J.M.; Recasens, A.; Lapedriza, À. Context based emotion recognition using EMOTIC dataset. IEEE T ransactions on Pattern Analysis and Machine Intelligence 2020, 42, 2755–

    2020

  65. [67]

    Alvarez, Adria Recasens, and Agata Lapedriza

    https://doi.org/10.1109/TPAMI.2019.2916866. https://doi.org/10.3390/buildings1010000

    work page doi:10.1109/tpami.2019.2916866 2019