A Multi-Layered, Progressive Model of Self-Driving Tourists’ Environmental Responsibility Behavior: Enriched Tourism Destination 6A Framework

Abstract

Amid growing environmental challenges associated with self-driving tourism, this study explores destination-level factors beyond the traditional 6A framework that influence tourists’ environmental responsibility behavior (ERB). Using a grounded theory approach supported by expert interviews, this study conducted 20 + 5 (theoretical saturation test) interviews with local government officials, academics and researchers, local tourism industry professionals, and local community representatives. The interview data underwent a three-stage coding process—open, axial, and selective coding. As a result, three additional drivers of ERB were identified: (1) governance capacity for sustainability, (2) green innovation practices, and (3) community-based environmental empathy. Together with the 6A framework, these drivers form a multi-layered, progressive model that explains how destinations shape ERB through three pathways: product and service experience, institutional regulation and technological enablement, and emotional connection as the deepest driver. The study enriches destination behavior theory by integrating tangible, managerial, and cultural mechanisms, offering theoretical advancement and practical strategies for promoting sustainable self-driving tourism.

1. Introduction

Addressing environmental challenges in tourism, particularly those posed by its rapid growth, is central to achieving the United Nations Sustainable Development Goals [1,2]. With modern tourism, self-driving tours have developed rapidly globally in recent years [3]. It is a travel model in which individuals or small groups use private cars as the primary means of transportation, autonomously planning travel routes, schedules, and activities, reflecting a pursuit of personalization and flexibility [4]. The Chinese government’s policy of free highways during major holidays and support for RV tourism have fueled urban residents’ preference for self-driving tours, making them a vital segment of China’s tourism market [5,6,7]. In 2019, it reached 3.84 billion person-times, accounting for 64% of China’s total number of tourists.

However, the environmental issues brought to tourist destinations have also become increasingly severe [8,9]. Self-driving tours pose significant threats to the environmental sustainability of tourist destinations, with numerous instances of ecological damage frequently publicized on online platforms. Xinjiang’s Duku Highway and the Sichuan–Tibet self-drive tourist routes have become increasingly contaminated with waste, while in Lijiang, automobile exhaust emissions have significantly exacerbated air pollution during the peak tourism season. This focus is evidence-led: Recent studies show that road transport is the dominant hotspot of tourism’s carbon footprint and that private car use by tourists contributes substantially to domestic tourism emissions, underscoring the policy salience of self-driving segments [10,11,12].

Given the mounting environmental pressures at popular destinations, authorities have adopted demand- and supply-side instruments to manage use and mitigate impacts. Managed-access tools—such as timed-entry systems and daily caps—disperse peak flows and protect fragile sites while retaining visitor support [13,14]). Waste governance has likewise shifted from ad hoc operations to coordinated, multi-actor networks that combine source separation, on-site collection, and circular practices embedded in destination operations (Xu, Luo, & Lai, 2024) [15]. The Chinese government has strengthened protection policies. On 18 October 2022, the Ministry of Culture and Tourism issued the Code of Conduct for Civilized Behavior in Tourist Attractions to promote civility and sustainability [16], followed by the Implementation Rules for the Lijiang Tourism Regulations (9 January 2025), requiring environmental protection and prohibiting entry into undeveloped areas [17]. Further, on 10 July 2025, three Yunnan provincial departments released the Notice on Regulating Tourism Activities in Nature Reserves in Yunnan Province, mandating visitor limits, waste management, and “Green Shield” patrols [18].

Despite these measures, sustainable tourism remains challenging [19]. According to the Global Sustainable Tourism Council, sustainable destination development relies on optimizing visitor experiences and reducing pressures through systematic planning [20], goals consistent with the tourism destination 6A framework (6A). The 6A model has become central in tourism research, particularly in interpreting tourist behavior [21,22,23], and significantly influences environmental responsibility behavior (ERB) [24,25,26]. However, while offering a comprehensive view of destination elements, it mainly emphasizes tangible and experiential services [27,28], overlooking deeper factors such as macro-management, innovation, and cultural permeation. Thus, the 6A framework alone cannot fully explain self-driving tourists’ ERB.

Additionally, existing research remains inadequate regarding the exploration of ERB among self-driving tourists as a specific group. Car-based trips often entail higher per capita emissions and distinct decision constraints compared with public transport users [10,12]. Second, car reliance remains salient in many destinations, making this segment both high-impact and managerially actionable [11]. This study intentionally focuses on self-driving tourists, whose high autonomy over routing, parking, and in-vehicle practices creates distinct intervention points at the destination. Other segments (e.g., public transport users, eco-tourists) operate under different mobility constraints and norms and thus require re-anchored levers rather than a different theory [10,11,12].

Building on the gaps identified above, this study pursues three objectives and related questions: (RQ1) to identify destination-side mechanisms beyond the 6A framework that influence self-driving tourists’ ERB; (RQ2) to theorize how these mechanisms integrate with 6A in a progressive, cross-level model; and (RQ3) to translate the model into actionable strategies for destination management.

Grounded theory was selected because the mechanisms beyond the 6A framework are under-theorized, and the method is well-suited for inductive model building [29,30]. This study contributes by extending the 6A model with governance, innovation, and community drivers. Practically, it offers strategies for destination managers to promote ERB among self-driving tourists.

2. Literature Review

2.1. Self-Driving Tours

Self-driving tours are trips in which individuals or small groups use private cars, independently planning routes, timing, and activities, reflecting personalization and flexibility [4]. In the 21st century, they have rapidly expanded in China, becoming a major tourism segment. Compared with traditional tourism, they allow flexible adjustment of destinations, durations, and schedules, offering richer experiences [9].

