How social media information processing influences tourist destination image: a heuristic-systematic model perspective

Social media has transformed how information is created and shared (Geyer et al., 2024), with user-generated content (UGC) now significantly influencing daily life. Within tourism contexts, travelers increasingly rely on social platforms to search for destination-related information, which significantly shapes their evaluations of destinations and ultimately determines travel-related decisions (Villamediana et al., 2019). Research has highlighted that the presentation and perception of platform-based information critically shape potential visitors’ behavioral patterns (Xiang and Gretzel, 2010). Therefore, although previous studies have extensively examined how social media and UGC affect destination perceptions, it is still necessary to elucidate how tourists cognitively process multi-source social media information and how these perceptions coalesce into destination image formation.

Social media provides users with an unparalleled engagement space where they can freely consume and contribute content without time or location constraints (Kim et al., 2017). In tourism, tourists increasingly share their experiences through various forms of tourist-generated content (TGC), which has become a critical reference for travel decision-making (Ruan et al., 2025). Meanwhile, supply-side stakeholders such as destination marketing organizations (DMOs) use these platforms to promote destination features through DMO-generated content (DGC) (Hernandez-Ortega et al., 2020). Both TGC and DGC play significant roles in shaping destination image (Geyer et al., 2024), and while systematic reviews have examined UGC in this context (Gurung and Goswami, 2016; Li et al., 2023), research comparing the differential impacts and synergistic effects of TGC and DGC remains limited. Although studies have analyzed destination image attributes via UGC (Qian et al., 2023; Xiao et al., 2022) or compared perceived and projected images through TGC and DGC (Geyer et al., 2024; Qu et al., 2022), few have quantitatively assessed how TGC and DGC distinctly and interactively influence destination image, leaving a key gap in understanding their combined role in tourists’ pre-trip evaluations.

Beyond the content-centric processing of UGC, tourists have heterogeneous information evaluation strategies. A segment of travelers may prioritize heuristic processing over the systematic analysis of UGC, opting for simplified destination assessments to reduce cognitive effort (Son et al., 2020). Concurrently, others who are aware of potential systemic biases in both TGC and DGC (García-Carrion et al., 2024) may seek supplementary verification through alternative channels. In such contexts, destination source credibility (DSC), defined as the credibility of destination managers (Veasna et al., 2013), emerges as a critical heuristic cue. DSC affects the acceptance of destination-related information and directly shapes destination image (Qiu et al., 2022). This dual role underscores the necessity of integrating DSC into theoretical framework, particularly in an era marked by information overload and declining trust in digital platforms (Shahbazi and Bunker, 2024).

This characteristic of tourists’ information processing can be explained by the heuristic-systematic model (HSM), which was proposed by Chaiken (1980) as a dual-process theory model for explaining individual information-processing behaviors. This model suggests that individuals mainly process information in social activities via two modes: heuristic and systematic (Chaiken and Ledgerwood, 2012). Systematic cues require substantial cognitive effort to process the actual content of a message (Tan et al., 2021); thus, both DGC and TGC are classified as such. Conversely, heuristic cues involve low-effort processing that relies on simple decision rules based on peripheral cues (Son et al., 2020); thus, DSC is classified as a heuristic cue (Table 1).

Although the elaboration likelihood model is also another important model of the dual-process theory, it assumes that the two processing methods are mutually exclusive and are influenced by the level of motivation; in contrast, HSM assumes that the two modes can coexist simultaneously and does not take into account motivational factors (Chen and Chaiken, 1999). Therefore, HSM provides a more appropriate and nuanced theoretical lens for the present study because it allows researchers to concurrently model the distinct yet potentially co-occurring influences of a heuristic cue (DSC) and systematic cues (DGC and TGC) on destination image formation.

Therefore, this study sought to address the aforementioned research gap by applying HSM to examine how social media information processing (including DGC, TGC and DSC) collectively shapes destination image. This study addresses the following research questions:

1. How do DGC, TGC and DSC collectively influence tourists’ perceptions of the cognitive and affective image of a destination?

2. Does perceived risk mediate the relationships between these social media cues (DGC, TGC and DSC) and destination image?

