Investigating the relationship between colour and visual perception in campus circulation spaces: a data-informed exploratory framework

Abstract

This study investigates self-reported associations between colour and visual perception in a high-circulation public space on a university campus. Focusing on attributes including attention, visibility, perceived safety, aesthetic contribution, mood, and perceived wayfinding effectiveness, it explores how different colour alternatives are perceived across a set of everyday spatial-experience criteria, with the aim of informing evidence-based colour-design discussion rather than establishing causal or behavioural claims. The empirical study involved 151 undergraduate students from the Department of Interior Architecture at Istanbul Aydın University. The investigated setting was the heavily used outdoor circulation area known as the Blue Stairs. The theoretical framework draws on colour-in-context theory to examine associations between warm, cool, and neutral colour alternatives and users’ self-reported perceptual responses. Data were collected through an online, image-based questionnaire in which participants viewed eight digitally simulated colour alternatives of the staircase and, for each of eleven perceptual attributes, selected the single colour alternative they associated most strongly with that attribute—a categorical, forced single-choice design rather than an independent rating of each colour. Responses were analysed using descriptive statistics, chi-square goodness-of-fit tests, Cohen’s w effect-size estimation, Correspondence Analysis, and Hierarchical Cluster Analysis to explore patterns of association between colour and perceptual attribute. The findings indicate that no single colour was selected consistently across all criteria; instead, different colours were associated with different perceptual attributes. Correspondence analysis positioned red and purple, as an exploratory aggregate-level pattern, closer to attention and visibility, and blue, grey, and green closer to perceived safety, aesthetic contribution, and harmony with surroundings. A complementary hierarchical cluster analysis, with cluster number compared via silhouette coefficients across k = 2–7, found that the two-cluster solution—separating red from the remaining seven colours—achieved the highest silhouette coefficient (0.313), although overall cluster separation was modest. Taken together, these exploratory analyses suggest that the observed colour-selection patterns are structured rather than randomly distributed across perceptual attributes, and that colour may be a useful evidence-informed design variable, offering architects and designers a data-informed starting point for colour strategies in campus circulation spaces and comparable public environments.

1 Introduction

Spatial perception is a multisensory cognitive process combining visual, auditory, tactile, olfactory, and temporal input with a space’s function, aesthetic qualities, and symbolic meaning into a multilayered mental experience (Blesser and Salter, 2007; Bacci and Melcher, 2011; Gifford, 2014; Neves, 2024; Pallasmaa, 2012; Spence, 2020). Within this framework, visual perception is the most dominant mode in architectural and interior design, with colour, light, texture, contrast, and spatial arrangement as its primary directive components (Arnheim, 2004; Wagemans et al., 2012).

Although often dismissed as a superficial ‘aesthetic preference’, colour is a potentially influential design parameter associated in previous research with attention, mood, wayfinding-related performance, behavioural responses, and measurable neural activity in the visual cortex, limbic system, and prefrontal cortex (Coburn et al., 2017; Elliot and Maier, 2014; Elliot, 2015; Higuera-Trujillo et al., 2021; Zeki and Marini, 1998), operating through its own qualitative properties—hue, saturation, value, contrast, warmth–coolness, and wavelength (Camgöz et al., 2002; Gao and Xin, 2006; Ou et al., 2004).

Colour’s qualitative dimensions have also been associated with differences in the morphological and geometric perception of space: light surfaces, through higher light-reflectance, make a space appear wider, taller, and airier, whereas darker tones render boundaries more distinct and gradual hue transitions reinforce depth and orientation (Albers, 2013; von Castell et al., 2014; Oberfeld and Hecht, 2011).

Colour also shapes acoustic and thermal expectations through the colour–material relationship, without altering auditory perception itself (Aletta et al., 2018; Blesser and Salter, 2007; Spence, 2020). Bright, smooth, cool tones make surfaces appear harder and more distant, while warm tones on matte surfaces evoke a softer, ‘quiet-enclosed’ atmosphere that absorbs light and sound (Bacci and Melcher, 2011; Spence, 2020); colour context has also been proposed as a cross-modal influence on auditory perception, though a controlled replication testing whether identical sounds are judged differently against different coloured backgrounds found no consistent colour effect (Parizet and Koehl, 2011). Colours may similarly influence perceived thermal qualities without altering actual temperature, with cool tones evoking coolness and distance and warm tones evoking warmth and closeness (Tsushima et al., 2020). Reviews of indoor lighting design likewise emphasise that the perceptual and wellbeing outcomes of a chromatic environment depend on the lighting conditions under which it is experienced (Vélez González et al., 2026).

Previous neurophysiological studies have reported associations between cool-colour exposure and alpha-band activity, and between warm-colour exposure and heightened arousal-related responses (Gage, 1999; Elliot and Maier, 2014; Elliot, 2015; Higuera-Trujillo et al., 2021; Zhao et al., 2025); however, such effects vary by stimulus parameters, context, and experimental design, and they were not measured in the present study. Reported cognitive load is lower in light-coloured, low-contrast settings, and moderate saturation has been associated with enhanced task performance, especially among young adults (Park and Guerin, 2002; Mehta and Zhu, 2009).

Previous studies suggest that colour and contrast can contribute to visual salience, environmental differentiation, and wayfinding-related performance under specific experimental and applied conditions: highly saturated colours and high-contrast markings have been reported to ease wayfinding (Hidayetoglu et al., 2012), while warm tones have been associated with more active exploration and cool tones with longer sitting, resting, and waiting (Arthur and Passini, 1992; Stamps, 2011).

The “color-in-context” view holds that colour’s emotional, cognitive, and behavioural effects are not fixed universals but cues shaped jointly by biological predisposition (e.g., threat and warning indicators) and by learned, cultural associations tied to social context, climate, geography, and the goal at hand—social interaction, safety, performance, focus, or wayfinding (Elliot and Maier, 2014; Elliot, 2015). A complementary ecological account, Palmer and Schloss’s (2010) Ecological Valence Theory (EVT), argues that colour preferences track the aggregate emotional valence of the objects and experiences a colour evokes: for one person blue may recall ‘sky/sea/calm’, for another ‘coldness/distance’, so the same colour need not produce the same response across individuals (Palmer and Schloss, 2010; Schloss and Palmer, 2017). Damasio’s (2006) somatic-marker hypothesis offers a mechanism for this individual variation: feelings accumulated from past experience become bodily-encoded and bias present-day reactions, so a colour paired with negative markers functions as an alarm cue while one paired with positive markers acquires an encouraging quality. Together, these accounts frame colour preference as shaped jointly by individual history, cultural learning, and context (Palmer and Schloss, 2010; Schloss and Palmer, 2017) — meaning colour applications in stimulating public spaces such as campuses should be approached through perceptual evaluations tied to contextual goals, rather than aesthetic preference alone (Arthur and Passini, 1992; Lynch, 1960; Passini, 1996).

Explaining colour effects through hue alone is also insufficient: brightness/lightness and saturation/chroma independently shape affect and arousal, captured by the Pleasure–Arousal–Dominance (PAD) dimensions (Suk and Irtel, 2010; Valdez and Mehrabian, 1994; Wilms and Oberfeld, 2018). Colour decisions should therefore be evaluated not merely as ‘which colour’ but as ‘at which value, saturation, and lighting condition’ (Knez, 2001; Küller et al., 2006), corroborated by interior and office studies linking colour and light to mood, performance, and satisfaction (Kwallek et al., 1988, 1996; Stone, 2003; Elliot and Maier, 2007; Mehta and Zhu, 2009). These effects reflect universal tendencies as well as cultural differences: colour–emotion patterns hold only loosely across cultures, varying with linguistic and geographic proximity (Hurlbert and Ling, 2007; Jonauskaite et al., 2020; Ou et al., 2004); ‘blue is generally liked’ should therefore be read as a starting point weighed against user profile, context, and purpose, not a design prescription (Palmer and Schloss, 2010).

