The impact of objective urban features on perception of neighbourhood environments

Abstract

Perceptions of the neighbourhood environment can play an important role in promoting public health, yet modifying perceptions is challenging. Adjusting the built environment may be a pathway to influence perceptions. In addition to the physical environment, intrapersonal factors may shape perceptions. This study analysed data from several Japanese major cities to explore the association between objective and perceived neighbourhood environment attributes, stratified by age and gender. Perceived neighbourhood environment measures were adapted from established scales, while objective measures were derived from participants’ geographic address data. Multivariate linear regression was employed to assess these associations. All objective measures were positively associated with overall neighbourhood environment perception, and destination diversity presented the strongest association. Perceptions among those 65–69 were more strongly influenced by the physical environments of their neighbourhood, whether positively or negatively. Objective environmental measures have a greater positive impact on perception for females than for males, while males are more negatively affected in terms of perceptions of crime and traffic safety. These findings highlight how objective built environment attributes may shape residents’ perceptions across different demographic groups.

Introduction

The impact of the built environment on human health is a widely discussed topic, with one key pathway being its influence on health behaviours to promote positive health outcomes¹. For instance, living in walkable neighbourhoods can encourage physical activity, and regular physical activity contributes to greater happiness, fewer depressive symptoms, and a lower risk of type 2 diabetes^{1, 2, 3–4}. In addition to the objective built environment, individuals’ perceptions also play a role in supporting health behaviours. Positive built environment perceptions increase the likelihood of walking, cycling, and engaging in moderate physical activity^5,6. Altering residents’ perceptions of their neighbourhoods may encourage active behaviours and support public health efforts. However, perceptions depend on built environment attributes, along with personal and social factors^7,8. These factors interact in ways that make perceptions difficult to modify. However, since the built environment has long-lasting effects on large populations⁹, modifying its objective attributes may provide a more effective and sustainable way to influence perceptions. Evidence suggests that objective attributes influence how individuals evaluate their neighbourhoods^10,11. A study in the United States found that higher objective shop density correlated with more positive perceptions of accessibility¹⁰. Another study among older females in the United States showed that higher objective neighbourhood walkability was associated with better perceived environment scores¹¹. Investigating these links may help identify actionable strategies to modify the built environment to improve perceived environments and encourage physical activity.

Many studies have focused on the degree of alignment between objective and perceived neighbourhood built environments. A systematic review summarised findings on the consistency between these measures⁸. However, existing literature has rarely explored the extent to which objective attributes influence perceptions, and it remains unclear which specific features of the built environment have the strongest influence. Limited findings on the determinants of perceived environments restrict understanding the mechanisms underlying environmental perception formation⁷. As well, most research on this topic has examined Western settings and provides little evidence from Asian cities⁷. Asian urban areas typically feature higher population densities and more complex land use patterns than their Western counterparts¹². These unique built environment characteristics may shape residents’ perceptions of the environment in distinctive ways¹⁰. However, the extent to which relationships observed between objective and perceived environments in Western contexts apply to Asian urban contexts is unclear. Finally, how the relationships between objective and perceived neighbourhood environments vary across demographic groups remains unclear. While several sociodemographic characteristics may be relevant, age and gender have received relatively more attention in the literature due to their relevance to mobility, safety concerns, and how people interact with their neighbourhoods in daily life^{13, 14–15}. However, a recent review highlights that findings on the influence of age and gender on environmental perception are inconsistent⁷. This inconsistency underscores the need for further investigation in diverse contexts.

Therefore, using data from 21 major cities in Japan, the purpose of this study was to examine how objective built environment attributes influence residents’ perceptions of their neighbourhood environments, including variations by age (adults and those aged 65–69) and gender (females and males).

Methods

Data source and participants

This study leveraged cross-sectional data from the “Geo-social Survey for Urban Lifestyle Preferences”, specifically focusing on one of its components, which is called “Metro Survey”¹⁶. The “Metro Survey” was carried out online, targeting residents aged 20 to 69 in 21 major cities across Japan, including Tokyo Special Wards and other ordinance-designated cities. Figure 1 displays the 21 major cities targeted by the “Metro Survey”. Table 1 summarises basic characteristics, including city names, prefectures, and population (based on the 2020 Japan Population Census). The method of conducting the survey has been thoroughly described elsewhere^16,17, as well as in the Supplementary Information. In brief, data collection occurred from October to November 2020 following a quota sampling procedure to align the proportions of city/region, age, and gender with those of the general population. Responses were collected from registered panellists by the Nippon Research Centre Ltd survey company. An initial invitation email was first distributed to 234,483 panellists, resulting in 90,676 respondents who completed the screening survey. From those, 30,000 were invited for the “Metro Survey”. The survey captured details about participants’ residential areas, including perceptions of their neighbourhood environment and sociodemographic characteristics. The majority of participants (98.6%) also provided complete household residential address information, including street address and house number. All participants provided written informed consent and received points equal to 46 yen as a reward. The survey was approved by the Ethics Committee of Human Subjects Research, Graduate School of Engineering, Tohoku University (20A-10) and the methods were carried out in accordance with relevant ethical guidelines.

Fig. 1 [Images not available. See PDF.]

The 21 major cities targeted by the Metro Survey.

Table 1. Basic characteristics of the 21 Japanese cities.

City name	Prefecture	Population
Sapporo City	Hokkaido Prefecture	1,973,395
Sendai City	Miyagi Prefecture	1,096,704
Saitama City	Saitama Prefecture	1,324,025
Chiba City	Chiba Prefecture	974,951
Tokyo Special Wards	Tokyo Metropolis	9,733,276
Yokohama City	Kanagawa Prefecture	3,777,491
Kawasaki City	Kanagawa Prefecture	1,538,262
Sagamihara City	Kanagawa Prefecture	725,493
Niigata City	Niigata Prefecture	789,275
Shizuoka City	Shizuoka Prefecture	693,389
Hamamatsu City	Shizuoka Prefecture	790,718
Nagoya City	Aichi Prefecture	2,332,176
Kyoto City	Kyoto Prefecture	1,463,723
Osaka City	Osaka Prefecture	2,752,412
Sakai City	Osaka Prefecture	826,161
Kobe City	Hyogo Prefecture	1,525,152
Okayama City	Okayama Prefecture	724,691
Hiroshima City	Hiroshima Prefecture	1,200,754
Kitakyushu City	Fukuoka Prefecture	939,029
Fukuoka City	Fukuoka Prefecture	1,612,392
Kumamoto City	Kumamoto Prefecture	738,865

Measures

Outcome: perceived neighbourhood environment

The perceived neighbourhood environment was measured by eight items adapted from the Japanese version of the International Physical Activity Questionnaire Environmental Module (IPAQ-E) and the Abbreviated Neighbourhood Environment Walkability Scale Japanese version (ANEWS-J). IPAQ-E and ANEWS-J have acceptable test-retest reliability in the Japanese context^18,19. Participants were asked: “Do you have the following in your neighbourhood? Please think of it as a 10-minute walk from your home.” The items included: (1) access to shops (e.g., grocery stores); (2) access to daily life facilities (e.g., post offices and medical facilities); (3) access to public green spaces (e.g., parks and open spaces for walking and exercise); (4) access to public transport (e.g., bus stops and stations); (5) presence of paths (e.g., paths suitable for walking or jogging); (6) traffic safety (e.g., streets and intersections where traffic accidents are a concern); (7) crime safety (e.g., streets and places of concern in terms of security); and (8) aesthetics (e.g., attractive-looking buildings and landscapes). A 5-point scale from 1 (many) to 5 (none) was used for all items. Six items, including items 1 through 5 and item 8, were reverse coded. The composite score for perceived environments was calculated by summing the scores of the eight items (Cronbach’s alpha = 0.75). A higher composite score indicated a more positively perceived neighbourhood environment in terms of its walkability.

