SOR model is an important framework for understanding the antecedents, interventions and outcomes of tourism activities in environmental psychology (Choi and Kandampully, 2019; Fakfare et al., 2024). According to the causal chain of SOR model, environmental stimuli affect the behavioral intention by affecting the cognitive and emotional states of the organism (Eroglu et al., 2001; Robert and John, 1982). Stimuli include internal and external environmental cues (Peng and Kim, 2014; Luqman et al., 2017), while organism refers to an individual’s cognitive and emotional states (such as pleasure, arousal and dominance), which mediate between stimuli and responses (Lin et al., 2023). The response was manifested as either positive approach behavior or negative avoidance behavior (Arora et al., 2020).

SOR model is widely applied in individual behavior research in heritage tourism, nature tourism, festival tourism and other fields (Hsu et al., 2021; Huang and Bu, 2022; Yang et al., 2022). On this basis, this study also applies SOR model to explain how tourists’ perceived destination restorative qualities affect their cognitive, emotional and behavioral intentions under the background of rural tourism. When tourists perceive the environment as restorative, this perception acts as a psychological stimulus, triggering their cognitive and emotional responses (Li and He, 2023). Therefore, this study considers the perceived destination restorative qualities as a stimulus, while traveler’s wellbeing and destination image are viewed as organic factors. Wellbeing is a comprehensive emotional experience, reflecting the overall emotional state of tourists during tourism, including hedonic experience and eudaimonic experience, both of which play an intermediary role in tourists’ behavior choices under environmental stimulation (Lee, 2024). Destination image represents tourists’ overall cognition and evaluation of the culture, landscape and services of a place, which not only affects their emotional response, but also further influences their behavioral intention (Jin et al., 2015; Kim, 2018). Finally, reactions focus on loyalty because reactions are expressed both in behavior and changes in attitude (Suhartanto et al., 2020).

ART provides a solid theoretical and empirical basis for the analysis of individual restorative perception. According to ART, people must focus their attention to avoid distractions when dealing with various challenges in life, a process that relies on directed attention mechanisms (Kaplan and Kaplan, 1989). However, prolonged, high-intensity use of directed attention can lead to fatigue, so people need to recover attention in a restorative environment (Huang et al., 2022). Based on ART, Hartig et al. (1997) proposed the perceptual restorative scale for the first time, evaluating the restorative quality of perceived destination through four dimensions: fascination, being away, extent and compatibility. This framework has been further improved and verified by many scholars, and the dimensions of restorative perception have been expanded in different contexts (Bagot, 2004; Felsten, 2009; Cole and Hall, 2010). Zhai et al. (2023) further proposed that restorative originates from the interaction between physical and social factors, adding the dimension of “life atmosphere” and expanding the original four-dimensional scale. Specifically, being away means withdrawing from an environment that causes fatigue and loss of concentration (Cho et al., 2016). Fascination is the ability to easily attract attention to an environment (Scopelliti and Vittoria Giuliani, 2004). Compatibility means that the environment fits with personal preferences and brings a sense of belonging (Herzog et al., 2003). Extent is related to the richness of environmental resources (Cole and Hall, 2010). And Life atmosphere captures the mood and spiritual character derived from local production and living conditions. It is defined by two key components: friendliness, which reflects tourists’ subjective impressions of the warmth and approachability in local interpersonal interactions, and vitality, which denotes the direct perception of local residents’ lifestyles and conditions. The interplay of these elements fosters a sense of social relaxation, thereby promoting restoration (Zhai et al., 2023). This five-dimension scale provides a powerful measurement tool for assessing the restorative perception of the rural tourism environment.

Wellbeing is a comprehensive concept that covers the subjective evaluation of overall life satisfaction, satisfaction in various areas, emotional and social wellbeing, and the evaluation of psychological resources (Huang et al., 2024). According to different philosophical origins, wellbeing can be divided into hedonic experience and eudaimonic experience (Lee, 2024). Hedonic experience mainly focuses on the emotional response or pleasure brought by tourism products, which usually reflects short-term emotional benefits (Wang et al., 2020). In contrast, eudaimonic experiences are more focused on mental health and quality of life enhancement, and are often related to self-eudaimonic, human development, personal purpose, and virtue (Sirgy and Uysal, 2016). Lengieza et al. (2019) pointed out that in current tourism research, hedonic experience is mainly reflected in two dimensions of pleasure and avoidance, while eudaimonic experience includes two dimensions of personal meaning and self-reflection.