According to Mafengwo and its “Polaris Tourism Big Data Service System,” self-driving tours grew by 39% year-on-year in 2023, ranking first among all travel categories (Figure 1). However, this growth has brought serious environmental pressures, including waste, vehicle emissions, ecological degradation, and water pollution [31].

In 2022, small cars—the primary transport for self-driving tourists—recorded the highest pollutant emissions among passenger vehicles (Figure 2). Given these environmental challenges and tourists’ insufficient ERB, it is crucial to explore deeper influencing factors and construct a more comprehensive destination-based model to promote the sustainable development of self-driving tourism.

2.2. Environmental Responsibility Behavior

Environmental responsibility behavior (ERB) refers to actions aimed at reducing environmental impacts, such as conserving energy, cutting emissions, protecting heritage, waste sorting, and wildlife conservation [32,33,34]. In self-driving tours, ERB involves conscious practices to minimize damage, including proper waste disposal, avoiding littering, reducing noise, and saving energy [35]. Tourists may also use reusable utensils to cut plastic waste [36] or choose electric/hybrid vehicles to lower emissions [37]. ERB is especially critical in self-driving tourism, as it helps mitigate resource consumption and pollution from travel activities [38,39].

Over the past decade, scholars have extensively examined tourists’ environmental behaviors. Using CiteSpace (6.3), this study analyzed nearly 300 Web of Science articles on environmental protection and sustainable tourism (Figure 3 and Figure 4,

Table 1. Frequency ranking of keywords in the literature on tourists’ environmental protection.

Sort	Count	Centrality	Keywords
1	38	0.17	Sustainable development
2	38	0.09	Management
3	37	0.13	Sustainable tourism
4	25	0.14	Tourism
5	24	0.46	Conservation
6	15	0.04	Model
7	15	0.27	Protected areas
8	15	0.26	Attitudes
9	15	0.03	Satisfaction
10	14	0.05	Impact

Source: By authors, 2025.

). As of 2024, the leading research areas include sustainable development (38), management (38), sustainable tourism (37), tourism (25), and conservation (24), with keywords such as “ecotourism” and “sustainable tourism development” remaining prominent. This prominence is also visible in social media discourse on sustainable tourism [40]. Building on these trends, this study adopts a destination-focused perspective, applying the 6A framework and grounded theory to identify new factors shaping tourists’ ERB.

2.3. Tourism Destination 6A Framework

With the rapid growth of global tourism, demand has shifted from simple sightseeing to personalized, in-depth experiences [41,42], rendering traditional destination management inadequate. This necessitates a multi-dimensional tool to assess and enhance tourism experiences [43]. The 6A, proposed by Buhalis, offers such an integrated approach, addressing management complexity and diverse visitor needs, and thereby filling a critical gap in tourism research [23]. The 6A framework encompasses six core dimensions—attraction appeal, accessibility, amenities, ancillary services, activities, and available packages—highlighting their synergistic role in enhancing visitor experience and advancing destination development [21,44,45]. Figure 5 illustrates the six components of the 6A framework.

Attractions refer to a destination’s unique natural and historical features [46], such as traditions, landscapes, and archeological sites, which encourage self-driving tourists to protect heritage. Accessibility refers to the ease of reaching destinations [47]; for example, eco-friendly parking lots and EV charging facilities promote sustainable travel. Amenities denote public and complementary infrastructure [48], with waste separation stations or water-saving restrooms fostering responsible behavior. Ancillary services refer to supplementary offerings such as information, financial, and security services [49]; for instance, visitor centers with eco-route guides and conservation materials support sustainable practices. Activities include recreational, cultural, and natural engagements [49], such as volunteer programs or low-carbon driving challenges that raise environmental awareness. Available packages represent integrated combinations of services and experiences [50]; for example, self-driving packages with green lodging and conservation-focused tickets incentivize sustainable tourism.

The 6A significantly influence ERB [24,25,26]. Although the 6A framework for tourism destinations provides a comprehensive understanding of destination elements, it primarily concentrates on the product and service supply dimension, thus limiting its ability to fully address dimensions supporting sustainable development in tourism destinations. Previous research has revealed that, apart from elements included in the 6A framework, macro-level destination management, technological innovation, and social and cultural factors also significantly influence tourists’ ERB [51,52,53,54]. Hence, it is essential to identify new critical factors beyond the scope of the 6A framework when examining influences on ERB, to enrich and further refine the existing 6A framework for tourism destinations.

Taken together, Section 2.1, Section 2.2 and Section 2.3 indicate that self-driving tourism faces salient environmental pressures and gaps in ERB. Recent research trends have increasingly emphasized ecotourism and sustainable tourism development. The 6A framework clarifies experience/supply yet leaves the link to ERB enactment under-specified. Accordingly, our study asks which destination factors beyond the 6A shape ERB among self-driving tourists and how they interact. A grounded-theory design is adopted, using the 6A as the starting lens to inductively develop a destination-level model. The next section states the research questions and methods.

3. Materials and Methods

3.1. Research Design

This study employed an iterative grounded-theory design to construct a destination-level framework for self-driving tourists’ ERB. The flowchart shows the grounded-theory process: from defining the phenomenon, literature review, problem framing, and data collection, through three-level coding to theory building, with iterative data collection until saturation is reached (Figure 6).