Within this scope, wayfinding, safety, and legibility are core components of user experience in high-circulation public spaces: legible environments strengthen cognitive maps and facilitate safe, controlled movement (Lynch, 1960), and environmental cues—layout, signage, colour/contrast—must be designed together, since wayfinding difficulties are linked to stress, dissatisfaction, and performance loss (Arthur and Passini, 1992; Passini, 1996). Colour coding, particularly in complex buildings and vertical circulation spaces, can support spatial experience as a differentiating cue (Jansen-Osmann and Wiedenbauer, 2004; Carpman and Grant, 2002); however, much of this literature comes from laboratory or decontextualised stimuli, with genuine campus interventions measuring multidimensional user evaluations remaining limited (Elliot, 2015; Gifford, 2014), and colour perception’s interaction with ambient light and background necessitates more systematic field measurement (Valdez and Mehrabian, 1994; Wilms and Oberfeld, 2018).

Colour should not be treated as an isolated environmental variable: perception emerges through the interaction of colour with spatial configuration, lighting, materiality, graphics, and other sensory cues within a scene. Comparative work on physical and virtual exhibition spaces shows that spatial perception depends on the joint integration of visual, auditory, tactile, olfactory, and thermal information rather than on any single sensory channel (Xu et al., 2025). EEG- and VR-based wayfinding research similarly indicates that colour operates alongside graphics, signage, and other architectural features in shaping orientation-related outcomes, rather than acting as an independent predictor (Kalantari et al., 2022). At a more basic level, experimental manipulation of privacy, biophilia, clutter, and colour temperature in built spaces has been linked to the fulfilment of psychological needs such as competence, autonomy, and relatedness, underscoring that colour is one of several interacting design features rather than a standalone determinant of user experience (Jutzi et al., 2025). Taken together, this literature indicates that the present study investigates subjective, image-based perceptual associations with colour within complete architectural scenes, rather than isolated behavioural or physiological responses to colour alone.

This positions the study to inform colour-related design decisions across interior typologies, from educational spaces to workspaces. Although existing literature shows how different colour tones alter spatial perception (Fekry et al., 2025), a holistic, measurement-based evaluation of university students’ perceptions and preferences regarding spatial colours—including, within the Turkish context, survey-based evidence on young adults’ visual perception—remains limited, underscoring the value of user-based data for grounding architectural design in scientific evidence.

Although previous studies have examined colour preference, emotion, visual salience, and wayfinding separately, fewer studies have evaluated multiple self-reported perceptual attributes within the same image-based architectural setting. Evidence is particularly limited for outdoor campus circulation spaces and for studies that combine respondent-level colour-selection frequencies with exploratory aggregate-level analyses. The present study addresses this gap by examining eleven perceptual attributes across eight digitally simulated colour alternatives in a real campus staircase context, while explicitly distinguishing self-reported perception from objectively measured behaviour.

2 Methods

2.1 Sample and study area

This study was conducted at and around the “Blue Stairs” — an environmental architectural design element located on the Florya Campus of Istanbul Aydın University (Figures 1, 2). The research sample consists of students from the Department of Interior Architecture, Faculty of Architecture and Design, who actively used the relevant campus during the 2025–2026 academic year. Sampling was based on accessibility and voluntary participation; a total of 151 students from the 1st, 2nd, 3rd, and 4th years took part in the survey administered within their courses.

FIGURE 1

FIGURE 2

Interior Architecture students were selected through convenience sampling because they were regular users of the investigated campus environment and were familiar with the Blue Stairs as part of their everyday circulation context. Their design education may have increased their sensitivity to colour, composition, and spatial representation; therefore, the sample should be regarded as a context-specific, design-literate university population rather than as representative of the general public. No formal a priori power analysis was conducted because the study was designed as an exploratory, image-based questionnaire study. The sample size nevertheless provided sufficient observations for descriptive frequency analysis and the reported exploratory aggregate-level procedures, but not for broad population-level inference.

The Blue Stairs were selected because they constitute a visually prominent, high-circulation outdoor transition space used for movement, sitting, meeting, and informal social activity. The site also functions as a familiar campus landmark and reference point. These characteristics made it suitable for examining colour-related perceptual associations in a setting where visibility, perceived safety, wayfinding, aesthetic contribution, and perceived sense of place are simultaneously relevant. At the same time, the existing blue identity of the site represents a familiarity-related confound that is explicitly acknowledged in the Limitations section.

2.2 Study method

This study adopts a quantitative research design supported by a structured literature review, combining theoretical foundations with quantitative questionnaire data derived from an image-based evaluation of a real campus setting. The literature synthesis underlying the methodological classification is summarized in Table 1, while the corresponding methodological categories are presented in Table 2. The methodological framework draws on a body of literature comprising experimental studies, neurophysiological research, field studies, and survey applications (Table 2). To ensure visual standardisation and comparability, eight different colour parameters (Blue, Red, Yellow, Blue–Orange, Green, Orange, Purple, Grey) were digitally applied to the image of the space to produce the visual stimuli (Table 3).

TABLE 3

Colour alternatives applied to the Blue Stairs (developed by the authors).

2.3 Stimulus standardisation and materials

These methodological details are reported to improve transparency, reproducibility, and reporting quality of the stimulus-presentation protocol. The absence of hardware-based monitor calibration and of randomised/counterbalanced presentation order is acknowledged as a limitation in Section 8.

The eight visual alternatives were produced from the same base photograph by modifying only the stair-surface colour treatment. Geometry, camera position, perspective, background elements, image dimensions, and nominal editing parameters were held constant. The alternatives were visually reviewed by the research team to confirm that the intended colour condition was clearly distinguishable and that no unintended spatial element had been altered. This review represented a procedural quality check rather than a formal perceptual pretest. No independent pilot study, colourimeter measurement, or hardware-based calibration was conducted.

The Blue–Orange condition was the only two-colour alternative and therefore differed from the single-colour conditions not only in hue, but also in chromatic contrast, spatial zoning, visual complexity, and salience. Its selections for perceived wayfinding effectiveness and perceived sense of place cannot therefore be attributed exclusively to the specific Blue–Orange pairing. Future experiments should compare multiple two-colour combinations and equivalent single-colour controls.

2.4 Data collection and analysis procedure

Following ethics committee approval, the data used in the final analyses were collected through the revised structured online questionnaire. Participants viewed the current design and eight digitally simulated colour alternatives of the staircase. For each of eleven perceptual attributes—attention-grabbing quality, visibility, perceived safety, aesthetic contribution, harmony with surroundings, positive effect on mood, perceived wayfinding effectiveness, invitingness, perceived sense of place, suitability for activities, and perceived spaciousness—participants selected the single colour alternative (1 of 8) they associated most strongly with that attribute. This was a categorical, forced single-choice design: for each attribute, a participant selected exactly one colour rather than rating every colour independently. The eight colour codes used in coding and analysis were: 1 = Blue, 2 = Red, 3 = Yellow, 4 = Blue–Orange combination, 5 = Green, 6 = Orange, 7 = Purple, 8 = Grey.

In addition to descriptive statistics and chi-square goodness-of-fit testing, the analytical strategy incorporated Cohen’s w effect-size estimation and two complementary exploratory multivariate analyses: (i) Correspondence Analysis (CA) of the colour × perceptual-attribute matrix, providing an aggregate-level view of colour–attribute associations; and (ii) Hierarchical Cluster Analysis (HCA), using Ward’s method and Euclidean distance, to examine similarities among colour profiles across the eleven perceptual attributes. Cluster solutions were compared using silhouette coefficients for k = 2–7. Valid sample size varies slightly by attribute because invalid/missing responses are excluded per attribute rather than per respondent. All analyses were conducted in Python (NumPy/SciPy) on the aggregated colour × attribute frequency table.