Exposure: objective neighbourhood environment

Four individual objective measures and two composite measures were considered in this study. The chocho-aza, a small neighbourhood unit in Japan, was first determined using the longitude and latitude information geocoded from the participant’s residential address²⁰. All objective measures were determined using a buffer zone that extends outward from the centre (most populated grid) of the chocho-aza, corresponding to a network distance of 1,000 m. This is roughly equal to a 10-minute walking distance for adults²¹.

One of the composite measures was the traditional walkability index, which was calculated using the average of the z-scores for three individual measures: population density, intersection density, and destination diversity. The 3-component index, adapted from a widely used method for calculating walkability by Frank, et al.²² and modified by Nakaya, et al.²³has been specially tailored for the Japanese context. This index has been validated in previous studies within Japan and related to residents’ walking behaviours²⁴. Population density data of the buffer were obtained from the 2020 Japanese census. Intersection density was derived from Japan road network data available in ArcGIS Geo Suite. This was done by dividing the number of intersections by the buffer area, identifying intersections as nodes where three or more road segments converge. Considering previous research and the Japanese setting, destination diversity was measured by calculating the types of destinations within the buffer, including a total of 18 different and common destinations such as parks, convenience shops, libraries, and more^24,25. Some of the destination data sources came from the National Land Numerical Information provided by the Ministry of Land, Infrastructure, Transport and Tourism of Japan. Other destination data were sourced from the 2021 Nippon Telegraph and Telephone Corporation telephone directory. More details are described in a previous study²⁴.

Another composite measure was space syntax walkability (SSW), developed by Koohsari, et al.²⁶ and included two individual measures. It was calculated by standardising the sum of the z-score for population density and twice the z-score for street integration²⁶. The data required to calculate SSW is relatively easy to access, making it suitable for areas where geographic information, such as retail floor area, is difficult to obtain. The population density data used in SSW aligns with the traditional walkability index. Street integration was automatically calculated by employing The Space Syntax Toolkit and Depthmap X, created by University College London^27,28. To measure integration values, this study applied angular analysis²⁹, which was suitable for complex street patterns and has been used in Japanese cities³⁰. This individual measure refers to the proximity of a specific street to others within the network³¹. Unlike intersection density, street integration captures street network features from another perspective³². Intersection density represents a numerical concept based on the number of intersections. In contrast, street integration reflects a topological concept considering how streets connect^31,32.

Stratification: age and gender

Age was self-reported by selecting from multiple age ranges. For stratification, it was treated as a binary variable. Participants aged 20–64 were categorised as adults, while those aged 65–69 were analysed as a separate age group. Gender was also self-reported and included male and female categories.

Covariates: sociodemographic variables

A systematic review identified several sociodemographic variables associated with the perceived walkable built environment⁷. These variables were included as covariates in the main effects model, including age, gender, education level, marital status, gross annual household income, length of residence in the neighbourhood, and employment status. In stratified analyses, age or gender was used for stratification, while the other variables remained as covariates.

Statistical analysis

Descriptive analysis of participants’ demographic features and neighbourhood environment measures was reported by computing the number and percentage (%) or mean and standard deviation. Differences between genders and age groups on neighbourhood environment measures were compared using independent t-tests. A one-way analysis of variance was used to calculate the city-level intra-cluster correlation (ICC). The variation in perceived environment attributes explained by differences among the 21 Japanese cities was negligible (e.g., the ICC for the overall perceived environment was 0.03). This value is below the standard minimum threshold of 0.05, indicating that the individual data may not exhibit significant nested group relationships³³. Therefore, multivariate linear regression was conducted to analyse objective environmental attributes in relation to perceived attributes, with perceived attributes treated as the dependent variable. Each objective environmental attribute was included in the model separately. When conducting analysis, all objective measures were standardised using z-scores. Unstandardised regression coefficients (B) and 95% confidence intervals (CI) were assessed for all estimates. Interactions between objective neighbourhood measures with gender and age were examined and stratified analyses were employed when significant interaction effects were identified. All analyses were performed using STATA 15.1, with statistical significance set at p < 0.05.

Results

Characteristics of participants

A total of 25,340 participants were included in this study after excluding those who selected “I don’t know” for certain sociodemographic items, as well as those with insufficiently precise geographic information for targeting their chocho-aza. Table 2 illustrates characteristics of the participants and compares them to the population living in 21 Japanese cities. Most of the participants were adults aged 20–64 years old (93.1%), with higher education attainment (81.5%) and currently employed (76.4%). About half of the participants were male (51.5%), married (57.4%), and reported a gross annual household income exceeding 5,000,000 yen (52.9%). Moreover, a larger proportion of participants had resided in their current neighbourhood for 5 to less than 20 years (38.1%). The sample was broadly comparable to the population in terms of age, gender, employment, marital status, and residential duration. However, individuals with tertiary education were overrepresented: 81.5% of participants had completed tertiary education compared with 61.5% in the population.

Table 2. Characteristics of participants and population residing in 21 Japanese cities.

Variable	Study participants (N = 25,340) N (%)	Population census (N = 23,378,992) N (%)
Age
20–64 years	23,589 (93.1)	21,354,704 (91.3)
65–69 years	1,751 (6.9)	2,024,288 (8.7)
Gender
Male	13,051 (51.5)	11,648,268 (49.8)
Female	12,289 (48.5)	11,730,724 (50.2)
Education level
Below tertiary	4,700 (18.5)	6,791,570 (38.5)
Tertiary or higher	20,640 (81.5)	10,830,266 (61.5)
Marital status
Married	14,534 (57.4)	12,909,172 (60.2)
Single	10,806 (42.6)	8,544,568 (39.8)
Gross annual household income
<¥5,000,000	11,927 (47.1)	–
≥¥5,000,000	13,413 (52.9)	–
Employment status
Employed	19,356 (76.4)	14,925,572 (78.3)
Unemployed	5,984 (23.6)	4,131,453 (21.7)
Length of residence
Less than 5 years	7,934 (31.3)	5,850,479 (30.7)
5 to less than 20 years	9,649 (38.1)	7,482,603 (39.3)
20 years or more	7,757 (30.6)	5,728,874 (30.1)