3. Do generational differences (between Gen Y and Gen Z) moderate these relationships?

Specifically, this study investigates the effects of DGC, TGC and DSC on tourists’ perceptions of the cognitive image and affective image of a destination and the mediating role of perceived risk. Furthermore, considering that Generation Y (Gen Y) and Generation Z (Gen Z) are the main forces driving contemporary tourism consumption and are also the primary users of social media, they differ significantly in how they handle social media information (Ong et al., 2024). Importantly, previous literature has barely discussed the differences between Gen Y versus Gen Z in the context of destination image via social media. Therefore, this study also investigates the differences between them in the relationship between social media information processing and destination image. This study contributes to further understanding of the mechanism of destination image formation based on social media information. The findings can guide DMOs in leveraging social media to strategically enhance the images of their destinations.

Social media features community engagement, open channels and dialog promotion, enabling the creation and exchange of UGC (Kaplan and Haenlein, 2010). Information related to destination travel is expressed by users via texts, pictures, emoticons and short videos on these platforms, forming UGC. This UGC plays a crucial role in enhancing the prominence of destination image (Tham et al., 2013). From the perspective of DMOs, UGC not only actively constructs the destination image by disseminating promotional narratives aligned with diverse thematic attributes but also provides critical insights into tourists’ perceived destination image.

UGC has become an essential data source for tourism research and destination image analysis (Adamış and Pınarbaşı, 2022; Wang et al., 2024). While numerous studies have used UGC to construct destination image (Geyer et al., 2024; Qu et al., 2022), few have explored the structural relationship between UGC and destination image (Jalilvand et al., 2012; Kim et al., 2017). Moreover, previous studies ignored the consideration of UGC from different sources. Given that the destination information received by tourists through DGC and TGC is often inconsistent (García-Carrion et al., 2024), it is vital to understand how tourists view UGC from different sources and the impact of UGC on destination image.

Following previous research (Huerta-Alvarez et al., 2020), this study categorizes social media information into two sources: the supply side (DMOs: DGC) and the demand side (tourists: TGC). As established in the literature, DMOs use social media to establish destination brand identity and positive image (Mariani et al., 2018), while tourists strongly shape and influence destination image by sharing knowledge, emotions and experiences (Lim et al., 2012). Both sources significantly influence tourists’ perceptions, but their unique effects on the cognitive and affective dimensions of the image remain under-explored.

DGC represents information that includes destination products and services published by DMOs on social media. The rise of social media has dramatically changed destination marketing practices. DMOs now actively disseminate visual and textual content to provide tourists with curated product information and to project a positive image. Furthermore, social media also provides destinations with better opportunities to communicate and interact with consumers, which is conducive to enhancing the destination’s image (Koltringer and Dickinger, 2015). Consequently, social media has become an indispensable strategic tool for DMOs.

However, the actual impact of DGC on tourist attitudes and behaviors is still being determined (Sano et al., 2024). Prior research on official destination websites has demonstrated that their information enhances destination image and visit intentions (Chung et al., 2015). Recent studies confirm that DGC on social platforms can modify tourists’ behavioral intentions and engagement patterns (Stojanovic et al., 2022). DGC typically provides comprehensive, accurate, factual and positive descriptions of destinations. According to HSM, such information triggers systematic processing, prompting tourists to exert cognitive effort to analyze and integrate this information, thereby systematically constructing their understanding of the attributes and functions of the destination. At the same time, DMOs also aim to evoke tourists’ emotional resonance through carefully planned content. Therefore, this study proposes the following hypotheses:

TGC refers to travel-related content (e.g. pictures, texts and videos) created and shared by tourists on social media (Mak, 2017). Within the context of Travel 2.0, tourists increasingly use social media to generate, share and exchange content (Sano et al., 2024), fundamentally reshaping information dissemination. TGC is considered more important, reliable and interesting, and consumers have more confidence in it (Llodrà-Riera et al., 2015). Therefore, it is an indispensable source for perceiving a destination image. The capacity of TGC to disseminate favorable evaluations and enhance destination image warrants significant scholarly and practical consideration. While prior research has robustly validated the utility of TGC in constructing and identifying destination image attributes (Wang et al., 2024), causal investigations into how TGC directly influences destination perception remain under-explored.

