Reframing Sustainable Informal Learning Environments: Integrating Multi-Domain Environmental Elements, Spatial Usage Patterns, and Student Experience

Abstract

Sustainable informal learning environments are increasingly recognized as critical components of educational architecture, yet their environmental and behavioral dynamics remain underexplored. Informal learning spaces (ILS) support flexible, student-driven learning beyond formal classrooms. While prior research often isolates individual environmental factors, integrated multi-domain interactions and reciprocal occupant–space dynamics receive less attention. This study adopts a dual-perspective analytical framework, combining spatial analysis and student surveys (n = 1048) across 130 ILS in five academic buildings in China. The findings highlight several environmental dimensions influencing student experience. One extracted factor combines acoustic and thermal comfort with learning atmosphere—domains seldom grouped together—indicating their collective relevance to student experience. Additionally, spatial openness and natural connectivity further enhance student experience. Importantly, the results show that frequently used spaces receive lower physical quality ratings, group collaboration areas outperform individual study zones, and spontaneously formed spaces—informally appropriated, unplanned areas such as corridors or leftover corners—score lowest. These patterns may reflect mismatches between spatial supply and use intensity, institutional investment priorities, and differing levels of student autonomy and environmental control. This research extends conventional post-occupancy evaluations by introducing a comprehensive dual-perspective framework that links spatial characteristics with user-driven dynamics, and by identifying the combined effects of multi-domain physical environmental and supportive elements on student experience. The insights offer empirical grounding and actionable strategies for campus planners and architects, including prioritizing sensory comfort, enhancing spatial diversity, and supporting student-led adaptations to promote sustainable learning environments.

1. Introduction

Higher education institutions worldwide increasingly recognize the pivotal role of campus environments in achieving sustainable educational outcomes [1,2,3,4]. Informal learning spaces (ILS)—such as lounges, collaborative areas, and atria—have emerged as essential components of contemporary campus design, supporting flexible, student-driven learning beyond traditional classrooms [2,5,6,7]. Within sustainable campus planning, these spaces contribute significantly not only to spatial efficiency but also to enhancing student experience (SE), well-being, and long-term institutional sustainability goals [3,8,9,10].

In recent decades, Chinese higher education has experienced rapid expansion and infrastructural transformation, driven by policies emphasizing educational innovation and sustainable development [11,12,13]. As a result, informal learning environments have been widely integrated into new academic buildings and campus renovation projects across China. Despite this increasing adoption, research in the Chinese context has predominantly focused on formal educational spaces such as traditional classrooms, lecture halls, and laboratories, resulting in fragmented insights regarding how informal learning environments specifically shape student experiences [14,15,16]. This gap significantly constrains both theoretical understanding and practical guidance in designing effective, sustainable informal learning spaces within higher education contexts.

Existing international research frequently highlights the importance of individual indoor environmental elements—including thermal comfort, acoustic conditions, lighting, and indoor air quality—as isolated factors influencing occupant satisfaction and behaviors [17,18,19]. While valuable, this isolated, single-domain approach neglects potential synergistic interactions among different environmental attributes. Zhao and Li [17] emphasize that research examining interactions between multiple indoor environmental elements often yields inconsistent or even contradictory results, noting a particular scarcity of studies that integrate more than two or three elements simultaneously. Makaremi et al. [20] similarly argue that comprehensive assessments of indoor environments should incorporate multiple factors simultaneously, considering their collective implications for occupants’ overall well-being.

As a conceptual extension to these environmental considerations, theories of affordances and 4E cognition argue that student learning arises through ongoing interaction with not only the physical environment, but also functional, psychological, and socio-cultural supports [9,21,22,23,24]. Consistent with this expanded understanding, the 50-year summary report on higher education underscores the importance of integrating physical elements into student experience as an essential innovation in contemporary learning and teaching practices [1]. Thus, examining how multi-domain indoor environmental and supportive elements collaboratively influence student experience represents a critical step toward optimizing the design of educational spaces [25].

Beyond physical conditions, existing literature frequently conceptualizes campus spaces primarily as static entities or management tools, neglecting the dynamic reciprocal relationship between occupants and spaces [4,26,27,28]. Recent sociological perspectives—including theories of co-construction and entanglement—have challenged this static view by highlighting space as an evolving social product continuously shaped by, and shaping, user behaviors and interactions [4,29,30,31,32]. Such dynamic frameworks reveal how occupant behaviors significantly influence the effectiveness, quality, and identity of spaces over time [28,33]. Yet, despite theoretical acknowledgment, empirical research explicitly exploring the reciprocal occupant–space dynamic, particularly in informal learning contexts, remains limited.

Informal learning spaces exemplify this reciprocal dynamic: Characterized by user-driven flexibility and adaptability, these environments are continually reshaped through varied usage patterns [2,10,34,35,36]. However, research and practice currently face several challenges. First, the investment cost for ILS construction is relatively high, given their inherent flexibility and multifunctionality [36]. Second, the demand and utilization of these spaces can be unpredictable, complicating effective design and management [10]. Third, existing design guidelines and evaluation criteria for informal learning spaces remain insufficiently defined [7]. Most evaluations are derived from formal classroom settings or libraries, whose metrics might not fully apply to ILS contexts. Additionally, previous studies often adopt the perspectives of designers or facility managers, with relatively few systematically incorporating student perceptions through mixed-method empirical approaches [37,38]. Finally, most empirical studies to date have focused on libraries or campus-wide informal spaces, overlooking informal learning areas within academic buildings, where student-driven learning activities frequently occur [9,10].

Addressing these critical knowledge gaps, this study adopts a dual-perspective analytical framework. Firstly, from a space-to-user perspective (“space–student experience”, space–SE), the research explores how multi-domain indoor environmental elements and supportive factors within informal learning spaces jointly contribute to shaping student experience. This involves not only conventional indoor environmental elements but also factors such as spatial openness, functional flexibility, learning atmosphere, and the provision of supportive facilities. Secondly, adopting a user-to-space perspective (“student experience–space”, SE–space), the research investigates how student usage patterns—including frequency and modes of use (e.g., individual versus collaborative, planned versus spontaneous)—reciprocally influence perceptions of space quality and effectiveness. This reciprocal approach provides a robust empirical foundation to rethink informal learning spaces as dynamically interactive entities, rather than passive, static environments.

Specifically, the research aims to answer the following questions:

1.

Which multi-domain indoor environmental and supportive elements within ILS collaboratively influence student experience (SE) and student satisfaction (SS) in university academic buildings?

1.1. How can the key elements of ILS that impact SE outcomes be identified?

1.2. How can these elements be continuously evaluated through SS?

2.