Highly restorative environments, such as natural landscapes, quiet parks or comfortable leisure areas, can effectively relieve individual fatigue and stress and promote emotional recovery (Chen et al., 2024). Baldwin et al. (2021) pointed out that individuals exposed to the natural environment are more likely to experience emotional recovery and enhancement of positive emotions, thus enhancing their hedonic experience. In the context of developing countries, Zhang et al. (2024) found that in suburban national forest parks, there was a positive correlation between perceived restorativeness and general subjective wellbeing (i.e., hedonic experience).

According to ART, natural and restorative environments can help restore focus, reduce mental fatigue, and promote deep thinking (Hartig et al., 2003; Kaplan and Kaplan, 1989). These environments provide a space for individuals to relax, think and reflect, helping to re-examine their own goals and meaning in life, thereby enhancing their sense of wellbeing (Huang et al., 2022). Furthermore, Løvoll et al. (2020) point out that there is a significant positive correlation between tourists’ positive perception of the destination environment, such as beauty, comfort and naturalness, and their hedonic and eudaimonic feelings. In summary, this paper proposes the following hypotheses:

Destination image is usually defined as a tourist’s psychological perception of the knowledge, attributes and overall impression of a tourist attraction (Fakeye and Crompton, 1991). Chi and Han (2021) pointed out that a highly restorative environment can not only improve tourists’ mood and psychological state, but also enhance their favorable impression of the destination. These positive emotional experiences are crucial to shaping a positive image of the destination (Joo et al., 2023). At the same time, Prayag et al. (2017) analyzed the relationship between emotional experience and destination image, and proposed that restorative environment can improve tourists’ overall destination image. Based on the above research, this study proposes the following hypothesis:

In tourism research and development, the idea of improving tourist wellbeing does not conflict with the goal of improving destination image (Aureliano-Silva et al., 2017). The pleasure and emotional satisfaction obtained by tourists during the travel process will improve their overall evaluation and cognition of the destination (Chi et al., 2022). This positive emotional experience reinforces the image of the destination in the minds of tourists, increases their willingness to revisit, and influences the perception of others about the destination through word of mouth (Joo et al., 2023). As Kim and Fesenmaier (2015) pointed out, the happiness and satisfaction experienced in tourism directly affect tourists’ cognition and attitude toward the destination image.

Visitors tend to have a deeper emotional connection with the destination when they gain a sense of personal growth and meaning (Ali et al., 2014). This deep experience not only deepens the tourists’ positive evaluation of the destination, but also further enhances the overall awareness of the destination’s image (Liu and Huang, 2020). Pung and Chiappa (2020) pointed out that the transformational travel experience promotes tourists’ eudaimonic experience, thus enhancing the perceived image of the destination. Similarly, Kuykendall et al. (2015) explored the role of self-actualization and psychological satisfaction in the travel experience, finding that these factors have a significant positive impact on the image of a destination, especially in the context of deep cultural and natural experiences. Therefore, this paper proposes the following hypotheses:

Loyalty is defined as an individual’s willingness and behavior to continue to use, consume, or patronize a product, service, or destination in the future (Oliver, 1999). In the process of travel, if tourists obtain higher subjective wellbeing, their sense of pleasure will increase, thus forming a higher loyalty (Wang et al., 2020). Balleyer and Fennis (2022) showed that hedonism is the key factor affecting people’s purchase of tourism and leisure products. At the same time, Lee (2024) pointed out that eudaimonic experience can help tourists better understand themselves, the world and their position in the world, so as to enhance their willingness to revisit and recommend the place. Further research also supported this view, and Al-okaily et al. (2023) found that achieving wellbeing has a direct impact on tourists’ destination loyalty. Keskin et al. (2024) showed that in the context of food festivals, high levels of hedonic experience and eudaimonic experience can enhance the influence of loyalty on behavioral intention. In summary, this paper proposes the following hypotheses:

A sound destination image can attract new visitors, and also cultivate loyal customers and destination promoters, thereby strengthening its position in a competitive tourism market (Benítez-Márquez et al., 2021; Chi et al., 2022). Although the positive correlation between destination image and tourist loyalty has been verified many times, the research on this relationship in the context of rural tourism is still limited (Zhang et al., 2014; Jeong and Kim, 2019). Therefore, this study proposes the following hypothesis:

Grounded on the above research and assumptions, a conceptual model is presented in Figure 1.