For each attribute, the chi-square goodness-of-fit test used an equal-probability null distribution across the eight alternatives. This null model was selected as a neutral baseline because the questionnaire did not specify an a priori expected preference distribution and no external normative probabilities were available for the eight digitally simulated conditions. Rejection of this null hypothesis indicates only that the alternatives were not selected equally often; it does not establish that colour caused psychological or behavioural outcomes.

Analyses were performed in Python using NumPy, SciPy, and scikit-learn. Hierarchical Cluster Analysis was performed on row-proportion colour profiles using Ward linkage and Euclidean distance, while the same profiles were multiplied by 100 only for visual presentation in the heatmap. Row-percentage standardisation was retained because the analysis was exploratory and based on only eight colour profiles, and because unstandardised profiles are dominated by differences in marginal selection totals rather than by profile shape. Alternative standardisation procedures were therefore not used as the basis for the reported solution, and systematic sensitivity testing across alternative standardisations is identified as a robustness check for future work; the cluster structure is accordingly interpreted as exploratory and descriptive rather than as a stable partition. The reproducibility materials are provided in Supplementary Material S1, S2. Supplementary Material S1 documents the questionnaire, visual stimuli, coding rules, and aggregated input matrix, whereas Supplementary Material S2 provides the statistical code, computational procedures, software information, and validation outputs.

Internal-consistency indicators such as Cronbach’s alpha were not calculated because the eleven items were not designed as interchangeable indicators of a single latent construct. Each item represented a distinct perceptual attribute and required a categorical forced choice among eight colour alternatives. Accordingly, the instrument was analysed as a set of attribute-specific categorical selections rather than as a unidimensional psychometric scale. Content validity was supported through alignment of the attributes with the colour-perception, environmental cognition, and wayfinding literature; however, formal construct-validation procedures were not conducted and are acknowledged as a limitation.

The full questionnaire, including the original Turkish item wording, English translation, response format, eight colour codes, operational definitions, and coding rules, is provided in Supplementary Material S1.

2.5 Ethics statement

The study protocol and the revised questionnaire used to generate the dataset reported in this manuscript were reviewed and approved by the Istanbul Aydın University Social and Human Sciences Ethics Committee (Meeting No. 2026/07, 30 July 2026). Following ethics approval, the questionnaire was re-administered to the participants, who were informed about the purpose and procedures of the study, the voluntary nature of participation, the intended academic use of the data, and their right to withdraw without consequence. Electronic informed consent was obtained before participation. Only responses obtained through this post-approval administration were used in the final analyses and findings reported in this manuscript. The post-approval responses showed no substantive changes in the overall response patterns compared with those observed during the preliminary study phase; however, the preliminary responses were not included in the final analytical dataset. Direct participant identifiers were removed before analysis and were not included in the analytical dataset.

3 Results

3.1 Sociodemographic and space-use findings

The majority of survey respondents were Turkish-national female students. In terms of academic level, respondents were predominantly first- and second-year students, and roughly half of this group had resided in Istanbul for between six and twenty years.

Regarding intensity of use of the space, those who never used the area and those who used it only occasionally formed the majority; ‘ascending/descending’ circulation and sitting were the most common purposes of use, and most participants reported that they did not prefer to spend time on the stairs itself. More than half of the participants indicated that colour was a determining factor in their use of the space, and that the stairway structure functioned as an effective reference point in the wayfinding (orientation) process (Table 4).

3.2 Colour selection findings

Looking at respondents’ colour selections (Table 5), red was the colour most frequently selected for attention-grabbing quality (78 of 146 valid responses, 53.4%) and visibility (78 of 147, 53.1%), consistent with a functional, alerting profile; purple was the next most frequently selected colour on both attributes, well behind red. For perceived safety, blue was the most frequently selected colour (75 of 146, 51.4%), followed by green (26, 17.8%) and grey (17, 11.6%). The measured safety-related construct in this study is perceived safety, not trust; no attribute in the questionnaire measured trust. For aesthetic contribution, blue was again the most frequently selected colour (72 of 143, 50.3%), followed by the blue–orange combination (27, 18.9%) and grey (19, 13.3%); green was selected by 3 respondents (2.1%). Green was therefore not the second most frequently selected colour for aesthetic contribution, contrary to an earlier draft. For harmony with surroundings, blue (56 of 147, 38.1%) and grey (54, 36.7%) were selected at broadly comparable rates, with orange (12, 8.2%) and green (10, 6.8%) selected considerably less often. Blue was also the most frequently selected colour for positive effect on mood (55 of 144, 38.2%), followed by green (34, 23.6%); for perceived wayfinding effectiveness (53 of 140, 37.9%), followed by the blue–orange combination (28, 20.0%) and red (26, 18.6%); for invitingness (54 of 145, 37.2%), followed by the blue–orange combination (25, 17.2%) and green (22, 15.2%); and for perceived sense of place (48 of 145, 33.1%), followed by the blue–orange combination (28, 19.3%) and red (21, 14.5%). For suitability for activities, grey was the most frequently selected colour (40 of 142, 28.2%), followed by blue (32, 22.5%) and the blue–orange combination (30, 21.1%). For perceived spaciousness, yellow was the most frequently selected colour (35 of 146, 24.0%), followed by grey (32, 21.9%) and blue (30, 20.5%). Across the eleven attributes, Blue was the most frequently selected colour for seven attributes, Red for two attributes, Grey for one attribute, and Yellow for one attribute. The Blue–Orange combination and Grey also appeared among the more frequently selected alternatives for several attributes. These findings represent descriptive selection patterns and do not imply causal effects of colour on perceptual outcomes.

3.3 Effect sizes and the structural pattern underlying colour–attribute associations

The descriptive results in Section 3.2 establish which colour was most frequently selected on each perceptual attribute, but they do not by themselves indicate the magnitude of departure from an equal-probability distribution or summarise the overall structure of the colour × attribute matrix. To address this, two complementary exploratory analyses were carried out on the aggregated colour × attribute frequency table (Table 4), alongside chi-square goodness-of-fit testing with Cohen’s w as effect size: (i) Correspondence Analysis (CA) of the full 8 (colour) × 11 (attribute) contingency table; and (ii) hierarchical cluster analysis (Ward’s method) of the eight colours’ profiles across the eleven attributes, with cluster count compared via silhouette coefficients.

3.3.1 Effect sizes (Cohen’s w)

Cohen’s w was calculated as w = √(χ²/n), where χ² is the chi-square statistic for the one-way goodness-of-fit test and n is the total number of respondents for that attribute. For every attribute, participants’ colour selections departed significantly from a uniform (chance) distribution across the eight alternatives (all p < 0.001). Effect sizes, computed as Cohen’s w, were uniformly large by conventional benchmarks (small ≈0.10, medium ≈0.30, large ≈0.50; Cohen, 1988). The chi-square and Cohen's w results are presented in Table 6. ‘Attention-grabbing’ (w = 1.277), ‘Visibility’ (w = 1.247), ‘Perceived safety’ (w = 1.247), ‘Aesthetic contribution’ (w = 1.228), and ‘Harmony with surroundings’ (w = 1.168) showed the largest effect sizes. ‘Perceived spaciousness’ (w = 0.628) and ‘Suitability for activities’ (w = 0.727) showed the comparatively smallest—though still large—effect sizes.

3.3.2 Correspondence analysis: an exploratory aggregate-level view

Correspondence Analysis was applied to the 8 × 11 colour × attribute matrix as an exploratory aggregate-level analysis of association patterns; it does not test or confirm respondent-level relationships. The first two dimensions jointly captured 76.5% of the total inertia (Dimension 1 = 58.4%, Dimension 2 = 18.0%) (Figure 3).