Table 3 shows the characteristics of neighbourhood environment measures for the entire sample and also stratified by gender and age. From the total sample, the average population density of the participants’ neighbourhood built environment was 13,580.6 (SD = 6,457.3) people per km². The average number of destination types was 15.4 (SD = 2.6), and there was an average of 241.3 (SD = 76.5) intersections per km². The mean value of street integration was 215.8 (SD = 84.7). In terms of the perceived environment, the highest average score was 4.0 (SD = 0.9) for perceived access to public transport, while the lowest average score was 2.5 (SD = 0.9) for perceived traffic safety. Significant differences in objective measures were found between adults and those aged 65–69 (p < 0.01). From the absolute differences in mean values, the 65–69 age group’s neighbourhood environment had 1,009.1 fewer people per km², 0.4 fewer destination types, 10.7 fewer intersections per km², and a street integration value that was 12.3 lower compared to that of adults. Significant gender differences were observed in perceived access to shops, daily life facilities, public green spaces, public transport, presence of paths, crime safety, and overall perceived walkability (all p < 0.01). Generally, females reported higher perceptions than males, with increases of 0.1 points in access to shops, public green spaces, and public transport, an increase of 0.2 points in daily life facilities, and a decrease of 0.1 points in crime safety. Females also rated overall walkability 0.3 points higher than males. Significant age-related differences were observed in all perception measures (p < 0.01) except for access to shops. Compared to adults, those aged 65–69 reported higher perceptions, with increases of 0.1 points for daily life facilities, public transport, and aesthetics, and increases of 0.2 points for public green spaces, presence of paths, traffic safety, and crime safety. Their overall perceived walkability score was 1.1 points higher than that of adults.

Table 3. Characteristics of neighbourhood environment measures (N = 25,340).

Neighbourhood environment measures	Total sample Mean (SD) (N = 25,340)	Mean (SD)		pa	Mean (SD)		pa
		Male (N = 13,051)	Female (N = 12,289)		Adults (N = 23,589)	Age 65–69 (N = 1,751)
Objective measures
Population density b	13,580.6 (6,457.3)	13,593.3 (6,465.7)	13,567.1 (6,448.5)	ns	13,650.4 (6,447.9)	12,641.3 (6,511.8)	< 0.01
Destination diversity c	15.4 (2.6)	15.4 (2.6)	15.4 (2.6)	ns	15.4 (2.6)	15.0 (3.0)	< 0.01
Intersection density d	241.3 (76.5)	241.7 (76.2)	240.9 (76.7)	ns	242.0 (76.3)	231.4 (78.0)	< 0.01
Street integration	215.8 (84.7)	215.9 (84.6)	215.6 (84.8)	ns	216.6 (84.5)	204.3 (86.7)	< 0.01
Traditional walkability index	0.0 (0.8)	0.0 (0.8)	−0.0 (0.8)	ns	0.0 (0.8)	−0.1 (0.9)	< 0.01
SSW e	0.0 (1.0)	0.0 (1.0)	−0.0 (1.0)	ns	0.0 (1.0)	−0.2 (1.0)	< 0.01
Perceived measures
Access to shops	3.9 (1.0)	3.8 (1.0)	3.9 (1.0)	< 0.01	3.9 (1.0)	3.8 (1.1)	ns
Access to daily life facilities	3.9 (0.9)	3.8 (0.9)	4.0 (0.9)	< 0.01	3.9 (0.9)	4.0 (0.9)	< 0.01
Access to public green spaces	3.9 (0.9)	3.8 (0.9)	3.9 (0.9)	< 0.01	3.9 (0.9)	4.1 (0.9)	< 0.01
Access to public transport	4.0 (0.9)	4.0 (0.9)	4.1 (0.9)	< 0.01	4.0 (0.9)	4.1 (0.8)	< 0.01
Presence of paths	3.8 (1.0)	3.8 (1.0)	3.8 (1.0)	< 0.01	3.8 (1.0)	4.0 (0.9)	< 0.01
Traffic safety	2.5 (0.9)	2.5 (0.9)	2.5 (1.0)	ns	2.5 (0.9)	2.7 (0.9)	< 0.01
Crime safety	2.7 (0.9)	2.8 (0.9)	2.7 (0.9)	< 0.01	2.7 (0.9)	2.9 (0.9)	< 0.01
Aesthetics	3.1 (1.0)	3.1 (1.0)	3.1 (1.0)	ns	3.1 (1.0)	3.2 (0.9)	< 0.01
Overall perceived walkability	27.8 (4.0)	27.6 (3.9)	27.9 (4.2)	< 0.01	27.7 (4.0)	28.8 (4.2)	< 0.01

^a Based on independent t-tests.

^b The number of people per km².

^c The total count for 18 types of destinations.

^d The number of intersections per km².

^e Space syntax walkability.

Associations between objective measures and perceived neighbourhood environment

The associations between objective and perceived environment measures are presented in Table 4, adjusted for covariates. Model 1 presents the main effects model. All individual objective measures were positively associated with overall environment perception, with destination diversity showing the strongest positive association (B = 0.87, 95% CI = 0.83 to 0.92). Both composite objective walkability measures, the traditional walkability index and SSW, were significantly and positively associated with perceived environment (B = 0.74, 95% CI = 0.69 to 0.79 and B = 0.54, 95% CI = 0.49 to 0.59, respectively), with the traditional walkability index showing a stronger association. Among the two individual objective street layout measures—intersection density and street integration—only intersection density showed a significant negative association with perceived access to public green spaces (B=−0.01, 95% CI=−0.03 to −0.00). For aesthetics perceptions, intersection density was negatively associated (B=−0.01, 95% CI=−0.03 to −0.00), whereas street integration had a significant positive relationship (B = 0.03, 95% CI = 0.02 to 0.04).

Table 4. Associations between objective and perceived neighbourhood environment measures (N = 25,340).