3.2. Research Site

Lijiang City in Yunnan Province, China, is selected as the case study site for this research. Yunnan Province is located in southwestern China and is renowned for its unique natural landscapes and diverse ethnic cultures. According to the China Mafengwo 2023 Tourism Big Data Report: Self-Driving Tours, Yunnan ranks first among self-driving tour destinations in China (see

Table 2. The top 20 popular provinces for self-driving tours in 2023.

Rank	Province	Region	Rank	Province	Region
1	Yunnan	outhwest	11	Guangxi	South China
2	Guangdong	South China	12	Gansu	Northwest
3	Sichuan	Southwest	13	Jiangsu	East China
4	Beijing	North China	14	Hunan	Central China
5	Zhejiang	East China	15	Fujian	East China
6	Shandong	East China	16	Hubei	Central China
7	Inner Mongolia	North China	17	Hainan	South China
8	Shaanxi	Northwest	18	Jiangxi	East China
9	Xinjiang	Northwest	19	Shanxi	North China
10	Guizhou	Southwest	20	Chongqing	Southwest

Source: China Mafengwo 2023 Tourism Big Data Report: Self-Driving Tours (https://www.dsb.cn/234359.html, 27 September 2025).

). This ranking indicates that Yunnan, with its abundant tourism resources, convenient transportation network, and well-developed tourism service system, has become one of the most favored destinations for self-driving tourists nationwide.

Among Yunnan’s tourist destinations, Lijiang stands out for its cultural heritage and natural resources, making it the province’s most popular city. According to Lijiang.cn, the city received 68.08 million tourists in 2023, generating 130.177 billion yuan in revenue—up 24% and 91% year-on-year, respectively. Data from the China Mafengwo Tourism Big Data Research Center show Lijiang ranked first with 38,461 visitors, surpassing Kunming (33,627) and Dali (28,781) (

Table 3. The top 10 popular tourist destinations in Yunnan Province.

Rank	Location	Visitors
1	Lijiang	38,461
2	Kunming	33,627
3	Dali	28,781
4	Shangri-La	17,408
5	Shuhe	15,596
6	Xishuangbanna	8543
7	Shuanglang	7313
8	Diqing	6738
8	Tengchong	4846
9	Deqen County	4714
10	Chuxiong	4531

Source: China Mafengwo Tourism Big Data Research Center (https://www.mafengwo.cn/mdd/citylist/12711.html, 27 September 2025).

). In 2023, it was also named by the General Office of the State Council as one of ten national incentive cities for the cultural and tourism industries. These figures highlight Lijiang’s pivotal role in Yunnan tourism and justify its selection as this study’s case site.

Lijiang is a major self-driving destination in southwestern China, benefiting from its location in northwestern Yunnan at the junction of Sichuan, Yunnan, and Tibet, with convenient access from Kunming, Dali, and Shangri-La to nearby attractions such as Lugu Lake and Jade Dragon Snow Mountain. It is connected nationwide by expressways (G0613, G5611, G4216) and highways (G348, G353, G214), while Lijiang Sanyi International Airport further enhances accessibility (Figure 7).

Despite its appeal, rapid tourist growth has intensified environmental pressures. In the 2023 Ranking of Low-Carbon Development Performance in Chinese Cities, Lijiang ranked last among 100 cities. Key challenges include: (1) ecological vulnerability, with Jade Dragon Snow Mountain glaciers retreating and Lugu Lake ecosystems stressed by tourism; (2) inadequate waste management, as visitor surges strain disposal systems and some areas lack recycling; and (3) rising carbon emissions and traffic pollution from self-driving tours. To address these, the government introduced the Implementation Plan for Air Pollution Control in the Old Town District and the Implementation Plan for the Elimination of Yellow-Label Vehicles in the Old Town District, though further policy refinement is required for sustainability.

Lijiang’s distinctive tourism resources, supportive tourism policies, status as a popular destination for self-driving tourism, and the environmental challenges it currently faces collectively make it the selected case study site for this research. Following the logic of analytical generalization, Lijiang is treated as a theoretically informative case rather than a sampling frame for statistical inference. Such positioning aligns with recent guidance on transferability in qualitative research and the role of theory-oriented inference in case-based designs [55,56].

3.3. Data Collection Administration

This study adopts a semi-structured interview approach for qualitative data collection, chosen for its flexibility in exploring themes beyond predefined categories [57]. To capture diverse perspectives, 20 participants were recruited, evenly distributed among four groups: government officials, academic scholars, tourism practitioners, and community representatives. This balance avoids bias and provides a comprehensive view of destination-level influences. Tourists were excluded due to limited professional knowledge of the 6A framework and their role as subjects rather than sources of influence; their perspectives will be addressed in the quantitative phase through large-scale surveys. This sampling strategy follows grounded theory principles of theoretical sampling [29,30].

Table 4. Demographic and professional background of expert interviewees (n = 20).

Interviewee Type	Number	ID	Professional Role	Average Years of Experience	Key Contributions to Research
Government Officials/Tourism Management	5	1–5	Director/Bureau Chief Community-Level Personnel	15 years (10–25)	Macro-level policies, regulations, inter-departmental coordination, and governance capacity.
Academic Scholars/Researchers	5	6–10	Professor/Research Fellow	12 years (8–20)	Theoretical insights, environmental psychology, sustainable tourism trends, and behavioral economics.
Tourism Industry Practitioners	5	11–15	Operator/Manager/Supervisor	10 years (7–18)	Industry practices, technological adoption, business strategies, market feedback, and operational challenges.
Local Community Representatives	5	16–20	Committee Head/Village Elder/NGO Leader/Local residents	20 years (15–30)	Local culture, resident behavior, community initiatives, traditional ecological wisdom, and social dynamics.
Total	20

Source: Authors’ fieldwork data, compiled by the authors.

presents participant details. Purposive sampling was used to select experts in tourism management and environmental sustainability, supplemented by snowball sampling, where initial interviewees recommended other qualified participants. The interview guide is provided in the Appendix A.

