Synergistic Effect of Community Environment on Cognitive Function in Elderly People

Abstract

With rapid global aging, the community environment has become a critical factor influencing cognitive health in older adults. However, most existing studies focus on single environmental attributes and rely on linear analytical methods, which fail to capture the complex and synergistic effects of community features. Guided by an integrated theoretical perspective on environmental psychology, aging, and cognitive health, this study examines how multiple community environmental factors jointly affect cognitive function in elderly people. A case study was conducted among 215 older residents in Shanghai, China. An exploratory factor analysis (EFA) identified the following five key dimensions of community environment: pedestrian friendliness, blue–green spaces, infrastructure, space attractiveness, and safety. We then applied both Partial Least Squares Structural Equation Modeling (PLS-SEM) and Fuzzy Set Qualitative Comparative Analysis (fsQCA) to reveal linear and configurational relationships. The findings showed that pedestrian friendliness, blue–green spaces, and space attractiveness significantly enhance cognitive health, while fsQCA highlighted multiple pathways that underscore the non-linear and synergistic interactions among environmental features. These results provide theoretical insights into the mechanisms linking community environments and cognitive function and offer practical guidance for designing age-friendly communities.

1. Introduction

With rapid global aging, community environments play an increasingly critical role in promoting healthy aging. Healthy aging depends not only on individual health conditions and medical services, but also on supportive community environments. Community design significantly impacts older adults’ physical activity, mental health, and social engagement. Well-designed communities enhance elderly health by encouraging daily activities [1], reducing psychological stress [2], and promoting social interactions [3]. Optimizing public spaces and built environments effectively delays aging and supports healthy aging goals.

Cognitive function is essential for maintaining independence and social participation among older adults [4]. With increasing elderly populations, cognitive decline and related dementias have become significant public health challenges, reducing older people’s quality of life and creating social burdens. Although cognitive health is influenced by lifestyle and chronic health conditions, community environments also significantly impact cognitive function. A supportive community environment can foster active lifestyles, social interaction, and mental well-being, helping to delay cognitive decline [3,5,6].

However, existing research has several limitations. First, studies often focus on individual environmental features rather than examining the community environment as an interconnected system. Second, linear analytical methods inadequately address non-linear and asymmetric relationships, such as synergistic or inhibitory effects among environmental features. Lastly, the lack of standardized environmental measurement methods limits comparative research. Therefore, a systematic analysis exploring complex environmental configurations is necessary.

Given these gaps, this study aims to address the following research questions: (1) How do multiple community environmental factors collectively affect cognitive function among older adults? (2) What are the key pathways or configurations that explain these relationships? This study addresses these gaps by systematically extracting key environmental features from the literature and constructing a measurement framework using exploratory factor analysis (EFA). Partial Least Squares Structural Equation Modeling (PLS-SEM) is applied to examine direct effects on cognitive function, while Fuzzy Set Qualitative Comparative Analysis (fsQCA) is used to uncover complex, non-linear relationships. This integrated approach provides a comprehensive understanding of how interconnected community environments influence older adults’ cognitive health.

The remainder of this paper is organized as follows: Section 2 reviews the theoretical background and develops the research hypotheses. Section 3 presents the data, case study, and research methods. Section 4 reports the results from both the PLS-SEM and fsQCA analyses. Section 5 discusses the key findings, theoretical contributions, and practical implications. Finally, Section 6 concludes the study and outlines future research directions.

2. Theoretical Background and Hypotheses

2.1. Community Environment and Cognitive Function in Elderly People

The physical environments of communities significantly influence older adults’ cognitive health. According to the General Ecological Model, human behavior and functioning result from dynamic interactions between individuals and their environments [7]. As physical and cognitive capacities decline with aging, environmental factors exert increasing impacts on daily behavior and functional autonomy, either facilitating or hindering adaptive capacities. Previous research broadly conceptualizes these environmental influences through the following two interrelated pathways: the Support–Insufficiency Pathway and the Adaptation–Risk Pathway.

In the Support–Insufficiency Pathway, communities rich in physical infrastructure, public services, and social opportunities enable frequent social interaction and cognitive engagement among older adults. Such environmental stimulation contributes to building cognitive reserves, which protect the brain from age-related cognitive deterioration by maintaining cognitive activity and social connectedness [8]. Conversely, insufficient community resources can reduce opportunities for cognitive stimulation and social engagement, indirectly contributing to cognitive decline through increased isolation, loneliness, and diminished emotional support.

The Adaptation–Risk Pathway focuses on how the community environment either supports or impedes older adults’ adaptation to age-related functional declines. Safe, accessible, and aesthetically appealing environments encourage older adults to engage in physical and social activities, mitigating the impact of physical and cognitive impairments. Natural environments (blue–green spaces), according to Attention Restoration Theory (ART) and Stress Reduction Theory (SRT), promote cognitive restoration and stress reduction. In contrast, poorly designed environments characterized by physical disorder or perceived threats increase stress and inhibit outdoor activity, negatively affecting cognitive function through chronic stress mechanisms.

Overall, the existing literature suggests that community environments significantly influence cognitive function among older adults through multiple complex pathways. Although prior studies have extensively investigated individual environmental features, few have systematically examined the community environment as a holistic, interconnected system. Therefore, this study systematically identifies and categorizes key environmental features derived from prior research (

Table 1. Community environmental factors affecting cognitive function in elderly people.