Objective neighbourhood environment measures	Perceived neighbourhood environment measures B (95% CI)
	Access to shops	Access to daily life facilities	Access to public green spaces	Access to public transport	Presence of paths	Traffic safety	Crime safety	Aesthetics	Overall perceived environment
	Model 1a
Population density	0.26 (0.25, 0.27) **	0.22 (0.21, 0.23) **	0.03 (0.02, 0.04) **	0.21 (0.20, 0.23) **	−0.01 (−0.03, −0.00) *	−0.06 (−0.07, −0.05) **	−0.02 (−0.03, −0.01) **	0.10 (0.09, 0.12) **	0.73 (0.68, 0.78) **
Destination diversity	0.33 (0.32, 0.35) **	0.30 (0.29, 0.31) **	0.04 (0.03, 0.05) **	0.25 (0.24, 0.26) **	−0.02 (−0.03, −0.01) **	−0.11 (−0.12, −0.10) **	−0.04 (−0.05, −0.03) **	0.11 (0.10, 0.13) **	0.87 (0.83, 0.92) **
Intersection density	0.18 (0.17, 0.19) **	0.14 (0.13, 0.15) **	−0.01 (−0.03, −0.00) *	0.11 (0.10, 0.12) **	−0.08 (−0.10, −0.07) **	−0.05 (−0.06, −0.04) **	−0.03 (−0.05, −0.02) **	−0.01 (−0.03, −0.00) *	0.24 (0.19, 0.29) **
Street integration	0.21 (0.20, 0.22) **	0.18 (0.17, 0.19) **	−0.01 (−0.02, 0.00)	0.14 (0.13, 0.16) **	−0.07 (−0.08, −0.06) **	−0.08 (−0.09, −0.07) **	−0.04 (−0.05, −0.03) **	0.03 (0.02, 0.04) **	0.36 (0.31, 0.41) **
Traditional walkability index	0.31 (0.30, 0.32) **	0.27 (0.26, 0.28) **	0.02 (0.01, 0.04) **	0.23 (0.22, 0.24) **	−0.05 (−0.06, −0.04) **	−0.09 (−0.10, −0.08) **	−0.04 (−0.05, −0.03) **	0.08 (0.07, 0.09) **	0.74 (0.69, 0.79) **
SSW d	0.25 (0.24, 0.27) **	0.21 (0.20, 0.22) **	0.01 (−0.01, 0.02)	0.19 (0.18, 0.20) **	−0.06 (−0.07, −0.05) **	−0.08 (−0.09, −0.07) **	−0.04 (−0.05, −0.03) **	0.06 (0.05, 0.07) **	0.54 (0.49, 0.59) **
	Model 2b
Population density	0.12 (0.08, 0.17) **	0.09 (0.05, 0.13) **	0.04 (−0.00, 0.08)	0.06 (0.02, 0.10) **	−0.02 (−0.07, 0.03)	−0.07 (−0.11, −0.02) **	−0.02 (−0.06, 0.03)	0.01 (−0.04, 0.06)	0.22 (0.03, 0.41) *
Destination diversity	0.04 (0.01, 0.08) *	0.04 (0.00, 0.07) *	0.00 (−0.03, 0.04)	−0.00 (−0.04, 0.03)	−0.07 (−0.11, −0.03) **	−0.07 (−0.11, −0.03) **	−0.01 (−0.05, 0.03)	−0.01 (−0.05, 0.04)	−0.06 (−0.23, 0.10)
Intersection density	0.10 (0.06, 0.15) **	0.08 (0.03, 0.12) **	0.01 (−0.04, 0.05)	0.04 (0.00, 0.09) *	−0.04 (−0.08, 0.01)	−0.08 (−0.13, −0.04) **	−0.02 (−0.07, 0.02)	0.02 (−0.03, 0.07)	0.11 (−0.08, 0.30)
Street integration	0.09 (0.04, 0.13) **	0.06 (0.02, 0.10) **	−0.00 (−0.05, 0.04)	0.02 (−0.02, 0.06)	−0.06 (−0.11, −0.02) **	−0.10 (−0.14, −0.05) **	−0.05 (−0.09, −0.00) *	−0.01 (−0.06, 0.04)	−0.05 (−0.24, 0.14)
Traditional walkability index	0.10 (0.06, 0.14) **	0.08 (0.04, 0.12) **	0.02 (−0.02, 0.06)	0.04 (−0.00, 0.07)	−0.05 (−0.09, −0.00) *	−0.08 (−0.13, −0.04) **	−0.02 (−0.06, 0.02)	0.01 (−0.03, 0.06)	0.09 (−0.09, 0.27)
SSW d	0.10 (0.05, 0.14) **	0.07 (0.03, 0.11) **	0.01 (−0.03, 0.05)	0.03 (−0.01, 0.07)	−0.05 (−0.10, −0.01) *	−0.09 (−0.14, −0.05) **	−0.04 (−0.08, 0.00)	−0.00 (−0.05, 0.04)	0.02 (−0.16, 0.21)
	Model 3c
Population density	0.05 (0.02, 0.07) **	0.05 (0.03, 0.07) **	0.01 (−0.02, 0.03)	0.05 (0.03, 0.07) **	−0.01 (−0.03, 0.02)	0.02 (−0.01, 0.04)	0.06 (0.04, 0.09) **	0.02 (−0.01, 0.04)	0.24 (0.14, 0.33) **
Destination diversity	0.07 (0.04, 0.09) **	0.06 (0.04, 0.08) **	0.02 (−0.01, 0.04)	0.06 (0.04, 0.08) **	0.01 (−0.01, 0.04)	0.01 (−0.02, 0.03)	0.04 (0.02, 0.07) **	0.02 (−0.01, 0.04)	0.28 (0.19, 0.38) **
Intersection density	0.03 (0.01, 0.06) **	0.03 (0.00, 0.05) *	0.00 (−0.02, 0.03)	0.03 (0.01, 0.05) *	−0.00 (−0.03, 0.02)	0.00 (−0.02, 0.03)	0.03 (0.01, 0.06) **	−0.02 (−0.04, 0.01)	0.11 (0.01, 0.21) *
Street integration	0.04 (0.02, 0.06) **	0.03 (0.01, 0.05) **	0.00 (−0.02, 0.03)	0.04 (0.02, 0.06) **	0.00 (−0.02, 0.02)	0.01 (−0.01, 0.03)	0.03 (0.01, 0.05) **	0.01 (−0.02, 0.03)	0.16 (0.06, 0.26) **
Traditional walkability index	0.06 (0.04, 0.08) **	0.05 (0.03, 0.07) **	0.01 (−0.01, 0.03)	0.06 (0.04, 0.08) **	0.00 (−0.02, 0.02)	0.01 (−0.01, 0.03)	0.06 (0.03, 0.08) **	0.01 (−0.02, 0.03)	0.25 (0.16, 0.35) **
SSW d	0.05 (0.02, 0.07) **	0.04 (0.02, 0.06) **	0.00 (−0.02, 0.03)	0.04 (0.02, 0.07) **	−0.00 (−0.03, 0.02)	0.01 (−0.01, 0.04)	0.05 (0.02, 0.07) **	0.01 (−0.01, 0.04)	0.21 (0.11, 0.30) **

^a Main effects model, adjusted for all covariates (age, gender, education level, marital status, gross annual household income, employment status, and length of residence), without interaction terms.

^b Age interaction model, adjusted for all covariates except age, using adults as the reference group.

^c Gender interaction model, adjusted for all covariates except gender, using males as the reference group.

^d Space syntax walkability.

B = unstandardised regression coefficient; 95% CI = 95% confidence interval. All objective built environment measures were standardised.

*p < 0.05, **p < 0.01.

Interaction effects between objective measures and gender or age in relation to perceived measures were also explored. Model 2 presents the interaction model for age. The interaction between population density and age was significantly and positively associated with overall perceived environment (B = 0.22, 95% CI = 0.03 to 0.41). Among all objective measures, this interaction showed the strongest positive associations with multiple individual perceived environment measures, including perceived access to shops (B = 0.12, 95% CI = 0.08 to 0.17), daily life facilities (B = 0.09, 95% CI = 0.05 to 0.13), and public transport (B = 0.06, 95% CI = 0.02 to 0.10). In contrast, the interaction between destination diversity and age had the strongest negative association with the perceived presence of paths (B=−0.07, 95% CI=−0.11 to −0.03). The interaction between street integration and age showed the strongest negative association with traffic safety perception (B=−0.10, 95% CI=−0.14 to −0.05) and was the only interaction with age that had a significant negative association with crime safety perception (B=−0.05, 95% CI=−0.09 to −0.00).