Qualitative research often prioritizes thematic saturation over large sample sizes. Based on previous research recommendations [58], 20–30 expert interviews are expected to be sufficient to achieve saturation. The final sample size will depend on when new themes stop emerging from the interviews. In the initial phase of the study, 20 participants were interviewed, and an additional 5 interviews were conducted during the later phase of the coding process to assess theoretical saturation. Before the interviews, participants received a briefing document and guiding questions, and the researcher engaged in pre-interview communication to clarify the study’s scope. These preparations allowed interviewees—who possessed strong professional expertise—to respond concisely and substantively. These preparatory tasks were carried out with great rigor and were sustained over an extended period. Semi-structured interviews were conducted either face-to-face or via online platforms (e.g., Zoom, Tencent Conference, WeChat). Each interview lasted approximately 30 min. All interviews were audio-recorded with participants’ consent and transcribed for analysis.

3.4. Data Analysis

Grounded theory emphasizes the development of theory from empirical data rather than the application of predefined theoretical models, which is closely aligned with the exploratory aim of this research to identify novel influencing factors [29,30]. The study will follow the three-stage coding procedure of grounded theory—open coding, axial coding, and selective coding—to analyze the interview data. To ensure the systematic rigor and traceability of the coding process, all interview recordings were transcribed verbatim and imported into NVivo 14. In addition, the completeness of the data analysis was validated through a theoretical saturation assessment.

To ensure the reliability of coding, the data analysis followed a process of iterative cross-checking and consensus validation. Codes were repeatedly compared, refined, and discussed until agreement was reached. This procedure enhanced intercoder reliability and minimized subjectivity in the interpretation of qualitative data [59].

4. Results

The coding process follows grounded theory procedures: open coding (600 initial concepts condensed into 30 codes), axial coding (three categories: B1 Governance, B2 Innovation, B3 Community), and selective coding (three core categories: C1, C2, C3). This process illustrates methodological transparency and theoretical rigor (Figure 8).

4.1. Open Coding

Open coding involves conducting a line-by-line analysis of the interview transcripts to extract concepts relevant to the research topic and categorize them as initial codes [59]. In this study, approximately 600 initial concepts were identified from 20 interview transcripts.

This word cloud visualizes the results of the open coding stage, highlighting high-frequency concepts derived from the interview data (Figure 9). Prominent terms such as “Strict Environmental Regulations”, “Joint Law Enforcement Mechanism”, “Household Waste Sorting”, “Smart Tourism App”, and “Community Collaboration” reveal the central themes emerging from expert perspectives. The visualization demonstrates the richness of the qualitative data and provides a foundation for subsequent axial and selective coding.

After merging similar items, 30 distinct open codes were generated (labeled A1 to A30). To improve readability, detailed coding tables (open, axial, selective) are relocated to Online

Table A1. Summary Table of Open-Coding Results.

Code	Open Code Label	Frequency	Supporting Excerpt	Interviewer ID
A1	Strict Environmental Regulations	18	“We have issued stricter vehicle restrictions and emission standards, equipped with smart monitoring and enforcement teams.”	1
A2	Joint Law Enforcement Mechanism	15	“We have established a joint enforcement mechanism with public security, transport, and tourism departments to jointly penalize serious environmental violations.”	2
A3	Tourist Credit Record	12	“Some serious violations are included in individuals’ tourism credit records, affecting their access to travel services in Lijiang.”	1
A4	Ecological Red Line Designation	10	“We have designated ecological red lines across the entire Lijiang region, clearly defining no-development zones.”	5
A5	Environmental Carrying Capacity Assessment	9	“We issue real-time alerts and divert visitor flow based on tourist volume to avoid excessive environmental pressure.”	4
A6	Multi-Stakeholder Collaboration	14	“We regularly hold environmental protection forums involving tourism enterprises and community residents to jointly develop action plans.”	1
A7	Ecological Compensation Mechanism	8	“For ecologically impacted areas due to tourism development, developers are required to conduct restoration or provide compensation.”	5
A8	Smart Tourism App	16	“The official tourism app provides real-time traffic updates and air quality index, and recommends green travel routes.”	1
A9	New Energy Charging Stations	13	“We piloted smart charging stations for new energy vehicles along popular self-driving routes and scenic entrances.”	1
A10	Smart Waste Sorting Systems	11	“Smart sorting equipment automatically recognizes garbage types and issues alerts based on waste volume.”	1
A11	Environmental Sensors	9	“We piloted the installation of smart sensors to monitor waste, water quality, and noise levels in real time.”	2
A12	AR-Based Environmental Tasks	8	“We designed AR-based environmental tasks where tourists can scan signs to learn about ecology and complete challenges.”	4
A13	Carbon Credit System	7	“We piloted a blockchain-based carbon credit system to encourage low-carbon travel and green activities.”	5
A14	Smart Parking System	10	“We piloted a smart parking management system allowing for advance reservations to reduce unnecessary idling and emissions.”	3
A15	Eco Navigation Alerts	9	“In collaboration with navigation companies, we marked ecologically sensitive areas and no-parking zones on maps.”	2
A16	Naxi Ecological Wisdom	17	“Naxi cultural values embody harmony between humans and nature, as reflected in everyday life.”	3
A17	Household Waste Sorting	15	“Residents in the old town sort waste meticulously, with some families composting food scraps themselves.”	1
A18	Community Environmental Role Models	14	“Tourists feel embarrassed to litter after witnessing residents’ strong environmental habits.”	1
A19	Eco-Friendly Homestays	12	“Eco-friendly homestay demonstration sites run by locals proactively share environmental stories with guests.”	4
A20	Cultural Experience Activities	11	“Visitors are invited to participate in traditional farming activities to learn about ecological agriculture.”	1
A21	Emotional Resonance	13	“When tourists see how much locals cherish their homeland, it evokes strong emotional resonance.”	2
A22	Ecological Interpreters	10	“Local residents act as ecological interpreters, sharing stories about land protection with visitors.”	5
A23	Environmental Storytelling	9	“Guesthouse owners tell stories about protecting the snow mountain, which inspires tourists’ sense of responsibility.”	1
A24	Volunteer Guidance	8	“Volunteers guide tourists in waste sorting and urban managers stop littering behaviors.”	1
A25	Green Cultural Products	7	“We promote handicrafts made from recycled materials that convey environmental values.”	4
A26	Environmental Point Incentives	8	“Tourists can earn points for eco-friendly behaviors and exchange them for souvenirs.”	11
A27	Leave-No-Trace Camping	7	“We emphasize the leave-no-trace concept to reduce environmental footprints during camping.”	11
A28	Eco-Driving Challenge Events	6	“We organize eco-driving challenges that record fuel-saving and waste-sorting performance.”	13
A29	Environmental Training	8	“The government conducts environmental training to communicate policy and technical requirements to practitioners.”	11
A30	Blacklist System	7	“Tourists who damage the environment are blacklisted and restricted from re-entering parks.”	15