How does the SE process affect the construction quality of ILS (CQILS) in university academic buildings?

2.1. How does the frequency of potential space use (FPS) by students affect CQILS?

2.2. Based on the FPS, how do ways in which students use space (WUS) influence CQILS?

This study addresses these questions through an integrated research approach combining spatial syntactic analysis, systematic on-site observation, and extensive surveys conducted with students across 130 informal learning spaces in five academic buildings at two major Chinese universities. By explicitly incorporating multi-domain indoor environmental factors and reciprocal occupant–space dynamics into a single analytical framework, the study extends conventional post-occupancy evaluation (POE) practices.

Overall, this research seeks to advance theoretical understanding and empirical methodologies in educational space design, environmental psychology, and facility management. It also aims to provide architects, campus planners, and educational policymakers with actionable insights to sustainably enhance informal learning environments, thereby contributing to improved student experience and long-term institutional sustainability objectives.

2. Materials and Methods

2.1. Material

This study was conducted at two universities in Wuhan: Huazhong University of Science and Technology (HUST) and Wuhan University (WHU). A total of five academic buildings were selected as case studies (see

Table 1. Overview of research subjects.

School	Huazhong University of Science and Technology (HUST)
Code	A-1	A-2	A-3
Type	Public Teaching Building	Science and Engineering + Arts	Science and Engineering
Name	Yifu Teaching Building (East Ninth Building) Source: https://www.hust.edu.cn/zjhkd/xyfg/jzdb.htm (accessed on 9 June 2025)	Caijian Building (School of Architecture and Urban Planning) Source: http://jjc.hust.edu.cn/ (accessed on 9 June 2025)	Faculty of Optics and Electronic Information Building Source: https://www.hust.edu.cn/zjhkd/xyfg/jzdb.htm (accessed on 9 June 2025)
School	Wuhan University (WHU)
Code	B-1		B-2
Type	Public Teaching Building		Humanities
Name	Main Teaching Building (Engineering Department) Source: https://edf.whu.edu.cn/info/1336/3601.htm (accessed on 9 June 2025)		Zhenhua Building (Comprehensive Building of Liberal Arts) Source: https://philosophy.whu.edu.cn/info/1468/17154.htm (accessed on 9 June 2025)

). The selection criteria are as follows:

• Comprehensive coverage: The selected buildings include general academic buildings and specialized buildings from the humanities, sciences, and arts.
• Representative design: All five buildings were either newly constructed or renovated between 2019 and 2021, making them the most recent facilities available at the time of the fieldwork in 2022. The ILSs were designed as part of the original building plan or optimized during renovation. These buildings offer more modern and innovative designs than traditional academic buildings, reflecting the latest trends in ILS design at these two leading universities. For ease of reference, the five buildings have been numbered, as shown in Table 1.

2.2. Measures

2.2.1. Space–SE

To reach a wider range of respondents and collect diverse data, this section uses a questionnaire survey to examine how ILSs in university academic buildings influence SE. Based on the aforementioned research questions, two questionnaires were designed to systematically explore the key ILS elements affecting SE and their impact on SS (see Figure 1).

The first questionnaire focuses on identifying the ILS elements that influence SE. The questions were developed based on the four-part framework summarized in the literature review: IEC, FASL, ASF, and SASB (see Table 2). Within this framework, 23 indicators were identified as potential factors influencing SE and were measured using a 5-point Likert scale. The full questionnaire is provided in the Supplementary Materials (see Supplementary Materials Files S1 and S2).

Table 2. Core elements of ILS and corresponding literature sources.

Key Factors of Informal Learning Spaces	Literature Sources
Indoor Environmental Comfort (Lighting, Ventilation, Noise, Temperature, Material, Color, Furniture Comfort)	[10,39,40,41,42,43,44,45,46]
Flexibility and Adaptability of Spatial Layout (Furniture Flexibility, Spatial Diversity, Openness and Privacy)	[9,10,34,39,40,41,42,46,47]
Spatial Autonomy and Sense of Belonging (Autonomy, Belonging, Interactivity, Sense of Community)	[5,39,40,41,42,48]
Availability of Supporting Facilities (WIFI, Power Outlets, Food and Beverages, Whiteboards, Computers)	[5,10,39,40,41,42,47,49]

Table 2. Core elements of ILS and corresponding literature sources.

Key Factors of Informal Learning Spaces	Literature Sources
Indoor Environmental Comfort (Lighting, Ventilation, Noise, Temperature, Material, Color, Furniture Comfort)	[10,39,40,41,42,43,44,45,46]
Flexibility and Adaptability of Spatial Layout (Furniture Flexibility, Spatial Diversity, Openness and Privacy)	[9,10,34,39,40,41,42,46,47]
Spatial Autonomy and Sense of Belonging (Autonomy, Belonging, Interactivity, Sense of Community)	[5,39,40,41,42,48]
Availability of Supporting Facilities (WIFI, Power Outlets, Food and Beverages, Whiteboards, Computers)	[5,10,39,40,41,42,47,49]

The second questionnaire builds on the results of the PCA, exploring how these key elements further affect SE through the lens of SS, also using a 5-point Likert scale.

2.2.2. SE–Space

Based on the results of the PCA, this section primarily examines how the FPS and WUS impact the CQILS in university academic buildings.

• Frequency of Potential Space Use: The FPS can be objectively measured using spatial syntactic analysis (SSA). SSA is a tool widely applied in urban planning and architectural studies to analyze spatial configurations and their impact on human behavior. One of its core metrics is integration, which describes the connectivity or accessibility of a given location in relation to other locations. Areas with higher integration are generally easier to access and, thus, more likely to become frequently used spaces by students. By measuring both global and local integration, SSA can reveal spatial characteristics within academic buildings, helping to identify areas more likely to be used by students. These data provide key quantitative support for understanding the FPS [50,51,52,53].
• Ways in which Students Use Space: WUS is determined through field observations, where observers record whether students are engaged in individual study, group collaboration, or both, in the ILSs of each academic building.
• Construction Quality of ILS: The CQILS in academic buildings is assessed using the results of PCA from the initial questionnaire. Each building’s CQILS scores are determined by averaging ratings from two independent observers on a scale of 1 to 5, where 1 represents very poor conditions and 5 represents excellent conditions.

2.2.3. Procedure

Field Observations were conducted in March 2022 by trained research team members following a standardized protocol. Each ILS was geo-located, photographed, and coded based on spatial features and observed usage. Spaces were classified as supporting either individual or group learning based on furniture configuration, actual usage observed during site visits, and brief informal conversations with student users. For each building, observations were conducted once on a weekday (when classes were in session) and once on a weekend (when most courses were suspended), to account for potential differences in space use. All team members underwent calibration training to ensure consistency in coding.