FIGURE 3

On Dimension 1, Red appears as the most extreme point on one side, positioned close to ‘Attention-grabbing’ and ‘Visibility’, broadly consistent with the descriptive pattern in Section 3.2. Purple appears on the same side of the axis, though less extremely, and is also positioned closer to attention/visibility than to aesthetic-related attributes. On the opposite side of Dimension 1, Blue, Grey, and Green appear closer to ‘Perceived safety’, ‘Aesthetic contribution’, and ‘Harmony with surroundings’. These three colours occupy the same general region of the map here, without this implying that they form a single perceptual category—the finer differences between them are set out in Section 3.2 and Table 4. Dimension 2 separates Yellow, positioned closer to ‘Perceived spaciousness’, from the remaining colours, suggesting spaciousness-related selections follow a somewhat different pattern than the safety/aesthetics/attention associations captured on Dimension 1. This is an exploratory aggregate-level colour–attribute association pattern, not evidence of a latent individual-level psychological structure.

3.3.3 Hierarchical cluster analysis of colour profiles

As a complementary exploratory analysis of the same aggregate matrix, the eight colours’ profile vectors across the eleven perceptual attributes (row-proportion profiles) were subjected to agglomerative hierarchical clustering using Ward’s method and Euclidean distance. The silhouette coefficient was computed for each solution between k = 2 and k = 7.

Among the tested solutions, k = 2 yielded the highest silhouette coefficient (0.313), although the overall cluster separation was modest (k = 3: 0.141; k = 4: 0.117; k = 5: 0.102; k = 6: 0.057; k = 7: 0.014) (Figure 4). The highest-scoring two-cluster solution separated Red from the remaining seven colours, but the modest silhouette coefficient indicates limited overall cluster separation rather than a sharply defined structural split. The dendrogram’s finer internal structure remains descriptively informative. Blue–Orange and Orange merge first, at the smallest distance of any pair (d = 0.1806), followed by Blue joining this pair (d = 0.1953). Yellow and Purple form a separate early pair (d = 0.2557), and Green joins the cluster comprising Blue, Blue–Orange, and Orange (d = 0.2905). Grey then joins the resulting structure (d = 0.3484), and the two larger substructures merge (d = 0.3854) before finally merging with Red, the last colour to join the tree and the largest single step in the dendrogram (d = 0.5455). This clustering is exploratory and descriptive rather than a validated taxonomy of colour categories: it summarises similarities among the eight colour profiles in the present aggregated matrix and should not be treated as a confirmed or generalisable classification.

FIGURE 4

Together, the effect-size, correspondence-analysis, and cluster-analysis results reflect two complementary exploratory analyses of the same aggregate matrix that produced broadly consistent patterns: colour profiles in the aggregate matrix show differentiated rather than uniform patterns across the eleven perceptual attributes, with Red displaying the most distinct overall profile, consistently standing apart from the other colours across both analyses. Within the remaining group, the correspondence analysis suggests further, exploratory differentiation (blue and grey toward safety/aesthetics/harmony, yellow toward spaciousness), but this finer structure is descriptive and exploratory rather than independently confirmed by the cluster analysis. The corresponding row-percentage heatmap of the colour × attribute profiles is shown in Figure 5.

FIGURE 5

4 Discussion

According to the research results, the associations between colour selections and self-reported spatial-perception attributes differed across aesthetic, functional, and psychological criteria. The selection patterns suggest that colour alternatives associated with visibility and signalling were not necessarily the same alternatives associated with aesthetic or atmospheric evaluations. This pattern is consistent with color-in-context theory (Elliot, 2015; Elliot and Maier, 2014). The structural analyses reported in Section 3.3 provide an exploratory aggregate-level representation consistent with this context-dependent interpretation: the aggregate colour–attribute matrix can be represented through a smaller number of exploratory dimensions, principally an attention/visibility contrast and a secondary spaciousness-related dimension, which is broadly consistent with the kind of contextual patterning that color-in-context theory predicts.

The findings indicate that blue was the most frequently selected colour for perceived safety (51.4% of valid responses) and aesthetic contribution (50.3%), and it was also the most frequently selected colour for perceived sense of place (33.1%). This result is consistent with the colour-psychology literature, which associates blue with positive evaluation, perceived safety, and calm (Elliot and Maier, 2014; Elliot, 2015). Previous neurophysiological studies have reported associations between blue-related visual stimuli and alpha-band activity (Gage, 1999; Higuera-Trujillo et al., 2021); however, such mechanisms were not investigated in the present study. Rather than a superficial generalisation, this pattern may be interpreted through Ecological Valence Theory: blue’s pairing with positive, life-supporting natural elements such as sky and clean water (Palmer and Schloss, 2010; Schloss and Palmer, 2017) may relate to previously formed positive somatic markers (Damasio, 2006); however, neither these associations nor the proposed somatic mechanism was measured directly in the present study. Blue’s prominence (38.2%) on the ‘positive effect on mood’ criterion likewise may be interpreted in relation to the pleasure and dominance components of the PAD model (Valdez and Mehrabian, 1994; Lynch, 1960; Passini, 1996), with brightness, saturation, and light-reflectance contributing alongside hue (Wilms and Oberfeld, 2018). In sum, Blue’s relatively frequent selection across several circulation-related attributes may reflect perceived psychological comfort rather than an abstract hue preference. This interpretation is broadly consistent with the correspondence-analysis map, in which Blue and Green occupy the same general safety-related region. However, the hierarchical cluster analysis does not identify Blue and Green as the closest pair; Blue–Orange and Orange form the earliest cluster. Unlike decontextualised colour-preference studies, however, the present setting was already known and experienced as the Blue Stairs. The Blue-related results may therefore reflect both general colour associations and context-specific familiarity, repeated exposure, and place identity, and should not be interpreted as evidence that Blue is universally perceived as safer or more aesthetic across populations and building types.

Red, displaying a dominant functional character, was identified as the most attention-grabbing and visible colour. Red’s frequent selection may be interpreted in relation to its established warning, priority, and salience associations in transition-space contexts. Consistent with color-in-context theory (Elliot, 2015), red’s learned and biological associations with warning and priority are associated with a perceived ‘visibility advantage’, potentially contributing to perceived visual salience and perceived wayfinding within vertical circulation settings (Arthur and Passini, 1992). However, a contrast was observed between Red’s frequent selection for attention-grabbing quality and visibility and its less frequent selection for aesthetic contribution and perceived safety (Wilms and Oberfeld, 2018). Previous neurophysiological studies have reported associations between red-related visual stimuli and elevated beta-band activity and attention (Elliot, 2015; Elliot and Maier, 2014); however, such mechanisms were not measured here. This aligns with red’s negative somatic-marker associations (Damasio, 2006) — stop, danger, alarm—which similarly explain its comparatively lower aesthetic and perceived-safety ratings. In sum, the selection pattern indicates that Red was strongly associated with attention-grabbing quality and visibility in this sample, while also showing that this salience effect does not always coincide with high emotional acceptance; colour choice in circulation areas and campus furnishings should therefore be aligned with the intended perceptual outcome—active signalling/alerting versus passive comfort/rest (Stamps, 2011). Structurally, this dual character is exactly what places Red at the extreme, isolated end of both the correspondence-analysis map and the cluster dendrogram in Section 3.3: it is the single colour whose profile is least compatible with the rest of the colour set. It should be emphasised that the questionnaire did not measure response time, detection threshold, or navigational performance; the result therefore indicates a subjective visibility association rather than demonstrated behavioural superiority.