Model 3 presents the interaction model for gender. All interactions between objective measures and gender were significantly and positively associated with the overall perceived environment, with the strongest association observed for destination diversity (B = 0.28, 95% CI = 0.19 to 0.38). Furthermore, all interactions between objective measures and gender were positively associated with four different individual perceived environment measures. The interaction between destination diversity and gender showed the strongest associations with three of these measures, including perceived access to shops (B = 0.07, 95% CI = 0.04 to 0.09), daily life facilities (B = 0.06, 95% CI = 0.04 to 0.08), and public transport (B = 0.06, 95% CI = 0.04 to 0.08). After identifying significant interactions, analyses were performed separately for different age groups or genders. Table 5 presents the stratification results. When stratified by gender, significant differences emerged between males and females: population density, traditional walkability index, and SSW were found to be negatively associated solely with males’ perception of crime safety (B=−0.05, 95% CI=−0.07 to −0.04, B=−0.06, 95% CI=−0.08 to −0.05, and B=−0.06, 95% CI=−0.08 to −0.05, respectively).

Table 5. Associations between objective and perceived neighbourhood environment measures: stratified by age and gender (N = 25,340).

Objective neighbourhood environment measures		Perceived neighbourhood environment measures
		B (95% CI)
		Access to shops	Access to daily life facilities	Access to public green spaces	Access to public transport	Presence of paths	Traffic safety	Crime safety	Aesthetics	Overall perceived environment
		Stratified by agea
Population density	Adults	0.25 (0.24, 0.26) **	0.21 (0.20, 0.22) **		0.21 (0.20, 0.22) **		−0.05 (−0.07, −0.04) **			0.72 (0.67, 0.77) **
	Age 65–69	0.37 (0.32, 0.42) **	0.30 (0.26, 0.34) **		0.27 (0.23, 0.31) **		−0.12 (−0.17, −0.08) **			0.92 (0.73, 1.11) **
Destination diversity	Adults	0.33 (0.32, 0.34) **	0.30 (0.29, 0.31) **			−0.01 (−0.03, −0.00) *	−0.10 (−0.11, −0.09) **
	Age 65–69	0.37 (0.33, 0.41) **	0.33 (0.30, 0.37) **			−0.09 (−0.12, −0.05) **	−0.17 (−0.21, −0.13) **
Intersection density	Adults	0.17 (0.16, 0.19) **	0.14 (0.13, 0.15) **		0.11 (0.10, 0.12) **		−0.04 (−0.06, −0.03) **
	Age 65–69	0.27 (0.23, 0.32) **	0.21 (0.17, 0.25) **		0.16 (0.12, 0.19) **		−0.13 (−0.17, −0.08) **
Street integration	Adults	0.21 (0.19, 0.22) **	0.17 (0.16, 0.18) **			−0.07 (−0.08, −0.05) **	−0.07 (−0.09, −0.06) **	−0.04 (−0.05, −0.03) **
	Age 65–69	0.29 (0.24, 0.33) **	0.23 (0.19, 0.27) **			−0.13 (−0.17, −0.09) **	−0.17 (−0.21, −0.13) **	−0.09 (−0.13, −0.05) **
Traditional walkability index	Adults	0.30 (0.29, 0.32) **	0.26 (0.25, 0.27) **			−0.04 (−0.06, −0.03) **	−0.08 (−0.09, −0.07) **
	Age 65–69	0.40 (0.36, 0.44) **	0.33 (0.30, 0.37) **			−0.10 (−0.14, −0.06) **	−0.17 (−0.21, −0.13) **
SSW c	Adults	0.25 (0.23, 0.26) **	0.21 (0.20, 0.22) **			−0.06 (−0.07, −0.04) **	−0.08 (−0.09, −0.06) **
	Age 65–69	0.34 (0.30, 0.39) **	0.27 (0.23, 0.31) **			−0.11 (−0.15, −0.07) **	−0.17 (−0.21, −0.13) **
		Stratified by genderb
Population density	Male	0.23 (0.22, 0.25) **	0.19 (0.18, 0.21) **		0.19 (0.18, 0.21) **			−0.05 (−0.07, −0.04) **		0.61 (0.55, 0.68) **
	Female	0.29 (0.27, 0.30) **	0.24 (0.23, 0.26) **		0.24 (0.22, 0.25) **			0.01 (−0.00, 0.03)		0.86 (0.79, 0.94) **
Destination diversity	Male	0.30 (0.29, 0.32) **	0.27 (0.26, 0.29) **		0.22 (0.20, 0.23) **			−0.06 (−0.08, −0.04) **		0.73 (0.67, 0.80) **
	Female	0.37 (0.35, 0.39) **	0.33 (0.32, 0.35) **		0.28 (0.26, 0.29) **			−0.02 (−0.03, −0.00) *		1.02 (0.95, 1.10) **
Intersection density	Male	0.16 (0.15, 0.18) **	0.13 (0.12, 0.15) **		0.10 (0.09, 0.12) **			−0.05 (−0.07, −0.04) **		0.19 (0.12, 0.26) **
	Female	0.20 (0.18, 0.22) **	0.16 (0.14, 0.17) **		0.13 (0.11, 0.14) **			−0.02 (−0.03, −0.00) *		0.29 (0.22, 0.37) **
Street integration	Male	0.19 (0.18, 0.21) **	0.16 (0.15, 0.18) **		0.13 (0.11, 0.14) **			−0.06 (−0.07, −0.04) **		0.28 (0.22, 0.35) **
	Female	0.23 (0.22, 0.25) **	0.19 (0.18, 0.21) **		0.16 (0.15, 0.18) **			−0.03 (−0.04, −0.01) **		0.44 (0.37, 0.51) **
Traditional walkability index	Male	0.28 (0.26, 0.30) **	0.24 (0.23, 0.25) **		0.20 (0.19, 0.22) **			−0.06 (−0.08, −0.05) **		0.62 (0.55, 0.68) **
	Female	0.34 (0.33, 0.36) **	0.30 (0.28, 0.31) **		0.26 (0.24, 0.27) **			−0.01 (−0.03, 0.01)		0.88 (0.80, 0.95) **
SSW c	Male	0.23 (0.21, 0.25) **	0.19 (0.18, 0.21) **		0.17 (0.15, 0.18) **			−0.06 (−0.08, −0.05) **		0.44 (0.37, 0.50) **
	Female	0.28 (0.26, 0.30) **	0.23 (0.22, 0.25) **		0.21 (0.19, 0.22) **			−0.02 (−0.03, 0.00)		0.64 (0.57, 0.72) **

^a Adjusted for gender, education level, marital status, gross annual household income, employment status, and length of residence.

^b Adjusted for age, education level, marital status, gross annual household income, employment status, and length of residence.

^c Space syntax walkability.