Questionnaire Survey was conducted using the online survey platform SoJump (Changsha Ranxing Information Technology Co., Ltd., Changsha, China) and distributed through two targeted channels. For general-purpose academic buildings commonly used by all students, the survey link was shared via campus-wide social networks such as WeChat and QQ groups. For specialized academic buildings, the questionnaire was distributed only to students in the corresponding departments, ensuring that all respondents had prior experience using the respective spaces. Additionally, face-to-face distribution in the buildings was carried out, with researchers providing guidance and obtaining informed consent. The survey period lasted from March to October 2022.

Principal Component Analysis (PCA) was applied to the first round of questionnaire responses to extract key perceptual dimensions of ILS. PCA was selected to reduce dimensionality and multicollinearity among the 23 assessment variables, and to reveal latent structures in student perceptions [54]. The Kaiser–Meyer–Olkin (KMO) measure (>0.70) and Bartlett’s test of sphericity (p < 0.001) indicated that the data were suitable for PCA. Components were retained based on a cumulative variance explanation threshold exceeding 60%, and varimax rotation was used to enhance interpretability [55].

Spatial Syntactic Analysis was conducted using AutoCAD 2020 (Autodesk Inc., San Rafael, CA, USA) and depthmapX (University College London, London, UK) software. The details of the process can be found in

Table A1. Steps and software used for generating integration metrics.

Step	Description	Software and Format
1. Creating Floor Plans	Floor plans of the five academic buildings were drawn with attention to detail. Saved for later use.	AutoCAD (.dwg)
2. Dividing Convex Spaces	Research areas were divided into convex spaces defined by walls and partitions. Saved for import.	AutoCAD (.dxf)
3. Generating Convex Space Maps	.dxf files were imported into depthmapX to generate maps with nodes and connecting pathways.	depthmapX
4. Calculating Integration	Analyzed the convex space maps to compute Global Integration (HH) and Local Integration (HHR3, HHR5).	depthmapX
5. Exporting Integration Data	Integration metrics for each building were exported for further analysis.	depthmapX

.

2.2.4. Analysis

• Analysis of Space–SE

Based on the data collected from the first questionnaire, reliability and validity analyses were conducted using SPSS S25 (IBM Corp., Armonk, NY, USA) to verify the internal consistency of the scale and the correlation between variables. Principal component analysis (PCA) was then used to extract factors from the 23 assessment indicators in the questionnaire, identifying statistically significant key elements of ILS. This analysis addresses research question 1.1: How can the key elements of ILS that impact SE outcomes be identified?

Next, based on the results of the second questionnaire, correlation analysis was performed using SPSS S25 to explore the relationship between the identified key elements and SS. This analysis addresses research question 1.2: How can these elements be continuously evaluated through SS?

The analysis is divided into two parts to address the two main research questions.

• Analysis of SE–Space

First, the global integration (Integration HH) and local integration (Integration HH3 and Integration HH5) scores for all spaces in each building were ranked. These spaces were then categorized into three groups—high (top 33%), medium (33–67%), and low (bottom 33%)—based on their integration scores. This classification establishes a basis for determining each space’s level of integration.

Next, integration data specific to ILSs, as identified through field observation numbering, were extracted, ranked from high to low, and similarly categorized.

Then, ILSs that ranked high or low in both global and local integration were identified. These spaces were analyzed qualitatively by reviewing the field observation photos and ratings to explore research question 2.1: How does the FPS by students affect CQILS?

Finally, the ways in which students used these spaces (e.g., individual study or collaborative learning), as observed in the field, were further analyzed to understand their impact on the CQILS. This analysis addresses research question 2.2: Based on the FPS, how do WUS influence CQILS?

3. Results

3.1. Space–SE

3.1.1. Identification of Multi-Domian Indoor Environmental and Supportive Elements Affecting SE

The first survey aimed to identify the ILS elements that influence SE. A total of 270 questionnaires were collected, with 241 valid responses after excluding invalid ones. Among these, 116 responses were from HUST and 125 from WHU, with an overall valid response rate of 89.2%. The gender distribution in the valid sample was fairly balanced, with 49.4% male and 50.6% female respondents.

• Reliability Analysis of the Questionnaire

The overall reliability of the questionnaire was 0.919, with results all exceeding 0.9, indicating a very high level of internal consistency in the scale (see

Table 3. Reliability statistics for SE questionnaire.

	HUST	WHU	Overall
Cronbach’s Alpha	0.909	0.928	0.919
Number of Items	24	24	24
Number of Cases	116	125	241

). A breakdown of reliability by dimension showed that all dimensions had reliability scores above 0.7, further confirming good internal consistency.

• Validity Analysis and Factor Analysis on Factors Influencing SE

The results from the questionnaire were entered into SPSS for exploratory factor analysis. The validity of the scale was found to be 0.878, with a KMO value greater than 0.6. In addition, the significance values in Bartlett’s test of sphericity were all less than 0.05, indicating a strong correlation among the variables and that the questionnaire was suitable for factor analysis.

Using PCA, five common elements were extracted from the 23 evaluation items in the questionnaire, with a cumulative variance contribution rate of 62.999% (see

Table 4. Total variance explained.

	Initial Eigenvalues			Sum of Squared Loadings for Extraction			Sum of Squared Loadings for Rotation
Com.	Tot.	Var.%	Cum.%	Tot.	Var.%	Cum.%	Tot.	Var.%	Cum.%
1	8.402	36.530	36.530	8.402	36.530	36.530	3.265	14.194	14.194
2	2.054	8.932	45.462	2.054	8.932	45.462	3.250	14.129	28.323
3	1.482	6.443	51.905	1.482	6.443	51.905	2.902	12.619	40.942
4	1.351	5.874	57.779	1.351	5.874	57.779	2.651	11.524	52.467
5	1.201	5.220	62.999	1.201	5.220	62.999	2.422	10.532	62.999

Extraction Method: principal component analysis. Abbreviations: Com. = component; Tot. = total; Var.% = variance percentage; Cum.% = cumulative percentage.

), which is greater than 60%, indicating that dividing the 23 evaluation elements into five dimensions is appropriate. The rotated component matrix showed that the loadings of the 23 evaluation items were all above 0.5 on a single dimension (see

Table A2. Rotated component matrix.