Source: Authors’ coding results based on interview transcripts.

,

Table A2. Summary Table of Axial Coding Results.

Axial Code	Included Open Codes	Logical Relationship Explanation	Frequency
B1: Policy, Regulation, and Collaborative Governance	A1, A2, A3, A4, A5, A6, A7, A24, A29, A30	Through strict regulations, joint law enforcement, and multi-stakeholder collaboration, an institutional constraint and social supervision system is formed to regulate tourist behavior.	101
B2: Smart Technology and Green Innovation	A8, A9, A10, A11, A12, A13, A14, A15, A26, A27, A28	By leveraging smart technologies and incentive mechanisms, the cost of engaging in ERB is reduced, thereby enhancing tourist participation.	94
B3: Community Culture and Emotional Resonance	A16, A17, A18, A19, A20, A21, A22, A23, A25	Through cultural transmission, local resident role modeling, and emotional interaction, tourists are inspired to resonate with and internalize environmental values.	103

Source: Authors’ coding results based on interview transcripts.

and

Table A3. Summary Table of Selective-Coding Results.

Selective Code	Included Axial Codes	Included Open Codes	Relationship Explanation
C1: Destination Governance Capacity for Sustainability	B1	A1, A2, A3, A4, A5, A6, A7, A24, A29, A30	Through regulatory constraints, multi-stakeholder collaboration, and social oversight, a systematic governance structure is established to regulate tourist behavior.
C2: Green Innovation Practices at the Destination	B2	A8, A9, A10, A11, A12, A13, A14, A15, A26, A27, A28	By leveraging smart technologies and incentive mechanisms, the cost of ERB is reduced, enhancing tourists’ willingness to participate.
C3: Community-Based Environmental Empathy	B3	A16, A17, A18, A19, A20, A21, A22, A23, A25	Through cultural transmission, resident role modeling, and emotional interaction, tourists are inspired to resonate with and internalize environmental values.

Source: Authors’ coding results based on interview transcripts.

.

4.2. Axial Coding

Axial coding builds upon open coding by integrating related concepts into higher-order categories centered on core themes, thereby revealing the underlying logical relationships among codes [60]. This study identifies three primary axial codes: Policy, Regulation, and Collaborative Governance (B1), Smart Technology and Green Innovation (B2), and Community Culture and Emotional Resonance (B3). Detailed axial-coding mappings are provided in Table A2.

Axial Code B1 (Frequency: 101) highlights the pivotal role of Lijiang in institutional and governance systems. This category reflects how the city has established a systematic environmental management framework through regulatory enforcement, interdepartmental coordination, and social oversight—effectively shaping self-driving tourists’ ERB through external constraints. For example, “tourists who damage the environment are blacklisted and restricted from re-entering parks.” (ID-15)

Axial Code B2 (Frequency: 94) underscores Lijiang’s commitment to technological empowerment and innovation-driven practice. This category demonstrates how the integration of smart technologies and gamified incentive mechanisms lowers the barriers to ERB and enhances tourists’ willingness to engage. For example, “the official tourism app provides real-time traffic updates and air quality index, and recommends green travel routes.” (ID-1)

Axial Code B3 (Frequency: 103), the most frequent among the three, accentuates the distinctive influence of local culture and community in fostering pro-environmental behavior. This category shows how Lijiang cultivates environmental empathy and intrinsic motivation among tourists through cultural heritage transmission, exemplary resident behavior, and emotional resonance. For example, “when tourists see how much locals cherish their homeland, it evokes strong emotional resonance.” (ID-2)

4.3. Selective Coding

Selective coding integrates axial codes around core theoretical categories to build a comprehensive theoretical framework [61]. This study identifies three core categories: C1, C2 and C3. Detailed selective-coding synthesis and the full codebook are provided in Table A3.