No.	Factor	Definition	Reference
1	Residential density	The concentration of housing units within a specific area.	[9,10]
2	Street connectivity	How the street network in an area is interconnected.	[5,9]
3	Land use mix	The diversity of land use in a defined area.	[9,11,12,13]
4	Proximity to destination	How conveniently residents can walk to various common locations within the community on foot.	[14,15]
5	Sidewalk coverage	The existence of paths for walking in the participant’s neighborhood.	[16,17]
6	Cleanliness	The level of cleanliness and hygiene in the community.	[18,19]
7	Public space maintenance	The state and quality of upkeep for publicly accessible areas, ensuring spaces remain usable and safe for the community.	[16]
8	Aesthetic appeal	The visual pleasure and appeal that community spaces and elements bring through their design, landscaping, and decoration.	[9,20,21]
9	Defensible space	Signs that distinguish between public and private domains.	[22]
10	Climate change	Extreme meteorological phenomena that lead to changes in the living environment and affect residents’ daily activities and comfort, such as heat waves, droughts, and heavy rains.	[23]
11	Infrastructure/service availability	The presence and variety of essential and non-essential services within a neighborhood or community.	[10,16,24,25]
12	Infrastructure/service accessibility	The ease of access to services and facilities within the community.	[26]
13	Blue space	Natural or man-made water features in the community, such as rivers, lakes, ponds, and fountains.	[5,12,27]
14	Green space	Areas covered by vegetation, such as parks, gardens, and forests.	[5,12,27,28,29,30]
15	Traffic safety	The safety of roads and infrastructure in the community, aimed at reducing accidents and protecting all road users.	[9]
16	Crime safety	To feel and to be safe, particularly with regard to threats of crime and at night.	[21]

) to comprehensively examine their configurational effects.

2.2. Conceptualization and Factor Extraction

Based on a comprehensive literature review, we identified 16 community environmental factors. To conceptually group these factors into higher-order dimensions, we employed an exploratory factor analysis (EFA) as a design step. Rather than presenting empirical results, this step aimed to refine the theoretical framework and guide the development of hypotheses.

Through this factor extraction process, we determined that a five-factor structure best represented the interrelationships among the 16 items. The analysis was based on data from 170 elderly community residents. The Kaiser–Meyer–Olkin (KMO) value was 0.77 (>0.6), and Bartlett’s test of sphericity was significant (p < 0.001), indicating that the data were suitable for conceptual factor grouping [31]. Five factors with eigenvalues greater than one were extracted, explaining 72.85% of the total variance after rotation. Specifically, these five factors accounted for 25.06%, 19.01%, 10.31%, 9.58%, and 9.31% of the variance, respectively. This conceptual grouping suggests that the extracted factors capture a substantial proportion of the variance in the data, supporting the validity of the underlying structure.

Subsequently, a factor loading criterion was applied to refine the factor structure. Items with factor loadings below 0.5 or communalities below 0.4 were removed [32]. Specifically, the factors “Defensible Space” (factor 9) and “Climate Change” (factor 10) were excluded because all of their items failed to meet these thresholds. The remaining items were then grouped into their respective factors based on their loadings, ensuring that each item adequately represented its corresponding latent variable. This approach streamlined the factor structure while retaining the most reliable items. The final factor structure was confirmed after this refinement, providing a solid foundation for further analysis and interpretation.

Table 2. Factor loading and factor component from EFA.

Factors	Factor Loading					Commonality
	Factor 1	Factor 2	Factor 3	Factor 4	Factor 5
1	0.894					0.828
2	0.886					0.791
3	0.869					0.77
4	0.78					0.628
5	0.875					0.772
6		0.925				0.867
7		0.939				0.889
8		0.915				0.844
11			0.886			0.799
12			0.877			0.785
13				0.861		0.745
14				0.795		0.666
15					0.756	0.687
16					0.609	0.685

presents the detailed classification of each factor, including the rotated factor matrix, while

Table 3. Latent and observed variables.

Latent Variable	Observed Variable
Pedestrian Friendliness	Residential density
	Street connectivity
	Land use mix
	Proximity to destination
	Sidewalk coverage
Space Attractiveness	Cleanliness
	Public space maintenance
	Aesthetic appeal
Infrastructure/Service	Infrastructure/service availability
	Infrastructure/service accessibility
Blue–Green Space	Blue space
	Green space
Safety	Traffic safety
	Crime safety

shows the structure of latent and observed variables.

2.3. Hypotheses Development

Building upon the conceptual framework outlined in Section 2.2, we now develop hypotheses regarding how community environment dimensions influence cognitive function among older adults. Previous studies have provided valuable insights into how specific community environmental features—such as walkability, green spaces, and infrastructure—affect the cognitive and mental health of older adults [9,21,33]. These studies form a solid foundation for our work, as they highlight the importance of community-level factors in shaping cognitive outcomes.

However, most prior research has examined such features individually and predominantly used linear analytical approaches, which may not fully capture the complex, synergistic relationships among environmental factors. Building on this literature, our study adopts an integrative approach by using PLS-SEM to assess direct effects and fsQCA to explore non-linear, configurational pathways.

Guided by this perspective, and to address our research questions, we conceptualize five higher-order dimensions of community environment—Pedestrian Friendliness, Space Attractiveness, Infrastructure/Service, Blue–Green Space, and Safety—and propose the following hypotheses.