	Component
	1	2	3	4	5
Availability of printers, copiers, etc.	0.795
Availability of computers, projectors, whiteboards, etc.	0.767
Availability of beverages and light snacks	0.699
Adjustable lighting, temperature, and ventilation	0.561
Adjustable sound and external visibility	0.514
High comfort of furniture		0.689
Sufficient number of furniture		0.609
High flexibility of furniture for easy rearrangement		0.605
Refined interior design		0.597
High-quality WiFi signal		0.584
Sufficient spaciousness		0.532
Sufficient power outlets		0.512
Adequate natural lighting			0.803
Spacious window view			0.802
Good ventilation			0.731
Appropriate artificial lighting			0.636
Low noise level				0.763
Suitable temperature				0.721
Strong learning atmosphere				0.646
Adequate openness					0.791
Diverse spatial types					0.701
Positive communication atmosphere					0.584
High level of privacy					0.501

Extraction Method: principal component analysis. Rotation Method: Kaiser normalization with varimax rotation.

), meaning that no items needed to be removed, and the validity of the data is reliable.

Based on the reclassification of the evaluation elements into five dimensions, the following names were assigned to the dimensions affecting SE (see Figure 2): Dimension 1, supporting services and spatial autonomy (SSSA); Dimension 2, spatial availability and furniture flexibility (SAFF); Dimension 3, lighting, ventilation, and window view (LVWV); Dimension 4, acoustic and thermal control and learning atmosphere (ATCLA); and Dimension 5, spatial diversity and openness (SDO).

3.1.2. Assessment of ILS Factors Affecting SE Based on SS

Based on the five dimensions identified from the PCA in the previous section, the research team proceeded to evaluate the key elements of ILS that impact SS. The aim is to provide empirical evidence to inform the more effective design of ILS in the future.

A total of 778 questionnaires were distributed, with 705 valid responses, including 431 from HUST and 274 from WHU, yielding an overall validity rate of 90.62%. The reliability of the questionnaire was high (0.961, greater than 0.9), and its validity was confirmed (KMO value was 0.927, greater than 0.8, and the significance p-value of Bartlett’s sphericity test was 0.000 ***) (see

Table 5. Reliability statistics for SS questionnaire.

	HUST			WHU		Overall
	A-1	A-2	A-3	B-1	B-2
KMO Value	0.925	0.785	0.813	0.745	0.898	0.927
Bartlett’s Test Sig.	0.000 ***	0.000 ***	0.000 ***	0.000 ***	0.000 ***	0.000 ***

Note: *** $p<0.001$ (significant correlation, based on Bartlett’s Test Sig.).

). The gender distribution was balanced, with undergraduate students accounting for more than 70%.

First, correlation analysis was conducted using SPSS S25 (IBM Corp., Armonk, NY, USA), and the results showed that the significance between the dimensions remained well below 0.01, indicating that all dimensions are positively and significantly correlated with each other. Based on this, the influence mechanism of the five dimensions on overall SS was further explored. According to the demographic variable comparison results from the preliminary questionnaire, the building, gender, major, usage frequency, and usage duration were identified as significant factors influencing overall satisfaction. To enhance the accuracy of the results, these five demographic variables were set as control variables in the linear regression analysis (see

Table 6. Model summary.

Model	R	R Square	Adjusted R Square	Errors in Standard Estimates
1	0.770	0.592	0.580	0.596

Predictors (constant): acoustic and thermal control and learning atmosphere (ATCLA); lighting, ventilation, and window view (LVWV); supporting services and spatial autonomy (SSSA); spatial diversity and openness (SDO); spatial availability and furniture flexibility (SAFF). Dependent variable: overall impact. Control variables: academic building, gender, major, usage frequency, duration of use.

).

Table 6 demonstrates a good fit of the linear regression model (R² = 0.592 > 0.4). The ANOVA results show that F = 49.680 and p < 0.001 (see

Table A4. ANOVA.

Model		Sum of Squares	df	Mean Square	F	Sig.
1	Regression	353.470	20	17.674	49.680	0.000
	Residual	243.330	684	0.356
	Total	596.800	704

Predictors (constant): acoustic and thermal control and learning atmosphere (ATCLA); lighting, ventilation, and window view (LVWV); supporting services and spatial autonomy (SSSA); spatial diversity and openness (SDO); spatial availability and furniture flexibility (SAFF). Dependent variable: overall impact. Control variables: academic building, gender, major, usage frequency, duration of use.

), suggesting that at least one of the five independent variables significantly influences the dependent variable. Additionally, the variance inflation factor (VIF) values for all five independent variables are below 10, with p < 0.001 (see

Table A3. Linear regression coefficients.

Model	Unstandardized Coefficients	Standard Coefficients		t	Sig.	Collinearity
	B	Standard Error	Beta			VIF
(Constant)	0.378	0.183		2.064	0.039 *
Independent Variables
SSSA	0.022	0.043	0.020	0.506	0.613	2.741
SAFF	0.086	0.073	0.068	1.182	0.238	5.584
LVWV	0.160	0.044	0.137	3.618	0.000 ***	2.416
ATCLA	0.313	0.049	0.275	6.349	0.000 ***	3.143
SDO	0.370	0.057	0.319	6.536	0.000 ***	3.989
Control Variables
Teaching Building
A-1	0.029	0.079	0.013	0.367	0.714	2.110
A-2	−0.396	0.098	−0.175	−4.042	0.000 ***	3.159
A-3	−0.372	0.100	−0.156	−3.739	0.000 ***	2.934
B-1	−0.156	0.088	−0.067	−1.778	0.076	2.353
B-2	0
Gender
Male	−0.167	0.047	−0.091	−3.508	0.000 ***	1.117
Female	0
Major
Philosophy, Economics, Law	0.069	0.082	0.030	0.849	0.396	2.146
Military Science, Management, Arts	0.219	0.113	0.059	1.935	0.053	1.549
Science, Engineering, Agriculture, Medicine	0.138	0.084	0.074	1.644	0.101	3.372
Education, Literature, History	0
Usage Frequency
Daily	−0.158	0.092	−0.069	−1.717	0.086	2.747
Weekly	−0.240	0.088	−0.129	−2.735	0.006 **	3.714
Monthly	−0.232	0.097	−0.096	−2.392	0.017 *	2.711
Quarterly	−0.107	0.114	−0.030	−0.943	0.346	1.753
Rarely	0
Duration of Use
Less than 30 min	0.208	0.087	0.086	2.399	0.017 *	2.176
30 min–1 h	0.066	0.073	0.028	0.905	0.366	1.651
1 h–3 h	0.129	0.061	0.068	2.109	0.035 *	1.769
More than 3 h	0

Note: ***p < 0.001; ** p < 0.01; * p < 0.05 (significant correlation).