As a reliable form of word-of-mouth, TGC is one of the most potent and dependable channels for influencing destination image (Ukpabi and Karjaluoto, 2018). Content shared by tourists generally expresses their true views and emotions toward a destination, which can affect and change other tourists’ perceptions (Llodrà-Riera et al., 2015). TGC, as personalized narratives, real experiences and emotional expressions shared by tourists, also undergo systematic processing. Tourists immerse themselves in these stories, empathize with the publishers and form emotional feelings toward the destination atmosphere. Simultaneously, TGC contains a large amount of practical information about facilities and experiences, allowing others to extract factual knowledge and systematically construct a cognitive image. Therefore, this study suggests that TGC is one of the critical factors of destination image and proposes the following hypotheses:

DSC is the perceived willingness and capability of destination managers to execute obligations aligned with a destination’s strategic objectives (Veasna et al., 2013). It determines the extent to which tourists believe destination-relevant statements are accurate and credible (Phau and Ong, 2007). As a critical non-content cue, DSC can affect tourists’ attitudes and behavioral intentions toward a specific destination. In today’s social media landscape, “source” has expanded beyond official DMOs to include influencers and peer contributors (Hernández-Méndez and Baute-Díaz, 2024; Ong et al., 2024). Nonetheless, the fundamental heuristic processing mechanism – relying on perceived source credibility to form judgments – remains central. This study follows Veasna et al.’s (2013) definition and uses DSC as a heuristic factor to examine its impact on destination image.

Destination image is formed by integrating heterogeneous informational stimuli (Zhang et al., 2014). Facing information overload, tourists increasingly use heuristic shortcuts (e.g. DSC) to form quick judgments without deep content processing (Chaiken, 1980). This low-effort heuristic processing can effectively reduce decision-making uncertainty and shape initial attitudes. This study posits that DSC serves as a vital cue; when information sources are perceived as credible, this credibility exerts a persuasive influence on their positive evaluations of the destination (Kani et al., 2017). This logic not only applies to traditional human or institutional sources but also extends to emerging digital entities. For instance, while virtual influencers represent a novel source type (Meng et al., 2025) and AI-generated content challenges the very nature of “source” authorship, their potential influence is still contingent on the audience’s heuristic assessment of their credibility – a core premise this study aims to test within the established framework. Therefore, this study proposes the following hypotheses:

In tourism, perceived risk refers to tourists’ subjective expectations of encountering various potential misfortunes or dangers during travel or at a destination (Tsaur et al., 1997). It is a multidimensional construct encompassing distinct facets such as physical, social-psychological and financial risks, all of which are critical to destination selection decisions (Chew and Jahari, 2014). The intangible, inseparable, variable and perishable nature of tourism services frequently elevates these risk perceptions during the decision-making process (Alcántara-Pilar et al., 2018). Consequently, analyzing risk perceptions is essential for understanding the mechanisms through which social media information processing shapes destination image.

Perceived risk serves as a pivotal determinant shaping tourists’ cognitive and affective evaluations of a destination and travel decision-making processes. Research confirms that travel-related risks can negatively impact destination image (Assaker and O’Connor, 2020; Xie et al., 2020). This study posits that social media information processing reduces these generalized risk perceptions. When tourists engage in systematic processing of DGC or TGC, the acquired information reduces uncertainty about the destination (Lepp et al., 2011). Simultaneously, a high DSC serves as a potent heuristic cue, acting as a strong low-risk signal that fosters initial trust (González-Rodríguez et al., 2022). The alleviation of overall perceived risk is, thus, hypothesized to facilitate more positive cognitive and affective destination evaluations.

To establish the fundamental mediating mechanism, this study first examines perceived risk as a global, higher-order construct (Quintal et al., 2010). This approach allows for a parsimonious test of the core theoretical proposition that risk reduction is a central pathway linking social media cues to image formation. Therefore, this study proposes the following hypotheses:

Subsequently, to uncover more granular insights and acknowledge the multidimensionality of the construct, this study also conducts exploratory analysis testing physical, social-psychological and financial risks as separate mediators. This secondary analysis aims to reveal whether specific risk dimensions are uniquely tied to cognitive or affective image outcomes, thereby refining the theoretical understanding of the risk mediation pathway.

Based on the above analysis, this study proposes the model shown in Figure 1.