Yellow was the most frequently selected colour for perceived spaciousness. This result is consistent with the literature on morphological and geometric perception, which holds that light-coloured surfaces, through higher light-reflectance levels, make a space appear wider, taller, airier, and more comfortable (Albers, 2013; von Castell et al., 2014; Oberfeld and Hecht, 2011). This aligns with the PAD framework (Valdez and Mehrabian, 1994; Knez, 2001; Küller et al., 2006): yellow’s comparatively high lightness and brightness may contribute to pleasure-related evaluations and reduced perceptions of boundary constriction, so the space is perceived as more spacious—a pattern also supported by experimental findings on the divergent effects of colour parameters on perceived spatial width (Wilms and Oberfeld, 2018). In sum, yellow’s prominence in spaciousness perception along the campus circulation axis may be interpreted in relation to its comparatively high brightness value on users’ perceived psychological responses, particularly relevant for optical/volumetric comfort in dense public and vertical transition spaces. This is precisely the secondary, largely independent axis (Dimension 2) that Yellow occupies alone in the correspondence-analysis solution in Section 3.3. The selection margin for Yellow over Grey and Blue on this attribute was nevertheless relatively modest, suggesting that spaciousness judgements may depend not only on hue but also on lightness, contrast, image composition, and display conditions.

Grey was one of the most frequently selected options for harmony with surroundings (36.7% of valid responses) and was the most frequently selected colour for suitability for activities (28.2%). Participants’ frequent selection of Grey for harmony with surroundings is consistent with the interpretation that its low-chroma character may support visual continuity, through its monochromatic structure and low chroma values, without conflicting with the campus’s existing architectural fabric and landscape elements (von Castell et al., 2014; Oberfeld and Hecht, 2011). Grey’s selection as the most suitable backdrop for different activities may be related to its neutral character, which does not dictate a specific behavioural or emotional mode and instead permits spatial flexibility. Within the PAD framework, previous research has associated low-chroma neutral environments with lower arousal and reduced visual demand, although cognitive load was not measured in the present study (Valdez and Mehrabian, 1994; Mehta and Zhu, 2009) — consistent with evidence that monochromatic colour schemes can balance task performance, focus, and mood in interior and work environments (Kwallek et al., 1988, 1996; Stone, 2003). In sum, the findings suggest that Grey may represent a useful neutral background colour for multifunctional campus settings, potentially reducing perceived visual complexity and supporting flexible spatial evaluation. This interpretation should nonetheless be treated cautiously, because the exact material–colour interaction and the real ambient-light conditions of the site were not experimentally controlled.

Building on the single-colour findings above, the Blue–Orange alternative was relatively frequently selected for perceived wayfinding effectiveness and perceived sense of place: the prominence of Blue (37.9%) and the Blue–Orange combination (20.0%) on the perceived wayfinding effectiveness attribute may be interpreted in relation to the wayfinding literature, which associates spatial legibility with reduced user stress and facilitated environmental orientation (Arthur and Passini, 1992; Passini, 1996). This pattern may be interpreted in relation to the visual contrast and differentiation provided by complementary colours such as blue and orange, as discussed by Jansen-Osmann and Wiedenbauer (2004) and Carpman and Grant (2002), particularly in vertical transition areas and complex campus structures. The finding that this combination was also selected as an aesthetic component associated with perceived sense of place (19.3%) is broadly consistent with architectural approaches advocating hierarchical, multidimensional visual cues over a single absolute colour solution in design (Arthur and Passini, 1992). However, because it was the only two-colour condition, the observed selections cannot be attributed specifically to the Blue–Orange pairing or interpreted as evidence of improved cognitive mapping or actual navigation performance.

Green’s identification as the second most frequently selected colour—after blue—for perceived safety (17.8% of valid responses) is consistent with the environmental-psychology literature. Green was not among the more frequently selected colours for aesthetic contribution in the corrected data (2.1%; Section 3.2), so the discussion below is restricted to the perceived-safety finding. In light of neuroarchitectural studies, green tones, like blue, have been associated in previous EEG-based research with increased alpha-wave activity and with reported associations including tranquillity, calm, perceived safety, and reduced cognitive load and stress (Gage, 1999; Higuera-Trujillo et al., 2021; Zhao et al., 2025), consistent with the PAD-model finding that blue-green tones combine lower arousal with higher positive valence (Valdez and Mehrabian, 1994; Wilms and Oberfeld, 2018). One possible interpretation of this selection pattern draws on biophilic-design principles and Ecological Valence Theory (Palmer and Schloss, 2010): green’s evocation of nature, vegetation, and vitality pairs the colour with life-supporting associations of high positive valence, and the therapeutic qualities of open spaces and landscape elements trigger spatial belonging and psychological comfort (Aletta et al., 2018). In sum, green’s prominence on the perceived-safety dimension may be interpreted in relation to lower-arousal and nature-associated meanings reported in previous literature. This possible interpretation is broadly consistent with Green’s location in the safety-related region of the correspondence-analysis map, in the same broad region as Blue, although the cluster dendrogram does not identify them as the closest pair (Section 3.3); the underlying psychological or neurophysiological mechanism was not tested.

The contribution of the multivariate analyses lies not in validating a psychological sequence, but in summarising the aggregate colour-selection matrix in a more interpretable form. The correspondence analysis distinguished an attention–visibility region, a safety–aesthetic–harmony region, and a secondary spaciousness-related direction. The hierarchical clustering provided a complementary profile-based view, identifying Red as the most distinct colour profile while showing only modest overall separation. These analyses therefore extend conventional colour-preference reporting by demonstrating how the same colour alternatives occupy different aggregate perceptual roles across multiple attributes.

The sample composition may also have shaped the observed selection patterns. The predominance of female participants and the exclusive inclusion of Interior Architecture students may have influenced sensitivity to hue, contrast, aesthetic coherence, and spatial representation. Recent evidence indicates that colour–emotion evaluations may differ between art-related and non-art university populations and may be jointly influenced by hue, saturation, and brightness (Song et al., 2025), reinforcing the need to interpret the present design-student sample cautiously. In addition, 92.1% of the sample consisted of Turkish nationals, and colour associations may vary across linguistic, cultural, and geographic contexts. Because respondent-level subgroup analyses were not performed, the present study cannot determine whether gender, nationality, year of study, prior colour-course training, or familiarity with the Blue Stairs systematically altered selections. These variables should therefore be treated as plausible moderators for future investigation.

A further interpretive caution concerns what the observed colour-selection patterns can and cannot represent. Because participants evaluated complete architectural scenes containing colour, geometry, depth, material appearance, environmental familiarity, and other contextual visual cues, the reported associations should not be read as responses to colour in isolation (Xu et al., 2025). The present design-image evaluations are accordingly distinct from EEG studies that record neural correlates of colour exposure (Gage, 1999; Higuera-Trujillo et al., 2021; Zhao et al., 2025), from behavioural wayfinding experiments that combine colour with graphics, signage, and other architectural features and record navigation-related outcomes (Kalantari et al., 2022), and from multisensory environmental studies that manipulate several design features simultaneously and measure need fulfilment or approach behaviour (Jutzi et al., 2025). The present study investigates only subjective, image-based perceptual associations with colour, collected through a single self-report instrument, and the findings should be interpreted within that scope rather than as evidence of physiological, behavioural, or multisensory effects.

5 Toward a conceptual framework: colour-mediated environmental cognition

The convergent evidence from Sections 3.3 and 4 — effect sizes indicating that colour selections were not equally distributed on any attribute; a correspondence-analysis map organised around a safety/aesthetics-versus-attention/visibility axis crossed by a spaciousness axis; and a cluster solution in which Red separates from the remaining colours, although overall cluster separation is modest—informs a proposed conceptual framework for organising possible relationships among colour and spatial-perception variables in campus circulation spaces. Rather than treating each perceptual outcome (safety, mood, wayfinding, perceived sense of place) as a separate dependent variable, the findings are consistent with a proposed conceptual organisation running from the physical properties of colour, through perception and emotion, to behaviour and ultimately to spatial identity; this sequence is offered as an interpretive structure rather than a tested causal path, since the present design did not include path analysis or other formal mediation testing. The proposed conceptual organisation is illustrated in Figure 6.