2.3.1. Pedestrian Friendliness and Cognitive Function in Older Adults

Within the ecological model of aging, cognitive and functional outcomes in later life emerge from the fit (or misfit) between individual competencies and environmental press. Neighborhoods that facilitate safe, frequent, and self-determined mobility help older adults preserve their autonomy, engage in social participation, and maintain cognitively stimulating routines [7,34,35]. Pedestrian-friendly environments, characterized by dense, connected street networks, adequate sidewalks, short distances to destinations, and barrier-free crossings, lower the “activation energy” required for daily out-of-home activity. Such environments support cognitive function through multiple reinforcing pathways, as follows: (i) biological—walking and outdoor activity enhance cerebral blood flow and upregulate neurotrophic factors such as BDNF, which promote neuroplasticity; (ii) behavioral—sustained mobility and social interaction expand cognitive reserves by engaging mental and social faculties; and (iii) psychosocial—exposure to pleasant, walkable surroundings reduces allostatic load by alleviating anxiety and depressive symptoms that drain cognitive resources [36,37,38,39].

A considerable body of research has highlighted the positive association between walkability and various cognitive and mental health indicators among older populations. Many studies point out that neighborhoods with well-connected pedestrian infrastructure and safe crossings encourage more frequent walking, which, in turn, leads to greater opportunities for social interaction and outdoor activity. This is consistent with findings that sustained mobility helps older adults preserve their independence and cognitive resilience. However, existing research also shows variations in how pedestrian environments are measured and perceived: while some focus on objective indices such as street connectivity and block length, others emphasize subjective perceptions of safety and comfort, both of which may influence walking behavior differently [40,41]. Additionally, certain studies underscore the role of micro-scale environmental features, such as greenery along sidewalks and the aesthetics of streetscapes, which can further enhance walking experiences and cognitive benefits. These insights suggest that Pedestrian Friendliness is a multidimensional construct, involving not only physical infrastructure, but also the perceived quality and comfort of walking environments.

Hypothesis 1.

Pedestrian Friendliness positively contributes to cognitive function among older residents.

2.3.2. Space Attractiveness and Cognitive Function in Older Adults

Space attractiveness refers to the aesthetic and perceptual qualities of the built environment, including visual appeal, architectural harmony, cleanliness, cultural elements, and the perceived pleasantness of public spaces. According to Attention Restoration Theory (ART) [42], aesthetically rich environments evoke “soft fascination”, allowing cognitive resources to recover from mental fatigue. Stress Reduction Theory (SRT) [43] similarly suggests that visually appealing and orderly surroundings elicit positive emotional responses and reduce physiological stress, which, in turn, supports cognitive functioning. Attractive public spaces also encourage longer voluntary outdoor stays and unplanned social encounters, both of which provide cognitive stimulation and enhance cognitive reserves [8].

A growing number of studies have indicated that the perceived attractiveness of neighborhoods is associated with better mental health and cognitive functioning among older adults. Research highlights that environments with aesthetic appeal—such as well-maintained streets, cultural landmarks, and visually pleasant green areas—can foster a positive sense of place and support emotional stability, indirectly benefiting cognition [44,45,46]. At the same time, the existing literature shows that space attractiveness is not a uniform concept: while some studies emphasize visual and design-related factors, others point to cultural elements or the sense of orderliness as key components of attractiveness [47]. Furthermore, there is evidence that the cognitive benefits of aesthetics may be mediated by behaviors such as walking or outdoor socializing, suggesting that attractiveness might work in synergy with other environmental features like pedestrian infrastructure and safety. Overall, prior work underscores that both the physical and perceived dimensions of attractiveness play an important role in shaping the cognitive outcomes of older adults.

Hypothesis 2.

Space Attractiveness positively contributes to cognitive function among older residents.

2.3.3. Infrastructure/Service and Cognitive Function in Older Adults

Community infrastructure and services provide the structural and social scaffolding that supports daily living and active engagement among older adults. Well-developed infrastructure lowers environmental press, enabling older individuals to maintain their autonomy and perform cognitively stimulating activities. Access to facilities such as libraries, parks, recreational centers, and healthcare services not only meets basic needs, but also promotes social interaction and lifelong learning, which are central to the cognitive reserve framework [8,48]. Cognitive reserve suggests that consistent engagement in intellectual and social activities strengthens neural pathways and delays cognitive decline. Moreover, services that provide emotional support or health guidance can mitigate stress and negative affect, indirectly enhancing cognitive function [35].

Research has consistently shown that neighborhoods with abundant and accessible infrastructure are linked to higher levels of physical activity, social interaction, and mental stimulation among older adults. For example, studies emphasize that proximity to public transport, community centers, and health services can improve mobility, reduce social isolation, and promote routine activities that are beneficial to cognitive functioning [49]. At the same time, some findings suggest that the mere presence of facilities may not be sufficient—perceived accessibility, quality of service, and cultural relevance are equally important for encouraging active engagement. Comparative evidence also shows that infrastructure in dense urban settings tends to offer more opportunities for incidental interactions, while suburban or rural areas often rely on fewer but more community-focused services. Collectively, these insights highlight that the cognitive benefits of infrastructure arise not only from physical availability, but also from how well these services align with the needs, preferences, and routines of older adults.

Hypothesis 3:

Infrastructure/Service positively contributes to cognitive function among older residents.