), indicating no severe multicollinearity issues. The resulting regression model is as follows:

$$\begin{array}{c}\hfill \mathrm{Overall}\mathrm{SS}=0.378+0.370\times \mathrm{SDO}+0.313\times \mathrm{ATCLA}+0.160\times \mathrm{LVWV}\end{array}$$

From the regression model, it can be seen that three of the five dimensions significantly affect overall SS. The contribution rate of SDO to satisfaction is 37%, ATCLA contributes 31%, and LVWV contributes 16%. The other two factors, SAFF and SSSA, do not significantly influence overall SS.

3.1.3. Summary of Results in Space–SE

Based on the reclassification of the evaluation factors into five dimensions, the following names were assigned to the dimensions affecting SE: SSSA, SAFF, LVWV, ATCLA, and SDO. The further analysis of the results is as follows:

• PCA revealed five dimensions influencing SE, highlighting complex interactions between multi-domain indoor environmental and supportive elements within ILS.
• SAFF and SSSA do not have a significant impact on SS.
• The three key factors—SDO, ATCLA, and LVWV—were found to significantly influence SS (see Figure 3).

3.2. SE–Space

3.2.1. ILSs Screening Through Integration Scores (FPS) for Each Building

To measure FPS by students, this study uses SSA integration data as a key indicator. Global integration (HH) measures the accessibility of a space within the entire building, while local integration (measured by HHR3 and HHR5) reflects accessibility within smaller areas, such as specific floors or adjacent areas. Higher integration scores indicate better accessibility and a higher likelihood of frequent student use.

For each building, the integration scores are categorized into high (top 33%), medium (33–67%), and low (bottom 33%) categories. Following the classification of the integration scores, the selected ILSs are categorized accordingly.

Table 7. ILSs screening through integration scores (FPS) for each building.

Building	Global Integration (HH)	Local Integration (HHR3)	Local Integration (HHR5)	High Integration ILS	Low Integration ILS
A-1	0.607–1.077	0.690–7.795	0.896–2.110	3F-4, 4F-5	2F-3
A-2	0.492–1.412	0.333–4.046	0.422–2.042	4F-6, 4F-7, 5F-10	1F-1-1, 1F-1-2, 1F-1-3, 2F-4-1, 2F-4-2, 3F-5-1, 3F-5-2, 3F-5-3
A-3	0.129–1.007	0.333–8.216	0.458–2.190	4F-3, 5F-3, 5F-12, 5F-16, 5F-20, 6F-14, 6F-17, 6F-21, 7F-13, 7F-18, 7F-22, 8F-19	2F-2
B-1	0.685–1.219	0.637–2.703	0.726–1.681	2F-1, 2F-2, 4F-6
B-2	0.443–1.062	0.333–2.945	0.574–1.667	2F-4, 3F-5, 3F-6, 3F-7, 3F-8, 3F-9, 4F-5, 4F-7, 4F-8, 4F-9, 4F-10, 4F-11, 5F-8, 5F-11	5F-17, 6F-18

summarizes the integration scores and categorization of ILSs in different buildings, based on global and local integration values. The selected ILSs are grouped into high and low integration categories according to global and local integration scores (ILS numbers and locations are detailed in Appendix C):

3.2.2. Comparative Analysis of CQILS Based on FPS

This section aims to further evaluate the quality of ILSs identified through integration scores (FPS). The objective is to compare CQILS scores between high- and low-FPS categories and explore how student FPS relates to CQILS.

A total of 26 ILSs were identified in the high category for global and local integration, while 12 ILSs were in the low category. Each ILS was evaluated on five dimensions derived from the questionnaire factor rotation analysis: LVWV (W1), ATCLA (W2), SDO (W3), SAFF (W4), and SSSA (W5). The average score across these dimensions determined the overall CQILS (W6), rated on a scale of 1 to 5, where 1 indicates poor conditions and 5 represents excellent conditions.

The results reveal significant differences in CQILS scores between high- and low-FPS ILSs (detailed CQILS scores for each ILS are provided in Appendix D):

• W1 (LVWV): High-FPS ILSs averaged 3.7, while low-FPS ILSs averaged 4.3.
• W2 (ATCLA): High-FPS ILSs averaged 2.4, compared to 3.3 for low-FPS ILSs.
• W3 (SDO): High-FPS ILSs averaged 2.5, compared to 3.0 for low-FPS ILSs.
• W4 (SAFF): High-FPS ILSs averaged 3.3, while low-FPS ILSs averaged 3.9.
• W5 (SSSA): High-FPS ILSs averaged 1.3, slightly lower than 1.7 for low-FPS ILSs, with overall low scores observed in both categories.
• W6 (CQILS): The average score for high-FPS ILSs was 2.6, while low-FPS ILSs averaged 3.2.

In summary, high-FPS ILSs tended to score lower across all dimensions compared to low-FPS ILSs, suggesting that spaces with higher FPS may have lower overall CQILS.

3.2.3. Comparative Analysis of WUS Based on FPS and CQILS

This section identifies the ILSs with CQILS (W5) scores in the top 33% (high category) and bottom 33% (low category) for high FPS, as well as those with CQILS scores in the top 33% for low FPS. By combining FPS, WUS, and CQILS, we can gain a more comprehensive understanding of how these factors collectively influence the usage and construction quality of ILSs.

Comparative analysis and results for different types of ILS (see

Table 8. Further ILSs screening through WUS and CQILS.

Building	ILS Number	FPS Category	WUS	CQILS Category
A-1	3F-4	High	Individual	High (3.1)
A-2	4F-6	High	Group	High (3.7)
A-2	5F-10	High	Individual	High (3.0)
A-3	5F-20	High	Individual	High (3.4)
A-3	6F-14	High	Individual	High (3.6)
B-2	3F-6	High	Individual	High (3.5)
A-2	4F-7	High	Individual	Low (2.1)
A-3	4F-3/5F-3	High	Spontaneously Formed	Low (2.0)
A-3	7F-13	High	Individual	Low (2.2)
B-1	2F-1/2F-2	High	Individual	Low (2.1)
B-1	4F-6	High	Individual	Low (2.4)
B-2	2F-4	High	Spontaneously Formed	Low (1.8)
B-2	4F-10	High	Spontaneously Formed	Low (2.1)
B-2	4F-11/5F-11	High	Individual	Low (2.2)
A-1	1F-1-1	Low	Individual	High (3.4)
A-2	3F-5-1	Low	Group	High (4.2)
A-2	3F-5-2	Low	Group	High (4.3)
B-2	5F-17	Low	Individual	High (3.7)

Spontaneously Formed: Refers to spaces informally adopted and rearranged by students for learning activities.

):

• Group Collaboration vs. Individual Learning—High-FPS ILS

In high-FPS spaces, group collaboration spaces had a higher CQILS score (3.7), while individual learning spaces scored lower (from 2.0 to 3.6).