A generation comprises individuals born within a specific period who share formative historical, socioeconomic, cultural and political experiences (Strauss and Howe, 2009). Generation cohorts exhibit different characteristics, values, interests, expectations and behaviors, providing a macro-level for understanding tourist behaviors (Gao et al., 2018). Previous studies have mainly compared Baby Boomers, Generation X and Gen Y (Luna-Cortés, 2018); only a few recent studies have begun to include Gen Z in their comparisons (Horpynich et al., 2025; Liu et al., 2023). Although inter-generational differences in attitudes and behaviors are recognized (Luna-Cortés, 2018), the extent to which generational differences, especially between Gen Y and Gen Z, affect the mechanisms underlying destination image remains under-explored.

Gen Y (born 1980–1995) and Gen Z (born 1995–2009) (Lebrun and Bouchet, 2024) together represent the dominant consumer force in today’s tourism market. A key distinction is that Gen Y tends to use peripheral persuasion approaches to handle information, whereas Gen Z prioritizes content quality (Ong et al., 2024). This distinction reflects broader sociocultural shifts in media socialization. Gen Z, as “digital natives”, has been enculturated into a participatory, networked media habitat from birth, whereas Gen Y, as “digital immigrants,” adapted to this landscape later in life (Prensky, 2001). These fundamentally different media habitats likely shape their default information processing strategies. This study suggests that there may be differences between these two groups in the process of assessing destination image through social media-based information. Therefore, Gen Y and Gen Z differences are incorporated as a moderating variable in the proposed conceptual model to examine how distinct cohorts differentially influence the relationship between social media information processing and destination image.

The measurement items were adapted from established scales in the prior literature, with minor adaptations for the rural tourism destination context ( AppendixTable A1). Specifically, DGC and TGC were measured separately using four items taken from Huerta-Alvarez et al. (2020). DSC was measured using six items adapted from Veasna et al. (2013). Perceived risk was operationalized using a 15-item scale from Chew and Jahari (2014), which included three dimensions: physical risk, social-psychological risk and financial risk. However, one item, “I worry that this trip will change the way my family thinks of me,” exhibited a low factor loading in this study and was subsequently removed to improve the average variance extracted (AVE) values for the construct.

For destination image, measurement items were adopted from Liu et al. (2024), whose scale operationalized the cognitive image and affective image of Hongcun. All items across constructs were recorded on a seven-point Likert scale ranging from 1 (strongly disagree) to 7 (strongly agree). Given that the original scales were developed in English, the Chinese version of the survey was rigorously translated through back-translation (Brislin, 1980) to ensure linguistic equivalence and conceptual consistency.

Rural tourism destinations are often in remote and secluded locations, and social media information is more important for these places (Liu et al., 2024). Therefore, choosing a rural tourism destination as a study site can provide valuable insights for this research model. This study used a random sampling method to collect data in Hongcun, a rural tourism destination in China. Data were collected via a paper-based, self-administered questionnaire distributed to participants in designated areas within the destination. The survey was administered during September and October 2023. Before the survey, participants read a survey explanation and were informed that the survey was strictly for academic purposes and that no personally identifiable information was collected to ensure anonymity and compliance with ethical standards. Eligible participants who satisfied the inclusion criteria and consented to participate were administered the survey instrument.

To qualify for participation, respondents had to confirm that they were social media users and had learned about Hongcun through social media. To ensure a clear understanding of the key constructs, this study provided concise operational definitions and illustrative examples of DGC (e.g. “official posts from Hongcun’s tourism board account”), TGC (e.g. “information about Hongcun shared by other tourists”) and DSC (e.g. “the credibility of the source of information about Hongcun”) immediately before the relevant survey sections. This approach was validated through a pilot study to confirm comprehension. A total of 800 responses were initially collected. Following rigorous data screening to exclude incomplete or inconsistent entries, 688 validated data sets were retained for analysis. The respondents were 48.1% male and 51.9% female. The majority of the participants belonged to Gen Y (37.2%) and Gen Z (56.5%) and had bachelor’s degrees or above (74%). Most of the sample had over three years of social media experience (88.1%) and spent at least one hour on social media every day (88.2%).