B = unstandardised regression coefficient; 95% CI = 95% confidence interval. All objective built environment measures were standardised.

*p < 0.05, **p < 0.01.

Discussion

Using a large sample of residents from 21 cities in Japan, this study investigated how objective built environment attributes influence residents’ perceptions of their neighbourhoods, a key factor associated with health outcomes³⁴. Consistent with some previous studies^7,11, objective measures—including two composite indices (the traditional walkability index and SSW) and their components—showed significant associations with perceived neighbourhood walkability. This study also extends the understanding of these associations by examining variations across demographic groups such as age and gender.

More specifically, destination diversity emerged as the strongest correlate of overall perceived neighbourhood walkability among all objective measures. One possible explanation could be the presence of diverse neighbourhood facilities that meet people’s daily needs, making walking in the area feel more meaningful³⁵. A study conducted in Canada reported different findings, showing null or adverse effects of objectively audited destination diversity on perceived walkability³⁶. Cultural differences may explain this discrepancy, but differences in how land-use mix was defined could also be a factor. Moreover, the Canadian study measured the environment along specific street segments rather than within participants’ residential neighbourhoods. Among the composite objective measures, both SSW and the traditional walkability index were positively associated with a better perceived neighbourhood environment, with the latter showing a stronger association. This finding aligns with previous research suggesting that these two measures are closely related and somewhat interchangeable²⁶. Given the challenges in obtaining parcel-level data, SSW serves as a practical and simplified alternative for capturing traditional walkability features.

We also found that two street layout attributes—intersection density and street integration—were positively associated with perceived neighbourhood walkability. However, when examining perceived access to public green spaces, only intersection density showed a significant negative association. Additionally, intersection density was negatively associated with aesthetics, whereas street integration showed a positive association. Previous studies have not directly examined the relationship between these street layout measures and aesthetics. A higher density of intersections may require pedestrians to stop frequently to cross streets³⁷. When navigating complex pedestrian environments, individuals must focus on safety-related elements such as traffic signals and vehicles³⁸, which may reduce attention to surrounding landscapes. Another study suggests that, from an aesthetic perspective, people may prefer sinuous street forms over straight ones³⁹, which contrasts with our findings on street integration. However, our measure of aesthetics was based on landscapes and buildings. Highly integrated streets often feature more commercial destinations³², which may create enjoyable building facades and potentially enhance aesthetic perceptions. The relationship between intersection density and perceived environment may also vary by urban context. In highly urbanised areas, higher intersection density may promote walkability, while in peri-urban or car-oriented areas, it may reflect traffic-dominant street patterns. Future studies should include measures of urbanity to explore how these contextual differences influence residents’ perceptions.

Furthermore, the interaction effects and stratified analyses showed that the associations between objective attributes and perceived walkability may be different across demographic groups. The age differences indicate that individuals aged 65–69 are generally more influenced by objective environmental factors, whether positively or negatively, compared to younger individuals. For example, population density had a stronger positive impact on the perceptions of those aged 65–69 regarding the overall neighbourhood environment, access to shops, daily life facilities, and public transport. Higher population density is usually associated with a greater number of local shops, daily amenities, and public transport options, which can better support those aged 65–69 in travelling outside^40,41. In this age group, the onset of physical decline may increase the importance of nearby destinations. Support from neighbourhood environment conditions might be more likely to enhance their perception of the environment. Another difference in the age group is that, compared to adults, destination diversity had a stronger negative impact on the perceived presence of paths among those aged 65–69. At the same time, higher street integration was more strongly associated with lower perceptions of traffic and crime safety. Paths are generally suitable for recreational walking⁴². Areas with high destination diversity tend to have more pedestrian traffic⁴³, which may not be ideal for leisure activities. Additionally, individuals aged 65–69 may be more concerned about potential risks in such environments, such as falling⁴⁴, which may reduce their positive perceptions of walking and jogging trails. Moreover, several studies indicate that higher street integration is linked to increased crime and traffic accidents^{45, 46–47}, supporting our findings on perceived safety. The decline in physical function among those aged 65–69 may heighten sensitivity to safety concerns⁴⁸.

Gender differences indicated that, compared to males, females’ overall perception of the environment was more influenced by objective environmental factors. However, when objective environmental conditions negatively affect specific perception attributes, males appeared more impacted. For instance, destination diversity was more strongly associated with improvements in females’ overall environmental perception, as well as perceived access to shops, daily life facilities, and public transport. In the Japanese cultural context, females are usually responsible for daily shopping, which might make them more sensitive to destination diversity⁴⁹. Moreover, this study found that population density and the walkability index were negatively associated with crime safety perceptions among males but not females. Males and females may experience and evaluate the built environment in distinctive ways⁵⁰. For instance, males might perceive higher density areas as less safe because of concerns about conflict in crowded environments⁵¹. However, these safety concerns may be mitigated for females, as their perceived insecurity may arise from the threat of individual physical violence, while the presence of more pedestrians can serve as a form of surveillance⁵². Future research should explore the mechanisms for these differences.

This study demonstrated that several objective attributes can be linked to perceived measures of the built environment associated with walkability. Urban designers and public health practitioners can target these measurable and quantifiable attributes to positively influence residents’ perceptions of their neighbourhoods and thereby influence behaviour. For example, several policies in Japan have been implemented to support people in living a comfortable and walk-oriented lifestyle, such as the “Urban Walkability Promotion Policy”⁵³. Some challenges of this approach should also be considered. Modifying the objective built environment cannot fully address personal and cultural biases that influence perceptions. Additionally, (re)designing cities can require significant costs and funding, and once built, the built environment is often difficult to modify due to various barriers and constraints. Complementary strategies, such as community engagement or education, may also be needed alongside physical environment changes⁵⁴. Future studies can explore how these complementary strategies can maximise the effectiveness of built environment interventions.

This study had some limitations. First, this cross-sectional study cannot determine the true causal relationship between objective built environment measures and perceived environments. Second, the 1,000 m buffer zone was used to define objective neighbourhoods which may not be relevant to perceived neighbourhoods. In this study, participants considered a 10-minute walk when evaluating their perceived neighbourhoods which may roughly equate to a distance of 1,000 m. Another limitation is the use of network buffers instead of Euclidean buffers. While network buffers more accurately reflect pedestrian access, their geographic sizes vary depending on the surrounding street layout. This may result in larger buffers capturing more destination types when diversity is measured by their count. Future research could explore sensitivity as associations may vary with buffer size definitions. Third, this study did not examine nonlinear relationships between objective and perceived built environment measures, though such effects may exist. Moreover, the representativeness of the study sample may be limited. Participants with tertiary education were overrepresented compared to the general population of the 21 Japanese cities, which may influence the generalisation of the findings. Finally, the R-squared values for objective environmental attributes in predicting overall environmental perception were relatively low in our regression models. This suggests that the objective attributes included in this study have limited explanatory power for variations in perception. While these attributes play a role, they account for only a small portion of the built environment factors that may shape perceptions. Other unmeasured factors may better explain perceptions, and future research should consider incorporating additional demographic variables and environmental characteristics. Despite these limitations, this study’s key strengths are its use of a large, diverse sample across multiple major cities in Japan and diverse measures of objective and perceived environments.