Located in Huangdao District, Qingdao City, Shandong Province, Yumingzui village is a traditional fishing village with a history of more than 300 years, surrounded by the sea on three sides and the mountains on one side. Local residents make full use of the geographical advantage and live a life combining agriculture and fishing. Although the village has been renovated in recent years and the whole village residents have moved to a nearby town, the villagers are still engaged in fishing activities here and sell seafood in the scenic spot, attracting a large number of tourists to come and buy. In addition, real estate has been developed on the original site, and villa groups have been built to form a rural tourism resort integrating business, catering and accommodation, providing tourists with an ideal residence for leisure and slow life. The coastal reefs provide an ideal spot for anglers, while the seaside lawns attract a large number of camping tourists.

The subjects of this study are tourists who went to Yumingzui village for vacation in 2024. The researchers used convenience sampling to collect data from voluntarily participating visitors at resort areas. Each visitor took about 15 to 20 min to complete the questionnaire. In addition, both the pilot survey and the primary data collection were aimed at visitors over the age of 18. Finally, a total of 489 valid samples were collected, which exceeded the minimum sample size recommended by Weston and Gore (2006).

The researchers of this study compiled a structured questionnaire by reviewing the relevant literature. The questionnaire consisted of six parts: (1) perceived destination restorative qualities, (2) hedonic experience, (3) eudaimonic experience, (4) destination image, (5) loyalty, and (6) background information of the interviewees. The 17 items of the perceived destination restorative qualities scale were verified in previous studies and modified appropriately according to the specific situation of Yumingzui village (Zhai et al., 2023). The 12-item scale created by Lengieza et al. (2019) was used to measure eudaimonic experience and hedonic experience. To assess tourists’ perception of a destination’s image, four items from a previous study were adapted (Chi and Han, 2021; Davras and Özperçin, 2023). The measurement of loyalty was based on four items in the previous research on rural tourism loyalty (Chen et al., 2024). The sixth part collected personal information such as gender, age and education level of visitors. All items were assessed using a five-point Likert scale, ranging from “strongly disagree” (=1) to “strongly agree” (=5).

Two Master’s students proficient in English independently translated the original English scale into Simplified Chinese (Mandarin). Subsequently, the research team compared the two translations, discussed and revised the discrepancies, and ultimately produced a unified Chinese version. Thereafter, two Master’s students fluent in both Chinese and English, who had not been exposed to the original English scale, conducted a back-translation of the Chinese version. The research team evaluated the back-translation results and convened an expert review committee for assessment. This expert review committee comprised both translation specialists and experts in the field of tourism. Prior to formal data collection, a pre-test was conducted to guarantee the reliability and validity of the questionnaire. Both the pre-test participants and the formal survey subjects were tourists on vacation in Yumingzui Village, who were invited to participate in the pre-test after consent. Data collection was conducted from April 6 to April 15, 2024, using the convenience sampling method. The target respondents were 120 people, and 103 valid questionnaires were eventually collected, with a response rate of about 86%. Subsequently, the data were evaluated for reliability. The results showed that Cronbach’s alpha reliability coefficient of each scale ranged from 0.73 to 0.91, indicating that the questionnaire had good reliability. As a result, all 37 items were retained in the official data collection.

The official data collection was conducted in Yumingzui village from April 27 to May 13, 2024, and the research team selected a total of 550 tourists to participate in the questionnaire. To ensure the availability of reliable samples, the researchers carried out an in-depth search for missing values and outliers. During the data preprocessing stage, a stringent filtering mechanism was implemented for the questionnaires. Specifically, any response with a time shorter than 180 s, as well as cases containing outliers, were systematically removed. Through this meticulous data-cleaning process, a total of 489 valid cases were retained, forming a solid basis for subsequent in-depth analysis. According to the data analysis, there were slightly more female respondents than men, accounting for 51.33 and 48.67%, respectively. The majority of respondents had a university degree (75.46%), of which 6.54% had a postgraduate degree. In terms of income, 74.48 percent of respondents earned less than 8,000 yuan a month, and 8.38 percent earned more than 15,000 yuan. The age of tourists is concentrated between 18 and 60 years old. The occupational distribution is more diverse. Table 1 details the demographic characteristics of the respondents.