FIGURE 6

In this model, Colour (hue, saturation, brightness) is the physical input. It is first processed as Perception, where salience and legibility determine whether a colour is noticed and functions as a landmark—a pattern consistent with Red’s frequent selection for attention and visibility. Perception may be associated with Emotion, operationalised here through the pleasure–arousal dimensions of the PAD model—a pattern consistent with Blue’s prominence in aesthetic contribution and with Blue and Green’s relatively frequent selection for perceived safety and positive effect on mood. The framework organises proposed relationships among colour properties, perception, emotional evaluation, possible behavioural tendencies, perceived wayfinding effectiveness, and perceived sense of place. Repeated perceptual and emotional evaluations may be relevant to the development of spatial identity over time; this proposed relationship was not tested in the present study. Grey and yellow occupy a partially distinct role within this chain, acting less on the emotion–behaviour pathway and more directly on the perception stage, by modulating perceived spatial volume and providing a neutral backdrop that does not itself dictate a behavioural mode.

This framework offers two advantages over a purely descriptive, colour-by-colour account. First, it is testable and extendable: each proposed relationship can be examined in future studies using appropriate respondent-level, behavioural, or physiological measures. Second, it provides a structure that can be tested in future studies beyond the specific case of the Blue Stairs: future circulation-space colour studies may use the framework to organise and test relationships among perceptual priorities, locating a given colour decision within the same conceptual organisation and evaluating it against whichever stage is the design priority (visibility and perceived wayfinding effectiveness versus comfort and perceived sense of place), rather than searching for a single universally ‘best’ colour.

Taken together, the convergence of descriptive statistics, effect-size estimation, correspondence analysis, and hierarchical clustering suggests that the aggregate colour-selection profiles varied across multiple perceptual attributes rather than following a single uniform pattern. Consequently, evidence-informed colour strategies may prioritise perceptual objectives, such as self-reported visibility, perceived safety, perceived sense of place, and perceived wayfinding effectiveness, instead of assuming the existence of a universally optimal colour.

6 Design implications

Although these observations may offer useful hypotheses for comparable transitional environments, the present findings derive from a single campus staircase and should not be generalised directly to other building types without further validation. Potential application to hospitals, offices, classrooms, or civic buildings should accordingly be treated as a hypothesis for future validation rather than as a direct design recommendation derived from this single-site study.

7 Conclusion

In this image-based study of a campus circulation space, colour emerged not merely as a decorative consideration but as a potentially relevant environmental design variable associated with differentiated self-reported perceptions of attention, visibility, perceived safety, aesthetic contribution, mood, perceived wayfinding effectiveness, and spatial experience. Moving beyond universal colour-psychology generalisations, the findings are broadly consistent with Elliot’s (2015) ‘color-in-context’ approach and Palmer and Schloss’s (2010) Ecological Valence Theory, while remaining subject to the contextual and methodological limitations of the image-based design. This consistency is not limited to the raw selection frequencies alone: the effect-size, correspondence-analysis, and cluster-analysis results reported in Section 3.3, synthesised into the exploratory conceptual framework in Section 5, provide an exploratory aggregate-level representation of the observed colour–attribute patterns—moving the contribution from ‘which colour was chosen’ to ‘how colour-selection patterns relate to multiple perceptual and experiential attributes in a public circulation space’.

The most fundamental dynamic revealed by the research outputs is that the observed colour-selection patterns did not follow a single homogeneous profile but differed according to the perceptual attribute under consideration. Indeed, the divergence observed in functional expectations such as visibility/signalling relative to atmospheric/aesthetic comfort expectations across colours suggests that colour-related perceptions should be interpreted in conjunction with contextual components such as environmental harmony, light, material, and purpose of use (Elliot and Maier, 2014).

Building on the empirical evidence and theoretical outputs obtained in this study, these findings can be synthesised into a compact, function-specific colour logic for dense, highly stimulating public transition spaces. In this sample, Red was most frequently associated with attention-grabbing quality and visibility; however, no actual wayfinding performance or physiological arousal was measured. Red was selected far less often for aesthetic contribution and perceived safety, a pattern that is consistent with an accent rather than a dominant role. Blue was the most frequently selected colour for perceived safety, aesthetic contribution, positive effect on mood, perceived wayfinding effectiveness, invitingness, and perceived sense of place. Green was the second most frequently selected colour for perceived safety and positive effect on mood, but it was selected infrequently for aesthetic contribution. Yellow was most frequently associated with perceived spaciousness, potentially reflecting the combined roles of hue, lightness, contrast, and image presentation. Grey was frequently selected as a neutral and potentially flexible backdrop that supports varied activities without imposing a behavioural mode, and was also the most frequently selected colour for harmony with surroundings (36.7%). Beyond single hues, the Blue–Orange alternative was relatively frequently selected for perceived wayfinding effectiveness (20.0%) and perceived sense of place (19.3%), although its two-colour composition prevents attributing this pattern solely to hue. Taken together, these patterns indicate that no single colour was universally optimal in this setting: each was associated with a different pattern of self-reported perceptual attributes within the exploratory framework outlined in Section 5, and design decisions should be matched to the specific perceptual goal at hand.

From a general perspective, these results indicate differentiated, context-dependent associations between colour selection and self-reported visual-perception attributes. These observed selection patterns may provide preliminary perceptual cues relevant to perceived wayfinding, spatial comfort, and future behavioural investigation.

It should be emphasised that the alpha- and beta-wave mechanisms invoked above are drawn from the prior neurophysiological literature to interpret the self-reported perceptual findings and the proposed behavioural relationships; they were not directly measured via EEG or other physiological instrumentation within the present study. As noted in Sections 8 and 9, empirical validation of these proposed mechanisms through EEG and related physiological measures remains a priority for future research.

The principal scientific contribution of this study is the integration of eleven attribute-specific colour-selection outcomes within a single descriptive and exploratory aggregate-level framework. The study shows that colour-selection profiles differ across attention, visibility, perceived safety, aesthetic contribution, mood, perceived wayfinding effectiveness, invitingness, perceived sense of place, suitability for activities, harmony with surroundings, and perceived spaciousness.

For campus stairs and comparable transitional spaces, the results suggest that colour selection should be aligned with the intended perceptual objective rather than with a single general preference. Red was most frequently associated with attention and visibility; Blue with perceived safety and aesthetic contribution; Grey with harmony with surroundings and suitability for activities; and Yellow with perceived spaciousness. These findings should be treated as context-specific design evidence rather than as universal prescriptions.

The proposed conceptual framework remains exploratory and requires validation through respondent-level modelling, randomised and calibrated stimulus presentation, immersive virtual reality, behavioural observation, eye-tracking, physiological measures, cross-cultural replication, and longitudinal investigation.

Taken together, these findings are limited to one student sample, drawn from one university, evaluating one familiar campus staircase through subjective, image-based evaluations; they should not be generalised beyond this context. Future work should accordingly prioritise larger and more heterogeneous samples, randomised presentation of stimuli, calibrated displays, and immersive or real-world environments, complemented by eye-tracking, EEG, and behavioural measurements, to test whether the exploratory patterns reported here extend beyond the present sample and setting.

8 Limitations

Future studies using independent Likert-type ratings for every colour alternative could test within-participant differences through repeated-measures procedures (RM-ANOVA or the Friedman test); studies that move beyond self-report to incorporate behavioural measures such as actual route choice and stopping/sitting tendencies, and that triangulate these with physiological indicators such as Virtual Reality (VR) and EEG, will help test the colour–space relationship in a multilayered fashion and produce stronger empirical evidence. Rather than weakening the present findings, these limitations define clear directions for methodological refinement and future evidence-based research.

9 Future research

Collectively, these developments would support the transition from descriptive colour preference studies toward predictive, adaptive, and human-centred environmental design models grounded in behavioural, physiological, and artificial intelligence-based evidence.

References

• 1

  AlbersJ. (2013). Interaction of Color. Yale University Press.