Author contributions

JL: Conceptualisation, Methodology, Writing—Original draft preparation. MJK: Supervision, Methodology, Writing—Reviewing and Editing. JZ: Writing—Reviewing and Editing. ATK: Writing—Reviewing and Editing. GRM: Writing—Reviewing and Editing. KO: Writing—Reviewing and Editing. TN: Writing—Reviewing and Editing. RT: Writing—Reviewing and Editing. RW: Writing—Reviewing and Editing. TH: Supervision, Methodology, Writing—Reviewing and Editing.

Funding

MJK is supported by the Japan Society for the Promotion of Science (JSPS) Grants-in-Aid for Scientific Research (KAKENHI) (grant JP23K09701). GRM is supported by a Canadian Institutes of Health Research Foundations Scheme Grant (FDN-154331). KO is supported by the JSPS KAKENHI (grant JP20H04113). TN was supported by the JSPS KAKENHI (grant JP20H00040). TH was supported by the JSPS KAKENHI (grant JP17H00947, JP18KK0371, and JP24K00176).

Data availability

The datasets generated and/or analysed during the current study are not publicly available due to ethical constraints but anonymised data are available from the corresponding author on reasonable request.

Declarations

Competing interests

The authors declare no competing interests. In particular, none of the authors has a financial interest in the Space Syntax Limited company.

Publisher’s note

Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

References

1. Frank, LD; Iroz-Elardo, N; MacLeod, KE; Hong, A. Pathways from built environment to health: a conceptual framework linking behavior and exposure-based impacts. J. Transp. Health; 2019; 12, pp. 319-335.

2. Riiser, A; Solbraa, A; Jenum, AK; Birkeland, KI; Andersen, LB. Cycling and walking for transport and their associations with diabetes and risk factors for cardiovascular disease. J. Transp. Health; 2018; 11, pp. 193-201.

3. Xu, Z et al. The effect of walking on depressive and anxiety symptoms: systematic review and meta-analysis. JMIR Public. Health Surveillance; 2024; 10, e48355. [PubMed: https://www.ncbi.nlm.nih.gov/pubmed/39045858][PubMedCentral: https://www.ncbi.nlm.nih.gov/pmc/articles/PMC11287235]

4. Zhang, Z; Chen, W. A systematic review of the relationship between physical activity and happiness. J. Happiness Stud.; 2019; 20, pp. 1305-1322.

5. Boakye, K et al. Perceived built environment characteristics associated with walking and cycling across 355 communities in 21 countries. Cities; 2023; 132, 104102.

6. Kaczynski, AT. Neighborhood walkability perceptions: associations with amount of neighborhood-based physical activity by intensity and purpose. J. Phys. Activity Health; 2010; 7, pp. 3-10.

7. De Vos, J; Lättman, K; Van der Vlugt, AL; Welsch, J; Otsuka, N. Determinants and effects of perceived walkability: a literature review, conceptual model and research agenda. Transp. Reviews; 2023; 43, pp. 303-324.

8. Orstad, SL; McDonough, MH; Stapleton, S; Altincekic, C; Troped, PJ. A systematic review of agreement between perceived and objective neighborhood environment measures and associations with physical activity outcomes. Environ. Behav.; 2017; 49, pp. 904-932. [DOI: https://dx.doi.org/10.1177/0013916516670982]

9. Chokshi, DA; Farley, TA. Changing behaviors to prevent noncommunicable diseases. Science; 2014; 345, pp. 1243-1244.2014Sci..345.1243C [PubMed: https://www.ncbi.nlm.nih.gov/pubmed/25214590]

10. Ma, L; Cao, J. How perceptions mediate the effects of the built environment on travel behavior?. Transportation; 2019; 46, pp. 175-197.

11. Orstad, SL et al. Neighborhood walkability and physical activity among older women: tests of mediation by environmental perceptions and moderation by depressive symptoms. Prev. Med.; 2018; 116, pp. 60-67. [PubMed: https://www.ncbi.nlm.nih.gov/pubmed/30092314][PubMedCentral: https://www.ncbi.nlm.nih.gov/pmc/articles/PMC6260982][DOI: https://dx.doi.org/10.1016/j.ypmed.2018.08.008]

12. Aarthi, GR et al. Associations of the built environment with type 2 diabetes in asia: a systematic review. BMJ Open.; 2023; 13, e065431. [PubMed: https://www.ncbi.nlm.nih.gov/pubmed/37015791][PubMedCentral: https://www.ncbi.nlm.nih.gov/pmc/articles/PMC10083821]

13. Valson, JS; Kutty, VR. Gender differences in the relationship between built environment and non-communicable diseases: a systematic review. J. Public Health Res.; 2018; 7, 1239. [PubMed: https://www.ncbi.nlm.nih.gov/pubmed/29780764][PubMedCentral: https://www.ncbi.nlm.nih.gov/pmc/articles/PMC5941255]

14. Núñez, MBF et al. Gender and sex differences in urban greenness’ mental health benefits: a systematic review. Health Place; 2022; 76, 102864.

15. Pollard, TM; Wagnild, JM. Gender differences in walking (for leisure, transport and in total) across adult life: a systematic review. BMC Public. Health; 2017; 17, pp. 1-11.

16. Hanibuchi, T. et al. Mapping Japan’s Major Cities Using a Social Survey (Kokon Shoin, 2022).

17. Hanibuchi, T. et al. A survey dataset on residents’ attitudes and behaviors in major cities in Japan: geo-social survey for urban lifestyle preferences (GULP). Data in Brief. 111859 (2025).

18. Inoue, S et al. Reliability of the abbreviated neighborhood environment walkability scale Japanese version. Japanese J. Phys. Fit. Sports Med.; 2009; 58, pp. 453-461.

19. Inoue, S et al. Association of physical activity and neighborhood environment among Japanese adults. Prev. Med.; 2009; 48, pp. 321-325. [PubMed: https://www.ncbi.nlm.nih.gov/pubmed/19463488]

20. Nakaya, T et al. Associations of all-cause mortality with census-based neighbourhood deprivation and population density in japan: a multilevel survival analysis. PloS One; 2014; 9, e97802.2014PLoSO..997802N [PubMed: https://www.ncbi.nlm.nih.gov/pubmed/24905731][PubMedCentral: https://www.ncbi.nlm.nih.gov/pmc/articles/PMC4048169]

21. Shoari, N; Ezzati, M; Baumgartner, J; Malacarne, D; Fecht, D. Accessibility and allocation of public parks and gardens in England and wales: a COVID-19 social distancing perspective. PloS One; 2020; 15, e0241102. [PubMed: https://www.ncbi.nlm.nih.gov/pubmed/33095838][PubMedCentral: https://www.ncbi.nlm.nih.gov/pmc/articles/PMC7584245]

22. Frank, LD et al. The development of a walkability index: application to the neighborhood quality of life study. Br. J. Sports Med.; 2010; 44, pp. 924-933. [PubMed: https://www.ncbi.nlm.nih.gov/pubmed/19406732]

23. Nakaya, T; Maeda, K; Nagata, S. A note on Building GIS-based walkability index of Japan. Ritsumeikan Bungaku; 2018; 656, pp. 693-707.