2.3.4. Blue–Green Space and Cognitive Function in Older Adults

Blue–green spaces—comprising vegetated areas, parks, water features, and tree canopies—contribute to both environmental quality and human health through multiple mechanisms. Humans have an innate affinity for nature, and exposure to natural elements can enhance psychological well-being and cognitive restoration. Urban ecology research also suggests that natural landscapes reduce exposure to environmental stressors such as heat, noise, and air pollution, which are known to negatively affect brain health and cognitive performance [34]. Moreover, the presence of blue–green spaces promotes outdoor activities and intergenerational social interactions, creating enriched environments that stimulate sensory and cognitive engagement [42,49].

Empirical evidence links regular contact with blue–green spaces to improved memory, attention, and emotional regulation in older adults. Studies have reported that individuals living near high-quality natural environments exhibit slower cognitive decline and better mental health compared with those in less vegetated areas [6,50]. However, findings differ regarding the relative importance of proximity versus quality: while some research highlights the amount of nearby green space, other studies emphasize the diversity, aesthetic value, and maintenance of these areas as more predictive of health outcomes [51]. Additionally, cultural and regional factors may shape the use and perception of blue–green spaces, suggesting that the benefits of nature exposure are not uniform across populations [52]. These studies collectively underscore the role of both physical characteristics and subjective experiences of natural environments in supporting cognitive function.

Hypothesis 4.

Blue–Green Space positively contributes to cognitive function among older residents.

2.3.5. Safety and Cognitive Function in Older Adults

Perceived and actual safety within a community are critical determinants of older adults’ well-being and cognitive health. Chronic exposure to unsafe or unpredictable environments increases psychological stress, which, over time, can impair attention, memory, and other cognitive functions. Safety concerns—stemming from crime, traffic hazards, and poorly maintained public spaces—can limit outdoor mobility and social participation, both of which are key for cognitive engagement [50]. Additionally, neighborhoods with lower safety often exhibit weaker social cohesion and fewer communal activities, reducing opportunities for mental stimulation and support networks that protect against cognitive decline [6].

Research has shown that older adults who perceive their neighborhoods as safe are more likely to engage in outdoor physical activities, interact with neighbors, and maintain higher levels of independence—all factors associated with better cognitive outcomes. Conversely, studies report that fear of crime and traffic hazards can lead to social withdrawal and reduced physical activity, contributing to cognitive deterioration [50,52,53]. However, findings vary on whether perceived safety or objective safety metrics (e.g., crime rates) are more predictive of cognitive health, with some evidence suggesting that subjective perceptions may exert a stronger influence on behavior [54]. Moreover, certain studies indicate that safety interacts with other environmental factors—such as walkability and infrastructure quality—where a safe environment can amplify the positive effects of these features [55]. This highlights the multifaceted role of safety as both a direct and indirect determinant of cognitive well-being.

Hypothesis 5.

Safety positively contributes to cognitive function among older residents.

3. Method

3.1. Study Area

This study was conducted in Shanghai, one of the largest metropolitan cities in China and a region undergoing rapid population aging. According to the 2022 Shanghai Statistical Yearbook and the 2022 Report on the Aging Population and Elderly Care Development of Shanghai issued by the Shanghai Municipal Government, residents aged 60 and above accounted for approximately 36.8% of the registered population by the end of 2022. This significant proportion of older adults underscores the importance of understanding how community environments affect cognitive health in this context.

To capture the diversity of community environments, seven districts with the largest permanent populations were selected, representing both dense urban centers and suburban residential areas. These districts vary widely in population density, infrastructure, green space availability, and public service accessibility, providing a rich context for exploring the relationships between environmental factors and cognitive function in older adults. Additionally, Shanghai has actively promoted age-friendly initiatives—such as upgrading public spaces, improving walkability, and enhancing community services—which further supports its relevance as a case study area for this research.

3.2. Sample and Data

The study targeted elderly residents aged 60 and above in the selected districts to explore the relationship between community environment and cognitive function. The research team collaborated with local community offices and resident committees to recruit participants through community announcements and direct outreach initiatives. All participants were required to be permanent residents of the selected communities and at least 60 years old.

Over a period of two months, we collected 215 valid questionnaires, with a mean age of 70.66 years (SD = 6.66) and a median age of 72.5 years. The sample consisted of 53% male and 47% female participants, ensuring a relatively balanced gender representation.

The study followed ethical guidelines approved by the Ethics Committee. In compliance with ethical standards, all participants were provided with detailed informed consent forms explaining the purpose, procedures, potential risks, and benefits of the study. They were assured of the confidentiality and anonymity of their responses, with all data used exclusively for research purposes. Data collection employed a structured questionnaire that included items measuring various aspects of the community environment and cognitive function.

3.3. Measures

The measurements of community environment elements and cognitive function among older adults were based on established scales or validated research instruments. Specifically, Pedestrian Friendliness and Safety were assessed using the Chinese version of the Neighborhood Environment Walkability Scale—Abbreviated (NEWS-A) [56]. Space Attractiveness was measured using the perceived built environment scale [57]. Infrastructure/Service was assessed based on the work of Zhang et al. [10] and Li et al. [58]. Blue–Green Space was measured by incorporating subjective perception scales for blue space [59] and green space [60]. Respondents rated these community environment elements using a 5-point Likert scale, ranging from 1 (strongly disagree) to 5 (strongly agree).

Cognitive function among older adults was evaluated using the Cognitive Function Instrument (CFI) [61,62], which includes assessments of memory decline, cognitive difficulties, and functional abilities. Response options for the CFI include Yes, No, or Maybe, with corresponding scores of Yes = 1, No = 0, and Maybe = 0.5, resulting in a total score ranging from 0 to 14. The specific questionnaire items are detailed in the supplementary material.