Result: In high-FPS spaces, CQILS scores for group collaboration ILSs were generally higher than those for individual learning ILSs, and individual learning spaces exhibit a larger range of score fluctuations.

• Group Collaboration vs. Individual Learning—Low-FPS ILS

In low-FPS spaces, group collaboration spaces (Building A-2’s 3F-5-1 and 3F-5-2, with CQILS scores of 4.2 and 4.3) had significantly higher scores than individual learning spaces (Building A-1’s 1F-1-1 and Building B-2’s 5F-17, with CQILS scores of 3.4 and 3.7).

Result: In low-FPS spaces, group collaboration spaces had better construction quality than individual learning spaces.

• High FPS vs. Low FPS—Group Collaboration ILS

In group collaboration ILSs, low-FPS spaces (Building A-2’s 3F-5-1 and 3F-5-2, with CQILS scores of 4.2 and 4.3) had higher scores than high-FPS spaces (Building A-2’s 4F-6, CQILS 3.7).

Result: In group collaboration spaces, low-FPS group collaboration ILSs had better construction quality than high-FPS group collaboration ILSs.

• High FPS vs. Low FPS—Individual Learning ILS

In individual learning spaces, low-FPS spaces (Building A-1’s 1F-1-1 and Building B-2’s 5F-17, with CQILS scores of 3.4 and 3.7) scored higher than most high-FPS individual learning spaces (except Building A-3’s 5F-20 and 6F-14, and Building B-2’s 3F-6, all others have CQILS scores below 3.4).

Result: In individual learning spaces, low-FPS individual learning ILSs performed better in terms of construction quality.

• Spontaneously Formed ILS

Student-generated ILSs are typically located in high-FPS areas, mainly used for rest, discussion, or storage, such as Building A-3’s 4F-3/5F-3 (CQILS 2.0) and Building B-2’s 2F-4 (CQILS 1.8) and 4F-10 (CQILS 2.1).

Result: CQILS scores for emerge spontaneous ILSs were generally below average, indicating that while these spaces lacked formal design and facilities, they remained important due to high usage frequency.

3.2.4. Summary of Results for SE–Space

• ILSs with higher FPS tend to have lower CQILS compared to those with lower FPS.
• Group collaboration spaces generally exhibit better CQILS than individual learning spaces, and individual learning spaces exhibit a larger range of score fluctuations.
• ILSs that are spontaneously formed by students tend to have lower CQILS.

4. Discussion

This section will discuss the following key findings from the study:

1.

PCA revealed five dimensions influencing SE, highlighting complex interactions between multi-domain indoor environmental and supportive elements within ILS.

2.

SAFF and SSSA do not have a significant impact on SS.

3.

The three key factors—SDO, ATCLA, and LVWV—were found to significantly influence SS.

4.

ILSs with higher FPS tend to have lower CQILS compared to those with lower FPS.

5.

Group collaboration spaces generally exhibit better CQILS than individual learning spaces, and individual learning spaces exhibit a larger range of score fluctuations.

6.

ILSs that are spontaneously formed by students tend to have lower CQILS.

4.1. Space–SE

In addressing Research Question 1 (Which multi-domain indoor environmental and supportive elements within ILSs collaboratively influence SE and SS in university academic buildings?), this study first identified five key dimensions of ILSs that affect the SE. A linear regression model was then used to analyze how these dimensions impact overall SS.

• Result 1: PCA revealed five dimensions influencing SE, highlighting complex interactions between multi-domain indoor environmental and supportive elements within ILSs

The results of principal component analysis (PCA) showed that the original four independent dimensions derived from the literature review—IEC, FASL, ASF, and SASB—were further refined into five dimensions: SSSA, SAFF, LVWV, ATCLA, and SDO.

Specifically, SSSA combines the availability of supporting services with students’ need for spatial control; SAFF emphasizes spatial availability as well as furniture flexibility and comfort; LVWV addresses the quality of light, air circulation, and window views within the indoor environment; ATCLA combines acoustic and thermal control with the learning atmosphere; and SDO highlights the role of spatial diversity and openness in shaping SE.

This refinement of dimensions suggests that SE is influenced by more diverse factors, reflecting more intricate interactions between multi-domian indoor environmental and supportive elements (e.g., spatial support, facility support, behavioral and psychological support) [56]. The various elements within learning spaces interact with one another and should not be viewed as independent mechanisms but rather as contributing to enhancing or diminishing the overall SE through multidimensional interactions [20,57]. This also supports the argument by Makaremi et al. [20] that research on student well-being related to campus buildings often focuses solely on physical dimensions, overlooking a broader perspective that considers environmental, individual, and collective factors to better understand and assess student well-being in campus environments.

• Result 2: SAFF and SSSA do not have a significant impact on SS

While previous studies have highlighted the importance of appealing interior design, flexible furniture, and supporting facilities for SE [44,58,59,60], this study found that students in ILS of university buildings prioritize functional needs—such as quiet environments, appropriate temperature, and good lighting—over aesthetics or furniture flexibility. This could be due to a threshold effect, where once basic requirements are met, further improvements have diminishing returns on satisfaction [9,61].

Although some studies suggest that greater control over surroundings increases satisfaction [62], this study found that supporting services and spatial autonomy had no significant impact on SS. This may be because students have low expectations for these elements or have become accustomed to current usage patterns. Makaremi et al. [20] also noted that, in academic environments, students often lack effective control over indoor environments, and even where control is possible, conflicting preferences in shared spaces can limit satisfaction improvements.

• Result 3: SDO, ATCLA, and LVWV were found to significantly influence SS

In the linear regression model, SDO contributed 37% to SS, indicating that universities should provide a variety of ILS types to meet students’ diverse learning needs. ATCLA contributed 31%, underscoring the importance of effective noise control, temperature regulation, and a conducive learning atmosphere. This can be achieved through the use of acoustic materials, optimization of ventilation systems, and appropriate use of color and materials. LVWV contributed 16%, highlighting the significance of natural light, air circulation, and outdoor views on students’ physical and mental well-being as well as their learning efficiency.

Additionally, the importance of lighting and ventilation indicates that universities should incorporate sustainable development principles in ILS construction, maximizing the use of natural light and ventilation to reduce reliance on artificial energy sources, thereby creating healthy, comfortable learning environments. This strategy not only meets current learning needs but also provides long-term support for environmental protection and energy conservation, contributing to the development of a more sustainable campus ecosystem [63].