  • Google Scholar

• 2

  AlettaF.ObermanT.KangJ. (2018). Associations between positive health-related effects and soundscapes perceptual constructs: a systematic review. Int. J. Environ. Res. Public Health15 (11), 2392. 10.3390/ijerph15112392

  • Pubmed Abstract
  • CrossRef
  • Google Scholar

• 3

  ArnheimR. (2004). Visual Thinking. University of California Press.

  • Google Scholar

• 4

  ArthurP.PassiniR. (1992). Wayfinding: People, Signs, and Architecture. McGraw-Hill.

  • Google Scholar

• 5

  AtkinsonR. LAtkinsonR. CSmithE. E.BemD. J.Nolen-HoeksemaS. (2002). Psikolojiye Giriş [Introduction to Psychology]. 12th Edn. Editor AloganY.Ankara: Arkadaş Yayınevi

  • Google Scholar

• 6

  BacciF.MelcherD. (2011). Art and the Senses (Oxford University Press).

  • Google Scholar

• 7

  BlesserB.SalterL.-R. (2007). Spaces Speak, are you Listening? Experiencing Aural Architecture. MIT Press.

  • Google Scholar

• 8

  CamgözN.YenerC.GüvençD. (2002). Effects of hue, saturation, and brightness on preference. Color Res. and Appl.27 (3), 199–207. 10.1002/col.10051

  • CrossRef
  • Google Scholar

• 9

  CarpmanJ. R.GrantM. A. (2002). Design that Cares: Planning Health Facilities for Patients and Visitors. 2nd ed. San Francisco, CA: Jossey-Bass.

  • Google Scholar

• 10

  CoburnA.VartanianO.ChatterjeeA. (2017). Buildings, beauty, and the brain: a neuroscience of architectural experience. J. Cognitive Neurosci.29 (9), 1521–1531. 10.1162/jocn_a_01146

  • Pubmed Abstract
  • CrossRef
  • Google Scholar

• 11

  CohenJ. (1988). Statistical Power Analysis for the Behavioral Sciences. 2nd ed. Lawrence Erlbaum Associates.

  • Google Scholar

• 12

  DamasioA. (2006). Descartes’ Error: Emotion, Reason, and the Human Brain. London: Vintage.

  • Google Scholar

• 13

  ElliotA. J. (2015). Color and psychological functioning: a review of theoretical and empirical work. Front. Psychol.6, 368. 10.3389/fpsyg.2015.00368

  • Pubmed Abstract
  • CrossRef
  • Google Scholar

• 14

  ElliotA. J.MaierM. A. (2007). Color and psychological functioning. Curr. Dir. Psychol. Sci.16 (5), 250–254. 10.1111/j.1467-8721.2007.00514.x

  • CrossRef
  • Google Scholar

• 15

  ElliotA. J.MaierM. A. (2014). Color psychology: effects of perceiving color on psychological functioning in humans. Annu. Rev. Psychol.65, 95–120. 10.1146/annurev-psych-010213-115035

  • Pubmed Abstract
  • CrossRef
  • Google Scholar

• 16

  FekryS. A.HosnyS. S.SabryH.FathyF. (2025). Lighting impact on stress reduction: a study of correlated color temperature in workplace design utilizing neuroarchitecture principles. Build. Environ.278, 112972. Article 112972. 10.1016/j.buildenv.2025.112972

  • CrossRef
  • Google Scholar

• 17

  GageJ. (1999). Color and Meaning: Art, Science, and Symbolism. University of California Press.

  • Google Scholar

• 18

  GaoX.-P.XinJ. H. (2006). Investigation of human’s emotional responses on colors. Color Res. and Appl.31 (5), 411–417. 10.1002/col.20246

  • CrossRef
  • Google Scholar

• 19

  GaoX.-P.XinJ. H.SatoT.HansuebsaiA.ScalzoM.KajiwaraK.et al (2007). Analysis of cross-cultural color emotion. Color Res. and Appl.32 (3), 223–229. 10.1002/col.20321

  • CrossRef
  • Google Scholar

• 20

  GiffordR. (2014). Environmental Psychology: Principles and Practice. 5th ed. Colville, WA: Optimal Books.

  • Google Scholar

• 21

  GoldsteinE. B.CacciamaniL. (2022). Sensation and Perception. 11th ed. Boston, MA: Cengage Learning.

  • Google Scholar

• 22

  HidayetogluM. L.YildirimK.AkalinA. (2012). The effects of color and light on indoor wayfinding and the evaluation of the perceived environment. J. Environ. Psychol.32 (1), 50–58. 10.1016/j.jenvp.2011.09.001

  • CrossRef
  • Google Scholar

• 23

  Higuera-TrujilloJ. L.LlinaresC.MacagnoE. (2021). The cognitive-emotional design and study of architectural space: a scoping review of neuroarchitecture and its precursor approaches. Sensors21 (6), 2193. 10.3390/s21062193

  • Pubmed Abstract
  • CrossRef
  • Google Scholar

• 24

  HurlbertA. C.LingY. (2007). Biological components of sex differences in color preference. Curr. Biol.17 (16), R623–R625. 10.1016/j.cub.2007.06.022

  • Pubmed Abstract
  • CrossRef
  • Google Scholar

• 25

  JaglarzA. (2023). Perception of color in architecture and urban space. Buildings13 (8), 2000. 10.3390/buildings13082000

  • CrossRef
  • Google Scholar

• 26

  Jansen-OsmannP.WiedenbauerG. (2004). Wayfinding performance in and the spatial knowledge of a color-coded building for adults and children. Spatial Cognition Comput.4 (4), 337–358. 10.1207/s15427633scc0404_3

  • CrossRef
  • Google Scholar

• 27

  JonauskaiteD.Abdel-KhalekA. M.Abu-AkelA.Al-RasheedA. S.AntoniettiJ.-P.BogushevskayaV.et al (2020). Universal patterns in color–emotion associations are further shaped by linguistic and geographic proximity. Psychol. Sci.31 (10), 1245–1260. 10.1177/0956797620931480

  • CrossRef
  • Google Scholar

• 28

  JutziC. A.MöllerJ.HansenJ.KlacklJ.JonasE. (2025). Psychological needs in the built environment. J. Environ. Psychol.101, 102419. 10.1016/j.jenvp.2024.102419

  • CrossRef
  • Google Scholar

• 29

  KalantariS.TripathiV.KanJ.RoundsJ. D.MostafaviA.SnellR.et al (2022). Evaluating the impacts of color, graphics, and architectural features on wayfinding in healthcare settings using EEG data and virtual response testing. J. Environ. Psychol.79, 101744. Article 101744. 10.1016/j.jenvp.2021.101744

  • CrossRef
  • Google Scholar

• 30

  KnezI. (2001). Effects of colour of light on nonvisual psychological processes. J. Environ. Psychol.21 (2), 201–208. 10.1006/jevp.2000.0198

  • CrossRef
  • Google Scholar

• 31

  KüllerR.BallalS.LaikeT.MikellidesB.TonelloG. (2006). The impact of light and colour on psychological mood: a cross-cultural study of indoor work environments. Ergonomics49 (14), 1496–1507. 10.1080/00140130600858142

  • Pubmed Abstract
  • CrossRef
  • Google Scholar

• 32

  KwallekN.LewisC. M.RobbinsA. S. (1988). Effects of office interior color on workers' mood and productivity. Percept. Mot. Ski.66 (1), 123–128. 10.2466/pms.1988.66.1.123

  • CrossRef
  • Google Scholar

• 33

  KwallekN.LewisC. M.Lin-HsiaoJ. W. D.WoodsonH. (1996). Effects of nine monochromatic office interior colors on clerical tasks and worker mood. Color Res. and Appl.21 (6), 448–458. 10.1002/(sici)1520-6378(199612)21:6<448::aid-col7>3.0.co;2-w

  • CrossRef
  • Google Scholar

• 34

  LavrenovaO. (2023). Color semantics of the cultural landscape. Arts12 (3), 111. 10.3390/arts12030111

  • CrossRef
  • Google Scholar

• 35

  LiuB.WangC. (2025). Elderly-centric chromatics: unraveling the color preferences and visual needs of the elderly in smart APP interfaces. Int. J. Human–Computer Interact.41 (5), 3527–3536. 10.1080/10447318.2024.2338659

  • CrossRef
  • Google Scholar

• 36

  LlinaresC.Higuera-TrujilloJ. L.SerraJ. (2021). Cold and warm coloured classrooms. effects on students’ attention and memory measured through psychological and neurophysiological responses. Building and Environment196, 107726. 10.1016/j.buildenv.2021.107726

  • CrossRef
  • Google Scholar

• 37

  LynchK. (1960). The Image of the City. MIT Press.