It is important to note that the CFI scale assesses the extent of cognitive impairment, whereas this study hypothesizes a positive influence of community environment factors on cognitive function among older adults. To mitigate measurement bias due to data inversion, we used the reverse conceptualization of the dependent variable in our analyses. Consequently, in subsequent symmetric analyses, a negative correlation should be interpreted, with attention paid to the absolute value of the path coefficient. Furthermore, in the interpretation of asymmetric analyses, it is essential to emphasize the conceptual inversion between the outcome variable (CF) and the non-outcome variable (~CF).

3.4. Symmetric and Asymmetric Methods

This study employed a combination of symmetric and asymmetric statistical analysis methods to investigate the effects of five latent variables related to the community environment on the cognitive function of elderly people. First, Partial Least Squares Structural Equation Modeling (PLS-SEM) was applied for analysis. PLS-SEM is particularly suitable for models with complex causal relationships, especially when the sample size is relatively limited and the model involves multiple causal pathways. It allows for the simultaneous estimation of relationships between multiple independent and dependent variables while also addressing the estimation of paths between latent and observed variables [63]. Data analysis was performed using SmartPLS v4.1.1.4 software, where standardized path coefficients were obtained through the path-weighting algorithm and the statistical significance of the paths was evaluated using 5000 bootstrap resamples [64]. The strength of this method lies in its ability to reveal the direct effects of various dimensions of the community environment on cognitive function in elderly people, while also providing insights into the predictive capabilities of the model and the causal relationships among the variables. Note that the cognitive function (CF) variable was reverse-scored, meaning that higher values indicate a poorer cognitive performance. Therefore, negative path coefficients should be interpreted as positive effects on cognitive function.

To further deepen our understanding of the complex configurations of community environmental factors and their impact on elderly cognitive function, Fuzzy Set Qualitative Comparative Analysis (fsQCA) was used to supplement the shortcomings of PLS-SEM. Grounded in configuration theory, fsQCA is well-suited for exploring non-linear and complex relationships by uncovering different combinations of conditions that influence the outcome variable. Unlike traditional regression models, fsQCA identifies multiple potential causal pathways and expresses the relationships between independent and dependent variables in fuzzy set terms, capturing the existence of multiple realities in complex systems [65]. Through fsQCA, this study identified various configurations of community environmental factors that contribute to positive or negative cognitive outcomes for elderly people, highlighting the necessary and sufficient conditions that influence cognitive health. The flexibility of this approach allows for the discovery of key factors and causal pathways in complex social environments, providing valuable theoretical insights for future policy development.

In comparison with prior studies in this field [6,12], which largely relied on regression- or covariance-based SEM approaches, our study’s integration of PLS-SEM and fsQCA represents both a formal and theoretical advancement. While previous research has provided valuable evidence on the linear effects of individual environmental factors, it often does not account for the configurational complexity inherent in community environments. By employing fsQCA, we were able to identify multiple combinations of factors that can yield high cognitive function outcomes. For example, we found that high Pedestrian Friendliness and Blue–Green Space can compensate for moderate Infrastructure Services, forming an alternative pathway to strong cognitive health. This finding illustrates how different configurations can lead to similar outcomes, an insight aligned with complexity and ecological aging theories, but less emphasized in earlier studies. Thus, our dual-method approach extends existing research by capturing both independent effects and synergistic configurations.

4. Symmetrical Analysis

4.1. Measurement Model

In this study, we evaluated the measurement model by assessing its reliability, convergent validity, and discriminant validity [66]. To ensure internal consistency reliability, we used Cronbach’s alpha and composite reliability (CR). As shown in

Table 4. Convergent and discriminant validity.

Variables	Structural Reliability			Fornell–Larcker Criterion
	α	CR	AVE	BS	SA	IS	CF	S	PF
BS	0.845	0.847	0.866	0.931
SA	0.908	0.908	0.845	0.877	0.919
IS	0.867	0.868	0.883	0.83	0.882	0.94
CF	0.862	0.884	0.783	0.833	0.849	0.826	0.885
S	0.864	0.864	0.88	0.836	0.851	0.842	0.8	0.938
PF	0.938	0.938	0.8	0.89	0.914	0.894	0.865	0.861	0.895

Note: The square root of the AVE is shown in bold diagonal lines. BS = Blue–Green Space, SA = Space Attractiveness, IS = Infrastructure/Service, CF = Cognitive Function, S = Safety, PF = Pedestrian Friendliness.

, these two values were both greater than 0.8 (>0.7), indicating a high reliability of the measurement model. Convergent validity was assessed with the average variance extracted (AVE), and all AVE values exceeded the recommended threshold of 0.5 (Table 4), demonstrating that a significant portion of the variance in the constructs was captured by their respective indicators. Furthermore, the factor loadings for all items were above 0.883 (

Table 5. Factor loading.

	BS	SA	IS	CF	S	PF
BS1	0.934
BS2	0.927
SA1		0.918
SA2		0.911
SA3		0.928
IS1			0.938
IS2			0.941
CF1				0.928
CF2				0.832
CF3				0.891
S1					0.937
S2					0.939
PF1						0.908
PF2						0.893
PF3						0.898
PF4						0.891
PF5						0.883

), indicating that the individual items had strong relationships with their respective constructs. Thus, the model exhibited a good convergent validity [67].