4.2. SE–Space

In addressing Research Question 2 (How does SE process affect the CQILS in university academic buildings?), this study first marked the locations of all ILSs in academic buildings through field observations. The SSA was then employed to identify the FPS by students. High- and low-FPS ILSs were subsequently selected and evaluated using the results of PCA from the first questionnaire (e.g., ATCLA). The relationship between FPS and CQILS was analyzed, along with the impact of WUS on CQILS.

• Result 4: ILSs with higher FPS tend to have lower CQILS compared to those with lower FPS

From a theoretical perspective, ILSs with high FPS should receive more investment, and, thus, their CQILS scores should be higher than those with low FPS. However, the findings of this study contradict this assumption, and several possible explanations are offered:

Firstly, students tend to choose more private spaces for individual study or group collaboration, while high-FPS spaces are typically more open. This may result in low-FPS spaces receiving more attention and investment. Secondly, many high-FPS ILSs may not have been originally planned as ILSs, with construction budgets prioritized for core functional areas like classrooms, leading to insufficient investment in these spaces. Additionally, high-FPS ILSs are likely used frequently by faculty or administrators, limiting students’ autonomy to modify these spaces and hindering further optimization. Finally, students may lack the initiative to actively modify high-FPS spaces [64], which could result in these spaces remaining under-optimized in both design and functionality.

• Result 5: Group collaboration spaces generally exhibit better CQILS than individual learning spaces, and individual learning spaces exhibit a larger range of score fluctuations

Although field observations indicated a preference for quiet individual study in academic buildings, group collaboration spaces received higher CQILS scores than individual learning spaces. This may be due to the following reasons:

Firstly, group collaboration spaces are often prioritized in design and construction, with greater attention and investment from institutions and stakeholders, reflecting a collectivist approach in spatial allocation [65,66]. Secondly, from an economic perspective, group spaces serve more students and meet diverse learning needs without additional investment for individual requirements. Additionally, current educational models emphasize collaborative learning [67,68], leading to increased investment in such spaces. From a sociological standpoint, group collaboration spaces foster social networks and social capital, which benefits both academic and future career development [69,70].

Furthermore, individual learning spaces exhibited greater variability in CQILS scores. This may be due to the limited availability of ILSs specifically designed for individual study in academic buildings, prompting students to repurpose group spaces for solitary activities. In these settings, students often rearrange furniture or create secluded areas for a conducive learning atmosphere, resulting in higher CQILS scores in some cases. However, ILSs dedicated to individual study are typically located in transitional areas like corridors or corners—high-FPS areas—where limited quantity and small size restrict student influence. This also supports Result 3, which found that high-FPS ILSs generally have lower CQILS than low-FPS ILSs.

• Result 6: ILSs that are spontaneously formed by students tend to have lower CQILS

The lower CQILS for spontaneously formed ILSs is an expected outcome. These spaces are typically located in more private or highly individualized areas of the building environment (e.g., corners or outside office doors), lacking formal design and planning. They often have poor lighting and ventilation and rely mostly on simple movable furniture without significant spatial reconfiguration or functional optimization. Spontaneously formed ILSs usually serve specific student groups, making it challenging to secure resources for construction and subsequent investment. However, there are exceptions. For example, the spontaneously formed ILS in Building A-2 benefits from favorable site conditions, relatively complete facilities, and flexible furniture provided by the faculty, which encourages students to reconfigure the space—turning it into an ideal location for self-organized study.

This phenomenon suggests that spontaneously formed ILSs have the potential to form spatial networks when appropriately guided and supported. Empowering students to modify these spaces can better reflect their actual needs, transforming these ILSs into anchors for broader indoor space optimization. Such an approach may contribute to the development of a spontaneous spatial network encompassing formal learning spaces, informal learning spaces, and public areas, ultimately enhancing and optimizing the overall learning environment.

4.3. Actionable Recommendations for Campus Planners and Designers

Based on the six major findings discussed, the following actionable recommendations are proposed to guide future campus planning and learning space design:

• Recognize the multidimensional nature of student experience (SE): Planning and design of ILS should account for the interplay between environmental and supportive elements—including layout, furniture, and atmosphere—as SE is not shaped by single factors alone. (Related to Result 1)
• Focus investment on elements with the greatest impact: Prioritize interventions targeting spatial diversity and openness (SDO), thermal/acoustic comfort and atmosphere (ATCLA), and lighting/ventilation/window views (LVWV), as these dimensions most significantly influence student satisfaction. (Related to Result 3)
• Avoid overinvestment in underperforming elements: While supportive facilities and spatial autonomy (SAFF and SSSA) are often highlighted in the literature, they may not always align with students’ actual priorities or satisfaction outcomes. Design resources should be reallocated accordingly. (Related to Result 2)
• Reinforce the quality of highly used spaces: Since high-FPS ILSs tend to have lower perceived quality, these should become focal points for environmental and spatial enhancement to maximize their impact. (Related to Result 4)
• Support both formal and spontaneous spatial formations: Planners should not only design for structured collaboration and individual learning but also empower students to appropriate and adapt informal spaces. This includes providing movable furniture and environmental support to strengthen CQILS in self-initiated ILSs. (Related to Result 5 & Result 6)

5. Future Research Implications

5.1. Deepening the Quantification of Student Experience: Beyond Generalized Models to Multidimensional Research

Zhao and Li [17] highlighted that existing overall satisfaction and indoor environmental auality (IEQ) models are mostly based on average user responses, falling under generalized models that struggle to effectively capture individual differences. Additionally, multi-domain IEQ studies have largely focused on the impact of the indoor physical environment on learning outcomes, often employing case study methods. However, learning outcomes are not always the best performance indicators, and students themselves may not be ideal self-assessors [20]. Therefore, considering SE as a key indicator for future research may provide a more comprehensive and effective approach than traditional metrics of learning outcomes.

While the SE model used in this study still belongs to the generalized category, analysis of students’ potential usage frequency and spatial utilization revealed variations suggesting that overall SE may influence space quality through multiple dimensions. Thus, future research could quantify SE in a more detailed, multidimensional manner.

Currently, several well-established SE scales have been developed in the educational field, such as the Learning Engagement Questionnaire [71] and the Learning Experience Survey [72]. However, these tools do not fully quantify students’ behavioral and psychological states in ILSs outside the classroom. This study used spatial syntactic analysis to quantify students’ potential frequency of ILS use and field observations to assess usage patterns, but these methods have limitations in precision and accuracy. Therefore, future studies could consider incorporating tools such as accelerometers, GPS, virtual reality (VR), eye trackers, and physiological sensors to provide a deeper quantification of students’ perceptions and behaviors in learning spaces, thereby offering a more comprehensive reflection of SE [73,74,75,76].