Data analysis was performed using SPSS 27.0 and AMOS 26.0. The test of the measurement model is divided into two steps. First, the reliability and validity of the CFA tool are tested. Secondly, Cronbach’s alpha coefficient was calculated to check the internal consistency of each subscale. The research model was verified by structural equation modeling (SEM), and finally the relationship between the direct effects, indirect effects and total effects was calculated. The efficacy and strength of our SEM were meticulously evaluated using a range of well-established fit indices, including the Chi-Square to Degrees of Freedom Ratio, the Tucker-Lewis Index (TLI), the Comparative Fit Index (CFI), the Standardized Root Mean Square Residual (SRMR), and the Root Mean Square Error (RMSE) of Approximation (Schermelleh-Engel et al., 2003). A model was considered adequate when the TLI and the CFI reached or exceeded 0.95, and SRMR as well as the RMSE of Approximation remained below 0.05, thereby affirming the model’s validity and reliability (Schermelleh-Engel et al., 2003).

The measurement model’s reliability and validity was assessed by the metrics below: standardized factor loadings, Cronbach’s alpha coefficient, convergent validity, discriminant validity, as well as model fit.

From the study by Schumacker and Lomax (2004), the standardized factor loadings of the items should not be below 0.50. Though the standardized factor loadings of the whole items in this study were higher than 0.50 (ranging from 0.516 to 0.930), LA1 and DI4 were deleted to improve the fit of the model. The Cronbach’s alpha coefficients for all five constructs exceeded 0.70 (0.748 to 0.912) (Table 2), showing a good internal consistency.

From the research by Fornell and Larcker (1981), for all the constructs, the CR should not be below 0.70 and the AVE should exceed 0.50, and the five constructs in this research met the two criteria (Table 3). In terms of the discriminant validity of the constructs, the square root of the AVE must exceed the correlation coefficient between the construct and the other constructs, and this criterion was met by the present measurement model, implying its good discriminant validity (Table 3).

Combined with the criteria recommended by Hu and Bentler (1999), in this research, the fit of the measurement model was satisfactory, χ²/df = 1.347, TLI = 0.976, CFI = 0.979, RMR = 0.025, RMSEA = 0.027 (Table 4).

SEM validation of the research model was conducted. The research results showed that the model’s fit was χ²/df = 1.381, TLI = 0.973, CFI = 0.976, RMR = 0.032, and RMSEA = 0.028 (Table 4). The test results showed that all seven hypotheses were verified except H7 (Table 5).

Perceived destination restorative qualities were significantly and positively correlated with eudaimonic experience (β = 0.213, t = 3.006, p < 0.01), destination image (β = 0.417, t = 5.549, p < 0.001), and hedonic experience (β = 0.678, t = 6.641, p < 0.001), which supported Hypotheses H1, H2, and H3. Eudaimonic experience was also significantly and positively correlated with destination image (β = 0.154, t = 2.577, p < 0.05) and tourist loyalty (β = 0.230, t = 3.419, p < 0.001), supporting hypotheses H4 and H6. Hedonic experience was also positively correlated with destination image (β = 0.211, t = 2.703, p < 0.01) and tourist loyalty (β = 0.262, t = 3.432, p < 0.001), supporting hypotheses H5 and H8. However, destination image was not significantly associated with tourist loyalty (β = 0.087, t = 1.396, p > 0.05), negating hypothesis H7. The validated structural model is shown in Figure 2.

The indirect, direct, and overall impacts between perceived destination restorative qualities, eudaimonic experience, destination image, hedonic experience, and destination loyalty were recalculated after excluding non-significant correlation (Table 6). The results showed that perceived destination restorative qualities had a weight of 0.234 on destination loyalty mediated by eudaimonic experience and hedonic experience, with hedonic experience having the largest indirect effect (β = 0.521 × 0.262 = 0.137), eudaimonic experience having the second largest indirect effect (β = 0.154 × 0.230 = 0.035), and destination image having the smallest indirect effect (β = 0.211 × 0.087 = 0.018).