Selective coding integrates the influencing factors of self-driving tourists’ ERB in Lijiang into three core categories—C1: Destination Governance Capacity for Sustainability, C2: Green Innovation Practices at the Destination, and C3: Community-Based Environmental Empathy—corresponding to institutional constraint, technological empowerment, and cultural engagement, together forming a multi-level, multi-stakeholder framework.

C1: Destination Governance Capacity for Sustainability (B1) highlights institutional mechanisms shaping ERB through strict regulations (A1, A2), social oversight (A3, A30), and multi-actor collaboration (A6, A24). Measures such as the ecological red line (A4) and carrying capacity assessments (A5) illustrate proactive governance. This aligns with perceived behavioral control theory, where external constraints strengthen responsibility [62]. However, governance gaps in remote areas (Interviewee 20) reveal the need for more precise management.

C2: Green Innovation Practices at the Destination (B2) emphasizes behavioral facilitation via smart technology and infrastructure, including digital guidance (A8, A14), charging and waste systems (A9, A10), and gamified incentives (A12, A26). Tools like the Smart Tourism App (Interviewee 1) and environmental point programs (Interviewee 11) reduce costs and enhance motivation, supported by carbon credit systems (A13) and AR tasks (A12). This reflects transaction cost reduction theory, where convenience drives behavior [63]. Yet, effectiveness depends on user acceptance and design (Interviewee 7).

C3: Community-Based Environmental Empathy (B3) draws on Naxi ecological wisdom (A16), role models (A17, A18), cultural activities (A20), and emotional engagement (A21, A23) to build intrinsic motivation. Examples include waste sorting, environmental storytelling, eco-homestays (A19), and green cultural products (A25), extending influence into tourist consumption. This aligns with collective consciousness theory, where role modeling fosters internalization [64,65]. Nonetheless, transferability is limited, as cultural influence depends on local depth (Interviewee 7).

These three core categories operate in a dynamic, complementary relationship: C1 establishes institutional boundaries through regulatory control, C2 empowers action via technological innovation, and C3 sustains behavior through cultural and emotional engagement. Together, this tripartite mechanism—regulatory constraint, technological facilitation, and cultural resonance—forms the theoretical foundation for understanding and promoting ERB among self-driving tourists in the context of sustainable tourism development in Lijiang.

Traditionally, analyses of destination influence on tourist behavior have centered on core products and services—most notably articulated through the widely adopted 6A framework. This framework has provided a foundational lens for understanding tourists’ perceptions and satisfaction with the basic experience layer of a destination, representing both the tangible and intangible carriers of tourism experiences [66,67]. However, qualitative findings from this study suggest that the 6A framework alone is insufficient to explain the ERB exhibited by self-driving tourists in Lijiang. Comparison between the traditional 6A framework (attraction, accessibility, amenities, ancillary services, activities, available packages) and the extended model developed in this study, which incorporates three additional categories (C1, C2, C3) (Figure 10).

Based on the preceding coding results, this study integrates the current theoretical positioning and future development trajectory of the 6A framework to construct A Multi-Layered, Progressive Model of Destination Influence on Self-Driving Tourists’ ERB (Figure 11).

Layer 1: Basic Experience Layer. Experiential Triggers through Product and Service Perception.

This foundational layer comprises the core elements of the 6A framework, which represent the tourism products and services that tourists directly experience. These elements shape tourists’ immediate satisfaction and travel experience and indirectly influence their environmental attitudes and behavioral intentions. For instance, convenient environmental facilities such as trash bins can reduce the physical effort required for environmentally responsible actions. Illustrative quotes: “Smart sorting equipment automatically recognizes garbage types and issues alerts based on waste volume.” (ID-1); “In collaboration with navigation companies, we marked ecologically sensitive areas and no-parking zones on maps.” (ID-2)

Layer 2: Macro-Governance and Innovation Layer: Institutional Regulation and Technological Enablement.

At this intermediate layer, C1 and C2 jointly shape ERB by setting normative boundaries and lowering participation costs. C1 captures Lijiang’s institutional governance—regulatory enforcement, inter-departmental coordination, and risk prevention—structuring tourists’ behavior via vehicle restrictions, joint enforcement, environmental credit records, and multi-agency cooperation. C2 reflects digital and innovative facilitation—eco-route recommendations, EV charging deployment, and interactive carbon incentives—making pro-environmental choices convenient and attractive. This layer thus couples external regulation (C1) with technical facilitation (C2), stimulating intrinsic motivation while “greening” 6A operations (e.g., smart waste systems strengthen amenities; green transport infrastructure enhances accessibility). Illustrative quotes: “We have issued stricter vehicle restrictions and emission standards, equipped with smart monitoring and enforcement teams.” (ID-1); “We piloted a smart parking management system allowing for advance reservations to reduce unnecessary idling and emissions.” (ID-3)

Layer 3: Deep Cultural and Social Driving Layer: Emotional Connection as the Deepest Driver.

This core layer activates intrinsic motivation through cultural identification and community-based empathy, i.e., C3. Rooted in Lijiang’s ethnic traditions and resident practices, C3 operates via role modeling, cultural transmission, and emotional engagement, exposing tourists to implicit norms that trigger personal environmental ethics. Illustrative quotes: “Naxi cultural values embody harmony between humans and nature, as reflected in everyday life.” (ID-3); “Guesthouse owners tell stories about protecting the snow mountain, which inspires tourists’ sense of responsibility.” (ID-1). Emotional resonance and moral identification foster the internalization of responsible behavior, turning it into a self-sustained part of the tourist’s value system. This cultural resonance provides the psychological and social grounding for the regulatory and technical mechanisms of the upper layers, ultimately encouraging voluntary, durable engagement in ERB.