The Fornell–Larcker criterion was employed to establish discriminant validity, which compares the square root of the AVE for each construct with the correlations between constructs [68]. The results confirmed that the square root of the AVE for each construct was greater than its correlations with other constructs, thereby satisfying the discriminant validity requirements (Table 4). Further examination showed that each indicator’s outer loading on its assigned construct was greater than its cross-loadings with other constructs [69]. Moreover, we checked for multicollinearity issues by applying the variance inflation factor (VIF), and all VIF values were below the commonly accepted threshold of 5, with the highest VIF value being 3.67, indicating no multicollinearity concerns in the model [70].

4.2. Structural Model

Using the bootstrapping method with 5000 subsamples [71], we assessed the significance of the path coefficients. Since CF is reverse-scored, negative path coefficients indicate that the predictor improves cognitive function. The results, shown in

Table 6. Path coefficients.

Path	β	T Value	p Value	97.5% Confidence Intervals
PF → CF	−0.318	3.205	0.001	[−0.508, −0.122]
BS → CF	−0.198	2.487	0.013	[−0.349, −0.037]
SA → CF	−0.199	2.128	0.033	[−0.385, −0.022]
IS → CF	−0.142	1.782	0.075	[−0.296, 0.014]
S → CF	−0.072	0.977	0.329	[−0.217, 0.068]

Note: PF = Pedestrian Friendliness, BS = Blue–Green Space, SA = Space Attractiveness, IS = Infrastructure/Service, S = Safety, CF = Cognitive Function.

, indicate that H1 (β = −0.318; t = 3.205; p < 0.01), H2 (β = −0.199; t = 2.128; p < 0.05), and H4 (β = 0.198; t = 2.487; p < 0.05) are supported. However, H3 and H5 are not supported (p > 0.05 and t < 1.96), indicating that they are not statistically significant in this research. Moreover, the 97.5% confidence intervals (CIs) for the path coefficients reveal that these two paths’ CI include zero, further indicating their lack of statistical significance.

Beyond the path significance tests, we evaluated the overall quality of the structural model using key fit and predictive indicators (

Table 7. Structural model quality and predictive relevance.

Indicator	Value
R2 (CF)	0.778
SRMR	0.051
NFI	0.869
f2 (PF → CF)	0.050
f2 (EQ → CF)	0.023
f2 (IS → CF)	0.015
f2 (BS → CF)	0.032
f2 (S → CF)	0.005
Q2_predict (R1)	0.615
Q2_predict (R2)	0.378
Q2_predict (R3)	0.770
R2 (CF)	0.778
SRMR	0.051
NFI	0.869

). The model exhibits a good fit, with SRMR = 0.051 (<0.08) and NFI = 0.869 (>0.80). The R² of CF is 0.778 (>0.75), demonstrating substantial explanatory power. The f² effect sizes of the predictors range from negligible to small (e.g., PF→CF = 0.050, BS→CF = 0.032), indicating varying levels of individual contribution. Furthermore, PLSpredict analysis yielded Q²_predict values of 0.615 (R1), 0.378 (R2), and 0.770 (R3), all of which are above zero, confirming the strong predictive relevance of the structural model [63,72].

Overall, the structural model results confirm that most of the hypothesized relationships between the community environment factors and the cognitive function of elderly people were significant (Figure 1). The bootstrap analysis provides robust support for these relationships. However, the non-significant paths suggest that the relationship between that specific community factor and cognitive function may be weaker or more complex. This does not imply that Infrastructure/Service and Safety are unimportant for the cognitive function of elderly people; rather, this non-significance may be due to the existence of complex configurational relationships between the factors. The lack of statistical significance in the current model may be attributed to these complex relationships, which cannot be fully captured by symmetrical methods such as PLS-SEM alone, requiring further analysis through fsQCA [73].

5. Asymmetrical Analysis

5.1. Necessity Analysis

The PLS-SEM analysis revealed two non-significant paths. To further explore the complex effects of community environmental factors on the cognitive function of elderly people, Fuzzy Set Qualitative Comparative Analysis (fsQCA) was employed as a complementary method. While PLS-SEM is effective in identifying linear relationships between variables, fsQCA offers a deeper perspective on multifaceted, causal pathways, particularly in scenarios involving multiple interacting conditions. Through fsQCA, we can uncover how various factors, in different combinations, collectively influence cognitive function in elderly people, thus providing a more comprehensive explanation of the complex phenomena that PLS-SEM alone may not fully capture.

The primary objective of necessity analysis is to identify whether certain conditions are prerequisites for the outcome variable. The criterion for a necessary condition is its consistency, with a value exceeding 0.9 generally considered indicative of necessity [74]. In this study, fuzzy-set calibration of the independent variables was performed using anchor points of 0.95, 0.5, and 0.05 [75], and an adjustment of +0.001 was applied to avoid fuzzy threshold data points [76,77]. In the calibrated necessity analysis, all consistency values were below 0.9, indicating that no single condition is necessary for cognitive function (

Table 8. Necessity analysis.

Conditional Variable	High CF		Low CF
	Consistency	Coverage	Consistency	Coverage
BS	0.821129	0.834506	0.524279	0.449105
~BS	0.457935	0.533157	0.806804	0.791747
S	0.834826	0.817793	0.527635	0.435661
~S	0.423906	0.515666	0.779327	0.799072
PF	0.81302	0.852479	0.523771	0.462904
~PF	0.487764	0.548561	0.833082	0.789714
SA	0.763665	0.870943	0.509432	0.489711
~SA	0.552565	0.57198	0.865745	0.755361
IS	0.81518	0.823725	0.519805	0.442727
~IS	0.448506	0.525641	0.793035	0.783394

Note: PF = Pedestrian Friendliness, BS = Blue–Green Space, SA = Space Attractiveness, IS = Infrastructure/Service, S = Safety, CF = Cognitive Function.