Based on SOR model, this study explores the impact of perceived destination restorative qualities on tourists’ hedonic experience, eudaimonic experience, destination image and loyalty, and verifies these relationships through empirical research. The results are as follows:

The perceived destination restorative qualities of rural tourism destinations have significant positive effects on tourists’ hedonic experience, eudaimonic experience and destination image. According to ART, the natural environment can effectively relieve the mental fatigue of tourists and improve their concentration and mental resources (Lee et al., 2015). Therefore, the sense of relaxation and pleasure in the rural environment significantly enhances the pleasure-seeking experience of visitors. Rural tourism provides an environment for tourists to relax, and more importantly, it creates a space for self-reflection (Sang, 2020). In the quiet and close-to-nature environment, it is easier for visitors to engage in inner thinking and self-examination, thereby reassessing life goals, values and their own potential, and enhancing the realization experience (Løvoll et al., 2020; Ding and Xu, 2024). Besides, rural tourism also incorporates rich cultural and social interactions, such as village folk activities and the warm reception of local residents (Chi and Han, 2021). These cultural and interpersonal experiences further deepen tourists’ goodwill and emotional connection with the destination and shape a more positive image of the destination.

Hedonic experience has a significant positive impact on destination image and loyalty, indicating that a pleasant and relaxed experience is the key factor to enhance tourist identity and loyalty. In contrast, although eudaimonic experience also has a positive impact on destination image and loyalty, its direct impact is weak, which may be due to that the immediacy of hedonic experience and strong emotional experience are more likely to directly affect tourists (Lee, 2024). Existing studies mainly focus on hedonistic wellbeing, with relatively few studies on self-achievement and personal development (Wang et al., 2020; Zhang et al., 2024). This study, for the first time, comprehensively examines the experience of pleasure and eudaimonic in rural tourism, and the results show that the hedonic experience strengthens the emotional connection between tourists and the destination, while the eudaimonic experience enhances the sense of identity of tourists through personal growth.

Although many tourism studies regard destination image as an important factor affecting loyalty (Zhang et al., 2014; Chi and Han, 2021), this study finds that its direct impact on loyalty is not significant, which may be related to the research background, sample selection, cultural differences, and differences in loyalty measurement methods (Charinsarn et al., 2023; Qi et al., 2023). In rural tourism, research shows that tourists often prioritize personal restorative experiences and emotional bonds over a straightforward cognitive evaluation of destination image (Huang and Bu, 2022; Ding and Xu, 2024). Moreover, studies by Chiu et al. (2016) and Tasci et al. (2022) indicate that factors like emotional attachment, satisfaction, and place attachment can moderate this relationship. These studies provide a theoretical basis for interpreting our results and suggest that the impact of destination image on loyalty may not always be direct.

By incorporating social factors such as cultural interaction and community cohesion, we address a dimension of restorative tourism that has traditionally centered on natural or physical environments, thereby offering a more comprehensive understanding of the rural tourism experience. The application of the SOR model in a rural setting further clarifies how environmental stimuli affect visitors’ psychological responses, extending the scope of environmental psychology in tourism research. We also investigate the mediating roles of hedonic and eudaimonic wellbeing between perceived restorative qualities and loyalty, addressing a gap in the literature that often focuses on a single dimension of wellbeing. Moreover, the finding that destination image does not directly influence loyalty challenges prevailing assumptions and suggests that emotional factors and overall experience quality may weigh more heavily than cognitive evaluations of destination image, particularly in rural tourism contexts.

This study confirms a positive relationship between the five dimensions of perceived environmental restorativeness and tourist loyalty, highlighting its crucial role in rural tourism development. Destinations can enhance tourist relaxation, interest, and exploration by creating tranquil nature areas, incorporating multisensory designs, and offering diverse activities for different visitor groups (e.g., families, backpackers, senior travelers). Additionally, developing ecotourism, hiking, fruit picking, and camping can further enrich tourist experiences and satisfaction. Moreover, the study reveals that wellbeing (both hedonic and eudaimonic) plays a key mediating role between perceived environmental restorativeness and loyalty. Therefore, rural tourism destinations should balance relaxation and personal growth by designing integrated experiences that combine cultural activities with nature-based healing. This approach enables tourists to achieve both enjoyment and self-fulfillment, ultimately enhancing their overall wellbeing and loyalty.