24. Tanimoto, R; Hanibuchi, T; Nakaya, T. Development of Japan postcode-level walkability index and examination of its usefulness. Q. J. Geogr.; 2023; 75, pp. 16-26.

25. Hanibuchi, T et al. Does walkable mean sociable? Neighborhood determinants of social capital among older adults in Japan. Health Place; 2012; 18, pp. 229-239. [PubMed: https://www.ncbi.nlm.nih.gov/pubmed/22000345]

26. Koohsari, MJ; Owen, N; Cerin, E; Giles-Corti, B; Sugiyama, T. Walkability and walking for transport: characterizing the built environment using space syntax. Int. J. Behav. Nutr. Phys. Activity; 2016; 13, pp. 1-9.

27. Gil, J., Varoudis, T., Karimi, K. & Penn, A. in SSS 2015-10th International Space Syntax Symposium. 148:1–148:12 (Space Syntax Laboratory, The Bartlett School of Architecture, UCL).

28. Turner, A. Depthmap 4: a researcher’s handbook. (2004).

29. Turner, A. in Proceedings of the 3rd international symposium on space syntax. 30 – 11 (Georgia Institute of Technology Atlanta, GA).

30. Watanabe, R; Hanibuchi, T; Yamamoto, R; Nakaya, T. Historical determinants of street connectivity in Japanese cities: relationship between Japanese castle town formation, war damage, and space syntax integration value. Japan Architectural Rev.; 2023; 6, e12408.

31. Hillier, B. The City as a socio-technical system: a spatial reformulation in the light of the levels problem and the parallel problem. In: Digital Urban Modeling and Simulation (Communications in Computer and Information Science, vol. 242), 24–48 (2012).

32. Koohsari, MJ et al. Street network measures and adults’ walking for transport: application of space syntax. Health Place; 2016; 38, pp. 89-95. [PubMed: https://www.ncbi.nlm.nih.gov/pubmed/26808485]

33. Snijders, T. A. & Bosker, R. Multilevel Analysis: An Introduction to Basic and Advanced Multilevel Modeling. (2011).

34. Wang, ML; Narcisse, MR; McElfish, PA. Higher walkability associated with increased physical activity and reduced obesity among united States adults. Obesity; 2023; 31, pp. 553-564. [PubMed: https://www.ncbi.nlm.nih.gov/pubmed/36504362]

35. Hsieh, HS; Chuang, MT. Association of perceived environment walkability with purposive and discursive walking for urban design strategies. J. Transp. Land. Use; 2021; 14, pp. 1099-1127.

36. Rodrigue, L et al. Factors influencing subjective walkability. J. Transp. Land. Use; 2022; 15, pp. 709-727.

37. Kim, S; Park, S; Lee, JS. Meso-or micro-scale? Environmental factors influencing pedestrian satisfaction. Transp. Res. Part. D: Transp. Environ.; 2014; 30, pp. 10-20.

38. Lee, M; Lee, H; Hwang, S; Choi, M. Understanding the impact of the walking environment on pedestrian perception and comprehension of the situation. J. Transp. Health; 2021; 23, 101267.

39. D’Acci, L. Aesthetical cognitive perceptions of urban street form. Pedestrian preferences towards straight or curvy route shapes. J. Urban Des.; 2019; 24, pp. 896-912.

40. Cerin, E et al. The neighbourhood physical environment and active travel in older adults: a systematic review and meta-analysis. Int. J. Behav. Nutr. Phys. Activity; 2017; 14, pp. 1-23.

41. Abe, T; Dogra, S; Owen, N; Sugiyama, T. Older adults’ staying at home in greater tokyo: association with population density and roles of car ownership and public transport. J. Transp. Health; 2024; 36, 101807.

42. Hinrichs, T et al. Perception of parks and trails as mobility facilitators and transportation walking in older adults: a study using digital geographical maps. Aging Clin. Exp. Res.; 2019; 31, pp. 673-683. [PubMed: https://www.ncbi.nlm.nih.gov/pubmed/30666515]

43. Eom, S; Kim, H; Hasegawa, D; Yamada, I. Pedestrian movement with large-scale GPS records and transit-oriented development attributes. Sustainable Cities Soc.; 2024; 102, 105223.

44. Zhang, CQ; Zhang, R; Baker, JS; Hagger, MS; Hamilton, K. A qualitative investigation exploring neighbourhood environment, risks and fear of falling, and fall prevention strategies among urban-dwelling older adults in a high-density City. Ageing Soc.; 2024; 44, pp. 2155-2176.

45. Xiang, L; Sheng, J; Liao, P. Understanding the relationship between the Spatial configuration and the crime rate of downtown Eastside in vancouver, Canada. Habitat Int.; 2023; 137, 102847.

46. Summers, L; Johnson, SD. Does the configuration of the street network influence where outdoor serious violence takes place? Using space syntax to test crime pattern theory. J. Quant. Criminol.; 2017; 33, pp. 397-420.

47. Iranmanesh, A; Kara, C; Tülbentçi, T. Mapping the relationship between traffic accidents, road network configuration, and urban land use. Int. J. Injury Control Saf. Promotion; 2024; 31, pp. 672-685.

48. Makita, M et al. Place (in) securities: older adults’ perceptions across urban environments in the united Kingdom. PsyEcology; 2020; 11, pp. 214-231.

49. Morioka, W; Kwan, MP; Hino, K; Yamada, I. How accessibility to neighborhood grocery stores is related to older people’s walking behavior: A study of yokohama, Japan. J. Transp. Health; 2023; 32, 101668.

50. Navarrete-Hernandez, P; Vetro, A; Concha, P. Building safer public spaces: exploring gender difference in the perception of safety in public space through urban design interventions. Landsc. Urban Plann.; 2021; 214, 104180.

51. Day, K; Stump, C; Carreon, D. Confrontation and loss of control: masculinity and men’s fear in public space. J. Environ. Psychol.; 2003; 23, pp. 311-322.

52. Zhu, Y; Su, F; Han, X; Fu, Q; Liu, J. Uncovering the drivers of gender inequality in perceptions of safety: an interdisciplinary approach combining street view imagery, socio-economic data and Spatial statistical modelling. Int. J. Appl. Earth Obs. Geoinf.; 2024; 134, 104230.

53. Japanese, M. L. I. T. Urban Walkability Promotion Policy, (2020). https://www.mlit.go.jp/toshi/toshi_gairo_tk_000092.html

54. Delitheou, V; Bakogiannis, E; Kyriakidis, C. Urban planning: integrating smart applications to promote community engagement. Heliyon; 2019; 5, e01672. [PubMed: https://www.ncbi.nlm.nih.gov/pubmed/31193667][PubMedCentral: https://www.ncbi.nlm.nih.gov/pmc/articles/PMC6538962]