  • Google Scholar

• 38

  MehtaR.ZhuR. (2009). Blue or red? Exploring the effect of color on cognitive task performances. Science323 (5918), 1226–1229. 10.1126/science.1169144

  • Pubmed Abstract
  • CrossRef
  • Google Scholar

• 39

  MengY.LiuW. (2025). “Design nature”: A color interpretation of Bauhaus school building in Dessau and its conceptual origins. Front. Archit. Res.14 (1), 32–61. 10.1016/j.foar.2024.06.012

  • CrossRef
  • Google Scholar

• 40

  NevesJ. D. (2024). Sensory architecture: the art of designing for all senses (2nd ed.). mapa lab.

  • Google Scholar

• 41

  Nolé FajardoM. L.Higuera-TrujilloJ. L.LlinaresC. (2023). Lighting, colour and geometry: which has the greatest influence on students’ cognitive processes?Front. Archit. Res.12 (3), 575–586. 10.1016/j.foar.2023.02.003

  • CrossRef
  • Google Scholar

• 42

  OberfeldD.HechtH. (2011). Fashion versus perception: the impact of surface lightness on the perceived dimensions of interior space. Hum. Factors53 (3), 284–298. 10.1177/0018720811407331

  • Pubmed Abstract
  • CrossRef
  • Google Scholar

• 43

  OuL.-C.LuoM. R.WoodcockA.WrightA. (2004). A study of colour emotion and colour preference. Part III: colour preference modeling. Color Res. and Appl.29 (5), 381–389. 10.1002/col.20047

  • CrossRef
  • Google Scholar

• 44

  PallasmaaJ. (2012). The Eyes of the Skin: Architecture and the Senses. 3rd ed. Wiley.

  • Google Scholar

• 45

  PalmerS. E.SchlossK. B. (2010). An ecological valence theory of human color preference. Proc. Natl. Acad. Sci.107 (19), 8877–8882. 10.1073/pnas.0906172107

  • Pubmed Abstract
  • CrossRef
  • Google Scholar

• 46

  ParizetE.KoehlV. (2011). Influence of train colour on loudness judgments. Acta Acustica United Acustica97 (2), 347–349. 10.3813/AAA.918414

  • CrossRef
  • Google Scholar

• 47

  ParkY.GuerinD. A. (2002). Meaning and preference of interior color palettes among four cultures. J. Interior Des.28 (1), 27–39. 10.1111/j.1939-1668.2002.tb00370.x

  • CrossRef
  • Google Scholar

• 48

  PassiniR. (1996). Wayfinding design: logic, application and some thoughts on universality. Des. Stud.17 (3), 319–331. 10.1016/0142-694X(96)00012-3

  • CrossRef
  • Google Scholar

• 49

  RasmussenS. E. (1964). Experiencing Architecture. 2nd ed. MIT Press.

  • Google Scholar

• 50

  SchlossK. B.PalmerS. E. (2017). An ecological framework for temporal and individual differences in color preferences. Vis. Res.141, 95–108. 10.1016/j.visres.2017.01.010

  • Pubmed Abstract
  • CrossRef
  • Google Scholar

• 51

  SongL.ZhangG.MaL.SilvennoinenJ.CongF. (2025). Comparative analysis of color emotional perception in art and non-art university students: hue, saturation, and brightness effects in the munsell color system. BMC Psychol.13, 650. 10.1186/s40359-025-03034-y

  • CrossRef
  • Google Scholar

• 52

  SpenceC. (2020). Using color to attract attention and influence behaviour. I-Perception11 (6), 1–34. 10.1177/2041669520968033

  • CrossRef
  • Google Scholar

• 53

  St-JeanP.ClarkO. G.JemtrudM. (2022). A review of the effects of architectural stimuli on human psychology and physiology. Build. Environ.219, 109182. 10.1016/j.buildenv.2022.109182

  • CrossRef
  • Google Scholar

• 54

  StampsA. E.III. (2011). Effects of area, height, elongation, and color on perceived spaciousness. Environ. Behav.43 (2), 252–273. 10.1177/0013916509354696

  • CrossRef
  • Google Scholar

• 55

  StoneN. J. (2003). Environmental view and color for a simulated telemarketing task. J. Environ. Psychol.23 (1), 63–78. 10.1016/S0272-4944(02)00107-X

  • CrossRef
  • Google Scholar

• 56

  SukH. J.IrtelH. (2010). Emotional response to color across media. Color Res. and Appl.35 (1), 64–77. 10.1002/col.20554

  • CrossRef
  • Google Scholar

• 57

  TsushimaY.OkadaS.KawaiY.SumitaA.AndoH.MikiM. (2020). Effect of illumination on perceived temperature. PLoS ONE15 (8), e0236321. 10.1371/journal.pone.0236321

  • Pubmed Abstract
  • CrossRef
  • Google Scholar

• 58

  ValdezP.MehrabianA. (1994). Effects of color on emotions. J. Exp. Psychol. General123 (4), 394–409. 10.1037//0096-3445.123.4.394

  • Pubmed Abstract
  • CrossRef
  • Google Scholar

• 59

  Vélez GonzálezL.Santamaría MelenchónA.Bailón MoyanoD.González-LezcanoR. A. (2026). Towards sustainable indoor lighting design: ensuring energy efficiency, health and human wellbeing—A review. Sustain. Dev.34 (S1), 1067–1095. 10.1002/sd.70201

  • CrossRef
  • Google Scholar

• 60

  von CastellC.OberfeldD.HechtH. (2014). The effect of furnishing on perceived spatial dimensions and spaciousness of interior space. PLoS ONE9 (11), e113267. 10.1371/journal.pone.0113267

  • Pubmed Abstract
  • CrossRef
  • Google Scholar

• 61

  WagemansJ.ElderJ. H.KubovyM.PalmerS. E.PetersonM. A.SinghM.et al (2012). A century of gestalt psychology in visual perception: I. Perceptual grouping and figure-ground organization. Psychol. Bull.138 (6), 1172–1217. 10.1037/a0029333

  • Pubmed Abstract
  • CrossRef
  • Google Scholar

• 62

  WilmsL.OberfeldD. (2018). Color and emotion: effects of hue, saturation, and brightness. Psychol. Res.82 (5), 896–914. 10.1007/s00426-017-0880-8

  • Pubmed Abstract
  • CrossRef
  • Google Scholar

• 63

  XuH.ZhaoJ.JinC.ZhuN.ChaiY. (2025). Research on the multi-sensory experience design of interior spaces from the perspective of spatial perception: a case study of suzhou coffee roasting factory. Buildings15 (8), 1393. 10.3390/buildings15081393

  • CrossRef
  • Google Scholar

• 64

  ZekiS.MariniL. (1998). Three cortical stages of colour processing in the human brain. Brain121, 1669–1685. 10.1093/brain/121.9.1669

  • Pubmed Abstract
  • CrossRef
  • Google Scholar

• 65

  ZhaoM.CrossleyT.ShinoharaH. (2025). The implications of EEG neurophysiological data in human-centered architectural design: a systematic review and bibliometric analysis. J. Environ. Psychol.103, 102550. 10.1016/j.jenvp.2025.102550

  • CrossRef
  • Google Scholar