). Therefore, further configurational analysis is required.

5.2. Configuration Analysis

To examine how the configurations of community environmental factors influence cognitive function in elderly people, a truth table was constructed based on a raw consistency threshold of 0.8, a case frequency of 3, and a PRI threshold of 0.75 [76]. This process generated complex, parsimonious, and intermediate solutions. The complex solution included all possible condition combinations, while the parsimonious solution retained only the core conditions, and the intermediate solution was simplified based on easy counterfactuals. A configurational path table was created from the parsimonious and intermediate solutions, highlighting four paths leading to high cognitive function and one path leading to low cognitive function (

Table 9. Configuration paths.

Configuration	Solution
	High Cognitive Function				Low Cognitive Function
	1	2	3	4	1
Pedestrian Friendliness					⊗
Space Attractiveness					⊗
Infrastructure/Service					⊗
Blue–Green Space					⊗
Safety					⊗
Consistency	0.913672	0.931653	0.93333	0.935141	0.954984
Raw coverage	0.672224	0.663412	0.664046	0.672549	0.61139
Unique coverage	0.041932	0.033121	0.033755	0.042258	0.61139
Overall solution consistency	0.886268				0.954984
Overall solution coverage	0.781357				0.61139

Note: indicates the presence of a condition, ⊗ indicates the absence of a condition. Large icons represent core conditions, while small icons represent peripheral conditions.

).

The following three indicators are used to assess the importance of the paths: consistency, which measures the degree to which the path aligns with the outcome; coverage, which evaluates the extent to which the path explains the cases; and unique coverage, which reflects the uniqueness of the path. As shown in Table 9, the consistency of all five configurations exceeds 0.9 (>0.8), and their coverage is above 0.6 (>0.5), indicating an acceptable reliability and explanatory power [78]. Robustness tests were conducted by adjusting the frequency threshold from 3 to 4 and the PRI threshold from 0.7 to 0.9. Since the paths remained unchanged, the results are considered robust [76].

6. Discussion

6.1. Findings and Theoretical Implications

This study integrated PLS-SEM and fsQCA to examine how community environmental factors influence cognitive function in older adults. Through this dual-method approach, the study identified significant direct effects and complex configurational relationships among environmental factors.

In the PLS-SEM analysis, the following three paths reached significance: Pedestrian Friendliness (PF; β = −0.318, p < 0.01), Space Attractiveness (SA; β = −0.199, p < 0.05), and Blue–Green Space (BS; β = −0.198, p < 0.05). These findings align with expectations. PF had the strongest impact, highlighting the critical role of pedestrian-friendly environments in supporting cognitive function through increased physical activity, enhanced social interactions, diverse sensory stimulation, and improved mental health. Regular walking activities promoted by pedestrian-friendly designs help maintain cognitive health by improving cerebral blood flow, stimulating neurotrophic factors, and increasing neuroplasticity [79,80,81]. Additionally, such environments facilitate social interaction, autonomy, self-efficacy, and psychological well-being, reducing the depression and anxiety risks associated with cognitive decline [46,82,83].

SA and BS also demonstrated significant effects of a similar magnitude, attributable to their roles in emotion regulation, cognitive stimulation, physical activity promotion, and social engagement. Attractive and natural environments alleviate anxiety and depression by reducing environmental stressors such as noise and clutter [84,85]. According to Attention Restoration Theory, natural environments restore cognitive resources and improve concentration [86,87]. Furthermore, natural elements like parks and gardens decrease cortisol levels, mitigating depression and anxiety—key risk factors for cognitive impairment [88,89]. Visually appealing environments stimulate cognitive processes through diverse sensory inputs and motivate older adults to participate in outdoor activities, enhancing executive function and memory through increased physical activity (Erickson et al., 2011 [79]; Warburton et al., 2006 [90]). Additionally, Blue–Green Spaces serve as social gathering points, fostering interactions that provide emotional support and cognitive stimulation, thus protecting cognitive health [46,82].

Infrastructure/Service (IS: β = −0.142, p > 0.05) and Safety (S: β = −0.072, p > 0.05) were not statistically significant in the linear analysis, but this does not imply irrelevance. These factors may exert indirect effects on cognitive health through complex pathways not captured by linear models. For example, convenient infrastructure and improved safety could indirectly support cognitive function by enhancing social participation, psychological security, and quality of life [91,92,93].

The fsQCA analysis (Table 9) identified four distinct configurations leading to high cognitive function and one configuration leading to low cognitive function.

Configuration 1 (IS • BS • S) shows that when Infrastructure/Service, Blue–Green Space, and Safety co-exist, older adults achieve high cognitive function even without strongly pronounced Pedestrian Friendliness or Space Attractiveness. This indicates that a combination of functional accessibility, natural resources, and perceived safety can compensate for lower walkability or visual appeal, supporting cognitive health through social interaction and psychological security.

Configuration 2 (PF • BS) demonstrates that the combination of Pedestrian Friendliness and Blue–Green space alone can suffice for high cognitive outcomes, confirming the powerful and robust effects of these two elements.

Configuration 3 (PF • IS • BS) suggests that Infrastructure can enhance or amplify the benefits of pedestrian-friendly and green environments by improving access to walking routes and activity spaces, thereby increasing opportunities for physical and social engagement.