5.2. Acoustic and Thermal Control and Learning Atmosphere: Combined Effects of Multi-Domain Physical Environmental and Supportive Elements on Student Experience

Current research often focuses on the individual effects of either indoor environmental elements or supportive elements on SE, with limited attention given to their combined impact. Therefore, future studies should focus on the synergy between multi-domain indoor environmental elements and supportive elements to fully understand which elements jointly influence SE. This will provide a scientific basis for designing learning spaces in educational buildings, ultimately creating environments that better support student learning and development.

In the context of school building environments, acoustic and thermal control are critical physical environmental elements that influence SE. Thermal comfort holds a high priority in educational environments as temperature significantly affects students’ attention, whereas acoustic factors are more likely to induce anxiety [77,78]. Learning atmosphere, as a crucial element of educational settings, has gained widespread interdisciplinary and cross-domain interest in recent years, with increasing attention given to emotional phenomena beyond the internal psychological realm [79]. In management and social sciences, atmosphere is typically defined as individuals’ subjective perception of a situation, with psychological attributes that can significantly influence behavior and attitudes [79]. Thus, focusing on the learning atmosphere, rather than solely on indoor environmental quality, may more effectively reveal the connection between the physical environment and student behavior [80,81].

Although learning atmosphere is challenging to observe directly, it manifests in the interaction among individuals within a space and influences the emotional state of groups [82]. Yan et al. [83] proposed a method based on facial expression recognition (FER) technology, using convolutional neural networks to monitor students’ emotional states in real time and applying shape–color–length (SCL) visualization to quantify classroom atmosphere attributes. These technologies provide feasibility for quantifying the learning atmosphere and offer a data foundation for exploring how learning atmosphere and physical environment jointly influence SE. Future research could draw on these methods to further investigate the effects of varying acoustic and thermal conditions on the learning atmosphere and its impact on SE, thereby optimizing learning space design to enhance SE.

5.3. Window Views, Daylighting, Ventilation, and Spatial Diversity: Combined Effects and Pathways to Enhance Student Experience and Sustainable Design

The positive impact of natural light, ventilation, and window views on physical and mental well-being is well documented [84,85,86]. However, most research on window views has primarily focused on aspects such as greenery and vegetation diversity, with limited exploration of their interaction with daylighting and ventilation. The study by Koyaz and Ünlü [87] provides an innovative perspective in this area, positioning daylighting, ventilation, and window views within the framework of user-building facade interactions. It emphasizes user experience as a holistic concept that encompasses all emotional and behavioral aspects of interaction with the facade and/or indoor environment. Recent empirical studies have increasingly converged on the view that integrating natural visual elements, daylighting, and ventilation can jointly enhance students’ cognitive and emotional functioning in learning environments. These effects span improved attention and academic performance [88], heightened creative thinking, particularly in early cognitive stages [89], and better emotional regulation underpinned by neurobiological responses to visual complexity and natural light [90].

Different space types influence individual and group behaviors differently, making spatial type and diversity crucial factors affecting SE. A study by Dong et al. [91] compared five types of university dormitory spaces and found that single and twin rooms with balconies offered higher satisfaction, and an increase in per capita area positively impacted occupant satisfaction. However, when the per capita area reached 13.5 square meters, the increase in satisfaction plateaued. Future research could consider applying similar methodologies to investigate the impact of ILS types and diversity on SE, providing empirical evidence to optimize the design of ILS.

Beyond immediate experiential and behavioral benefits, integrating spatial diversity, adaptive space configurations, and environmental qualities such as daylighting, ventilation, and visual connections to nature collectively contribute to the sustainability of informal learning environments. Diverse spatial typologies and flexible layouts support adaptability and resilience, reducing the need for frequent structural adjustments or renovations [92]. Simultaneously, environmental strategies like natural ventilation, optimized daylight utilization, and biophilic elements can significantly reduce energy consumption and enhance ecological performance [93]. Future research should holistically examine how combinations of these spatial and environmental design principles can create educational spaces that are not only conducive to student well-being and performance but also inherently sustainable in terms of both ecological and social dimensions.

5.4. Spontaneous Use and Dynamic Utilization of Informal Learning Spaces

This study found that the CQILS formed spontaneously by students was generally lower, revealing a potential conflict between bottom-up space usage and top-down space design. In many educational buildings, the structure functions as a shared environment that accommodates different users and resources. The dynamic use of learning spaces—such as individual and collaborative learning—is a significant aspect of human interaction but remains underexplored [73].

Behavioral geography and time geography provide effective methodologies for studying the dynamic use of space by individuals [94]. For example, Villarreal et al. [73] conducted a study combining qualitative surveys (n = 50) and mixed social data (n = 196) with accelerometer-based spatial use data in a building featuring classrooms, laboratories, and informal learning lounges. This building was equipped with 241 accelerometers connected to 136 sensor mounts, which sensitively measured floor vibrations to capture human foot traffic and other activities. This approach provides valuable empirical data for analyzing human dynamic behavior in ILSs. Another method for studying students’ dynamic use of space is collecting temporal-spatial behavior data using GPS software during their daily activities, then analyzing these behaviors to assess the effectiveness of campus planning [74,95]. Moreover, utilizing these behavioral data to train agent-based decision models represents a forward-looking research direction, aiding in predicting and optimizing student use of ILSs [96,97,98].

Future research could further explore spontaneous student use of ILSs and how these spaces support learning needs and social interactions. Leveraging technologies such as accelerometers and GPS to gather dynamic behavior data, combined with agent-based decision modeling for simulation and optimization, could provide empirical support for better designing and managing ILSs.

6. Cross-Cultural Comparisons Limitations

Although this study provides valuable insights into informal learning spaces (ILSs) within Chinese university campuses, cross-cultural comparisons remain limited due to scope and methodological constraints. International studies clearly illustrate how cultural and institutional contexts shape distinct ILS practices. For instance, Salih et al. [99] conducted a comprehensive systematic review highlighting key themes and strategies in designing sustainable and inclusive ILSs, emphasizing multifunctionality and flexibility. Baker [100] explored an ILS project within a university in Dubai, revealing the significant influence of cultural values and the tension between tradition and modernity in spatial behaviors. Yau et al. [101] proposed a practical, data-driven planning framework based on empirical data from Hong Kong universities, advocating for the optimization of spatial layouts to support self-directed learning. Similarly, Kumar and Bhatt [102] identified through surveys in Indian universities that students strongly preferred flexible spatial arrangements responsive to actual usage behaviors, particularly in resource-constrained contexts. These studies underscore the importance of context-specificity and suggest that future research should systematically explore how diverse socio-cultural and institutional factors influence ILS use and effectiveness across different countries.