4.4. Theoretical Saturation Test

To ensure the robustness of the theoretical framework, this study conducted a theoretical saturation assessment to verify the sufficiency of the data analysis (

Table 5. Theoretical saturation test table.

Testing Stage	Data Source	New Codes Identified	Frequency Change	Results
Initial Coding (Participants 1–20)	20 interviews	A1–A30	Total frequency: 600	Formed 30 open codes, 3 axial codes, and 3 core categories
Supplementary Coding (Participants 21–25)	5 interviews	None	Approx. 15% increase (e.g., A1 + 2, A8 + 3, etc.)	No new codes emerged; existing category relationships remained stable
Category Relationship Review	Reanalysis of the entire dataset	None	No change in logical relationships	Complementarity among core categories was further reinforced

Source: Authors’ compilation based on additional interviews for saturation test.

). The assessment involved:

(1)

Adding five additional interviews during the later stage of coding (participants 21–25, including government officials, academics, industry practitioners, and local residents) to assess whether any new conceptual codes emerged.

(2)

Reanalyzing the frequencies and interrelationships of existing codes to confirm structural stability and coherence among the categories.

The results of this assessment are presented below.

The theoretical saturation assessment confirmed that the initial set of 20 interviews had sufficiently identified the key factors influencing ERB among self-driving tourists, resulting in a well-established theoretical framework. The five additional interviews did not produce any new codes but led to modest increases in the frequency of existing ones (e.g., A1 and A8 increased by approximately 15%). Importantly, no changes were observed in the logical structure of the axial and selective coding. For instance, interviewee 21 (a government official) emphasized “strengthening vehicle restriction policies,” which aligns with code A1, while interviewee 23 (a tourism practitioner) mentioned the app’s “green route recommendation,” corresponding to code A8. These results indicate that the existing coding system has reached theoretical saturation, and the core categories C1, C2, and C3, along with their interrelationships, provide a robust explanation of the mechanisms shaping ERB in the context of self-driving tourism in Lijiang. Taken together, the absence of new codes and the stability of inter-category relations function as a pragmatic, internal validation of the model’s structure [68].

5. Discussion

Governance capacity establishes behavioral boundaries and reduces ambiguity, thereby enhancing perceived responsibility [69,70]. Destination governance capacity (C1) aligns with the concept of perceived behavioral control in the Theory of Planned Behavior, where external constraints and regulatory systems strengthen tourists’ sense of responsibility to act pro-environmentally [62]. Green innovation practices lower transaction costs and embed pro-environmental choices into tourists’ default routines [71,72]. Green innovation practices (C2) resonate with the theory of transaction cost reduction, emphasizing that technological facilitation and incentive structures reduce participation barriers and encourage behavioral change [58]. Community-based empathy mobilizes emotions and role modeling, promoting the internalization of norms [73,74]. Community-based environmental empathy (C3) corresponds to the sociological theory of collective consciousness, showing how cultural identity and role modeling foster emotional resonance and the internalization of pro-environmental norms [64,65]. Together, these layers explain not only which factors matter, but how they interact to generate persistent ERB.

While the model recognizes community-based empathy (C3) as a deep, durable driver, it does not downplay pragmatic motivations. Our data show that many ERB actions are enacted when cost, time, and hassle are reduced or when small incentives are present—mechanisms captured by C2 green innovation (e.g., default options, app prompts, gamified rewards) and C1 policy mixes (e.g., operator pricing and capacity management). Recent evidence likewise indicates that economic incentives and low-friction design can shift sustainable mobility and recycling behaviors in tourism and related settings [70,75,76]. Flashback and app-based nudges further sustain behaviors beyond the trip [71,72]. Thus, affective–normative cues complement, rather than replace, instrumental levers.

While C3 (community-based environmental empathy) mobilizes emotions and role modeling, destination communities are heterogeneous. Divergent interests among residents, SMEs, platform operators, and authorities may generate explicit or implicit conflicts that dampen empathy effects. Power asymmetries and perceived distributional unfairness can erode trust and reduce tourists’ exposure to consistent cues. Conflict-sensitive, participatory processes and clear benefit-sharing rules help stabilize community-anchored mechanisms of ERB [77,78,79,80].

5.1. Theoretical Contributions

To transition from empirical patterns to theory-building, we first clarify the study’s theoretical contributions and the cross-level mechanisms that underpin them.

(1)

Extending the 6A lens from triggers to a layered process. We position the classic 6A framework as the Basic Experience Layer that provides proximal experiential triggers but theorize that ERB becomes robust only when macro-governance structures and community-level culture/empathy are engaged. This reframes ERB formation from single-layer stimuli to a progressive, multi-layer process [69,74].

(2)

Articulating cross-level mechanisms and sequencing. We specify a three-step pathway—regulatory clarity → technological enablement → cultural internalization—that explains how destinations convert momentary pro-environmental intentions into enacted and persistent behaviors [29,70,71]. This sequencing operates through two complementary routes: an instrumental route that reduces price/friction via C1–C2, and a normative–affective route consolidated by C3. This sequencing aligns with recent findings on policy-supported PBC, low-friction digital nudges, and community contagion of ERB [69,70,71,73].

(3)

Boundary conditions for transferability. We propose that the effects are stronger under higher governance maturity, higher digital readiness, and stronger community cohesion. Recent studies corroborate that policy regimes and affect-laden experiences (e.g., awe) condition PEB formation [70,81]. These propositions invite future comparative tests across destination types and regulatory environments. Additionally, the influence of C3 is likely contingent on stakeholder alignment and perceived fairness within communities. Heterogeneous interests and power asymmetries can moderate empathy-driven pathways to ERB [77,80].