Configuration 4 (PF • BS • S) highlights that adding Safety to the core PF–BS combination strengthens psychological well-being and encourages outdoor activities, offering a secure and supportive environment for cognitive resilience.

In contrast, the low CF configuration (~PF • ~SA • ~IS • ~BS • ~S) shows that the combined absence of all five environmental features leads to poor cognitive outcomes, confirming a cumulative disadvantage effect where multiple adverse conditions jointly exacerbate cognitive decline.

Overall, these findings reveal that community environmental factors do not act in isolation but form synergistic configurations, aligning with the principles of equifinality and causal complexity in the Social Ecological Model [80]. While IS and S were not significant in the PLS-SEM analysis, their presence in certain fsQCA configurations underlines their context-dependent importance, complementing the linear results and enriching our theoretical understanding of environmental influences on cognitive function [94].

Notably, SA, despite its significant direct effect in PLS-SEM, did not emerge as a sufficient condition in the fsQCA pathways. This finding suggests that its role might be overshadowed by more critical factors like PF and BS when considering configurational effects. Thus, cognitive function can remain high even in environments lacking SA if other essential conditions are satisfied. Moreover, the fsQCA pathway leading to low cognitive function involved the absence of all five environmental factors, confirming a cumulative disadvantage effect where the aggregation of adverse factors contributes to significantly worse cognitive outcomes [95].

In summary, combining PLS-SEM and fsQCA provided complementary insights into how community environments affect cognitive function, identifying both direct and complex configurational effects. This integrative approach offers a nuanced understanding and robust evidence to guide community planning, public policy, and the development of age-friendly environments aimed at promoting cognitive health among older adults.

6.2. Practical Implications

The findings of this study have policy implications for community planning aimed at promoting healthy aging. Different configurations of community environmental factors exhibit distinct impacts on the cognitive health of older adults, and the configuration and quality of the environment have subtle moderating effects on health outcomes. The study suggests the following priorities for optimizing community environments:

(1)

Pedestrian Friendliness and Blue–Green Space showed significant independent effects on promoting cognitive health in older adults and emerged as core conditions in the configurational analysis. Policy efforts should prioritize these aspects, particularly when resources are limited. Improvements should focus on enhancing walkability and accessibility within communities to encourage physical activity, as well as expanding the coverage and accessibility of blue–green spaces to ensure easy access to these beneficial resources. It is also essential to maintain the quality of blue–green spaces to meet the diverse needs of older adults.

(2)

The Configurational Effects of Community Environment Factors indicate that combinations of factors may have greater impacts on cognitive function than any single factor alone. Therefore, an integrated approach to environmental optimization is needed, rather than isolated improvements. For example, enhancing walkability or increasing green spaces alone may not yield optimal outcomes; however, a comprehensive enhancement of safety and infrastructure alongside these elements can significantly boost their positive impacts on cognitive health. Additionally, based on the cumulative disadvantage theory, the accumulation of multiple adverse factors can significantly exacerbate cognitive decline. Thus, community planning should focus on simultaneous improvements across multiple domains to ensure that older adults benefit from an optimized environment.

6.3. Limitations and Future Research Directions

This study has several limitations. First, the data were collected from specific areas in Shanghai with a limited sample size, restricting the generalizability of the findings. Future studies should include larger, more diverse samples from both urban and rural contexts to enhance external validity. Second, the current study relied on subjective environmental assessments, which may introduce measurement biases. Future research could integrate objective measures to improve the robustness of findings. Lastly, this research employed a simplified theoretical model and did not thoroughly examine potential mediating or moderating mechanisms. Future research should investigate more complex relationships, incorporating mediators and moderators and exploring non-linear interactions to deepen the understanding of how community environments impact cognitive health.

7. Conclusions

This study provides robust empirical evidence that community environmental factors play a crucial role in supporting cognitive function among older adults. By integrating PLS-SEM and fsQCA, we confirmed that Pedestrian Friendliness, Space Attractiveness, and Blue–Green Space are particularly influential, exerting both independent and synergistic effects on cognitive health. Our findings reveal that environments that enhance mobility, aesthetic quality, and access to nature contribute to better cognitive outcomes by promoting regular physical activity, offering psychological restoration, and fostering social interactions.

Regarding theoretical contributions, beyond confirming established relationships, this study advances the field of environmental gerontology by introducing a multi-dimensional and configuration-oriented framework. While PLS-SEM captures the direct and linear impacts of each environmental factor, fsQCA reveals the combinational pathways through which different features jointly shape cognitive function. This dual-method approach aligns with the ecological model of aging and extends prior research by uncovering the non-linear, reinforcing mechanisms among environmental factors.

Regarding practical implications, the results offer actionable insights for urban planners, community designers, and public health policymakers. Prioritizing pedestrian-friendly infrastructure, accessible green and blue spaces, and aesthetically attractive public areas can enhance cognitive well-being in aging populations, especially when resources are constrained. Our findings highlight that community interventions should focus on integrated environmental improvements rather than isolated enhancements to maximize impact.

Regarding broader significance and future research, this study contributes to the ongoing dialogue on healthy aging by demonstrating how community-level interventions can complement individual-level health strategies. Future research could strengthen these findings by adopting longitudinal designs, exploring cross-cultural differences, and integrating objective cognitive performance metrics. By extending this hybrid analytical framework to other contexts, scholars and practitioners can gain a more comprehensive understanding of how urban environments shape cognitive and psychological well-being.