5.2. Practical Implications

The practical implications of this study are particularly relevant to China’s tourism governance context. For instance, the 2025 Implementation Rules for the Lijiang Tourism Regulations emphasize ecological protection and restrictions on entering undeveloped areas, while the Yunnan Provincial “Green Shield” initiative highlights visitor capacity management and waste control. The model developed in this study provides a theoretical rationale for these measures by showing how governance capacity, technological innovation, and community empathy interact to shape ERB. Beyond Lijiang, the framework applies to other self-driving destinations facing similar environmental challenges, such as Xinjiang’s Duku Highway or Sichuan–Tibet routes. This situates the findings within both the local and national contexts of sustainable tourism development.

Translating these mechanisms into management action, we next outline the practical implications for destination governance, technological enablement, and community engagement.

(1)

Governance and capacity management. Destinations should pair clear rules with graduated enforcement (e.g., on-site guidance + soft penalties for first offenses) and implement capacity limits/time-phased reservations on sensitive sites. Such policy mixes are shown to narrow the intention–behavior gap and stabilize ERB [70].

(2)

Technological enablement (“design for low-effort green”). Deploy digital nudges in navigation and booking flows (default green routes/parking, refill prompts, waste-sorting cues), and use post-trip flashback nudges via chatbots to sustain spillover behaviors at home [71,72]. These tools specifically target friction and attentional limits that commonly block ERB enactment. Pair digital design with light economic instruments—deposit–refund, time-of-day/proximity parking fees, and micro-rewards in apps—to shift behavior with low burden [70,75,76].

(3)

Community-anchored empathy. Support programs where residents model visible PEB (e.g., attendant-led sorting, local volunteer crews) and micro-narratives that humanize ecosystem impacts; both mechanisms heighten empathy and pro-environmental atmosphere, which in turn lift tourists’ ERB [73,74]. Ethical guardrails (privacy, transparency, opt-out) should accompany all digital interventions. On the other hand, map stakeholders and salience; co-design programs with residents and SMEs; specify benefit-sharing and rotate volunteer workloads; and establish low-friction grievance and feedback channels. Such participatory and meta-governance measures help manage conflict and sustain community-anchored empathy [78,79].

5.3. Limitations and Future Research

As a single-site grounded-theory study in Lijiang, our claims target analytical generalization; external applicability depends on the stated scope conditions rather than statistical representativeness [55,56]. To assess portability, we encourage multi-site theoretical replication across destination types [82]. Future studies are encouraged to pursue the following directions:

First, building upon the multi-layered driving model developed in this study, future research can formulate testable hypotheses and employ quantitative methods—such as large-scale questionnaire surveys and Structural Equation Modeling—to empirically examine the influence of each hierarchical factor and its underlying variables on self-driving tourists’ ERB. This approach can also help quantify the relative importance of each factor. It should be acknowledged that this study did not include real tourist participants. The exclusive reliance on expert interviews may constrain the generalizability of findings to actual tourist behavior. Future research should combine expert perspectives with large-scale real and professionally knowledgeable tourist surveys or in-depth case studies to validate and expand the present framework.

Second, future research should conduct cross-segment replications—extending tests beyond self-driving tourists to public-transport users and ecotourists—and across diverse destination types (e.g., urban, coastal, alpine/snow) and stages of tourism development, to evaluate the model’s applicability, generalizability, and boundary conditions.

Third, comparative analyses can be conducted to explore how destination-level factors influence ERB across different cultural contexts, such as between domestic and international self-driving tourists. Such studies could deepen the understanding of cultural variability in environmental responsibility, motivations, and responses.

Finally, qualitative grounding enables theory building, yet future work should test the sequencing proposition (governance → enablement → empathy) and the moderating roles of governance maturity, digital readiness, and community cohesion across destination types. Multi-method designs that combine field experiments (for nudges) with comparative case studies (for governance) would help adjudicate causality and generalizability [70,71].

6. Conclusions

This study pursued three objectives and related questions. Addressing RQ1. Grounded-theory analysis identified three destination-level drivers beyond 6A—C1 governance capacity, C2 green innovation, and C3 community-based environmental empathy. Addressing RQ2. The 6A framework is positioned as the Basic Experience Layer. A clear sequence is specified—governance → technological enablement → community empathy—that turns intention into enacted and persistent ERB. Addressing RQ3. A concise agenda follows: (i) policy mixes with capacity management; (ii) digital nudges and supportive infrastructure; and (iii) community programs that model and narrate environmental care.

This study synthesizes destination influence on ERB as a layered mechanism: 6A provides experiential triggers; C1–C2 lower intention–action frictions; C3 consolidates norms. The governance → enablement → empathy sequence and stated scope conditions define when and how the model transfers. Practically, we implicate that managers in tourism department should pair clear, capacity-based rules with low-effort digital defaults and real-time prompts and make resident role-modeling and micro-narratives visible through fair benefit-sharing. These steps operationalize the layered mechanism with minimal disruption to visitor experience.

While derived from Lijiang, the model is posed for analytical generalization under explicit scope conditions and to be examined via multi-site theoretical replication [55,82]. Therefore, as a single-site grounded-theory study using expert interviews, claims target analytical generalization; future work should conduct multi-site/large-sample tests and field/natural experiments to examine layer weights, the proposed sequence, and boundary conditions.