Data analysis was performed using partial least squares structural equation modeling (PLS-SEM), following Chin’s (2010) two-step approach: evaluating the measurement model’s reliability and validity and testing structural relationships among latent constructs. This study also used Henseler’s multi-group analysis (MGA) method to determine the significance of the differences between Gen Y and Gen Z.

Harman’s single-factor test was conducted to evaluate potential common method bias (Podsakoff et al., 2012). The rotated factor solution demonstrated that no single factor explained more than 38.09% of the total variance, well below the critical 50% threshold. Furthermore, Kock’s (2017) full collinearity assessment was applied, yielding variance inflation factor values between 1.611 and 1.945 across latent variables, significantly lower than the stringent cutoff value of 3.3. These diagnostic results collectively confirm the absence of substantial common method bias in this study.

The measurement model was rigorously assessed through confirmatory factor analysis, focusing on key indicators: factor loadings, Cronbach’s α, composite reliability (CR) and AVE. As shown in Table 2, except for a few items with slightly lower factor loadings for perceived risk and cognitive image, most items had factor loadings above the cutoff value of 0.7. As both CR and AVE values exceeded the recommended threshold, convergent validity was acceptable, and there was no need to delete other items with factor loadings lower than 0.7 (Rasoolimanesh et al., 2017). Additionally, Cronbach’s α values were greater than 0.7 for all variables, confirming the reliability of the measurements.

Discriminant validity was assessed using the heterotrait-monotrait (HTMT) ratio, a method validated as more robust than conventional approaches, such as the Fornell–Larcker criterion (Liu et al., 2023). As reported in Table 3, the HTMT values for all constructs were below the threshold of 0.85, thereby establishing robust discriminant validity.

The structural model was assessed using the coefficient of determination (R²) for endogenous latent constructs, which measures how much variance is explained by the predictor variables (Ringle et al., 2010). In this current study, R² values for perceived risk, cognitive image and affective image reached 0.391, 0.510 and 0.375, respectively, which signify moderate explanatory power of the predictor variables.

The goodness-of-fit index (GOF), proposed by Tenenhaus et al. (2005), is a model evaluation tool used to indicate the degree of fitting between the collected data and the constructed model. For this study, the GOF value was 0.522, which is higher than the 0.36 standard, indicating that the model fit well. The standardized root mean square residual (SRMR), a key diagnostic metric for PLS-SEM (Henseler et al., 2016), was further evaluated. In this study, the computed SRMR of 0.068 fell well within the threshold of 0.08, corroborating the robust model fit.

The results of the hypothesis tests are presented in Table 4 and Figure 2. The results indicated that the DGC exerted a significant positive effect on cognitive image (β = 0.144 and p < 0.01) but showed no significant impact on affective image (β = 0.073 and p > 0.05), thereby supporting H1a and rejecting H1b. TGC significantly influenced affective image (β = 0.140 and p < 0.05) but had no statistically meaningful effect on cognitive image (β = 0.084 and p > 0.05), rejecting H2a and supporting H2b. DSC demonstrated a significant positive effect on both cognitive image (β = 0.405 and p < 0.001) and affective image (β = 0.351 and p < 0.001), supporting H3a and H3b.

To assess the mediating role of perceived risk, this study adopted a non-parametric bootstrapping method (Chin, 2010). The findings demonstrated that perceived risk significantly mediated the effects of DGC, TGC and DSC on both cognitive and affective destination image (Table 5), supporting H4a–H4b, H5a–H5b and H6a–H6b. Moreover, to gain a deeper understanding of the mediating effect of perceived risks, this study also explored different types of perceived risk as independent mediating variables. The findings show that financial risk mediates the relationship between DGC, TGC and DSC and cognitive image. Similarly, social-psychological risk mediates the influence of these three factors on affective image. However, physical risk was found to have no mediating effect whatsoever.

To test the differences in the structural relationships between Gen Y and Gen Z in the research hypotheses, this study used MGA. The results (Table 6) revealed significant differences in structural path coefficients between generations. For Gen Y, DGC (β = 0.176 and p < 0.01) and DSC (β = 0.400 and p < 0.001) exerted significant positive effects on cognitive image, and DSC (β = 0.412 and p < 0.001) demonstrated a significant positive effect on affective image. For Gen Z, DSC (β = 0.396 and p < 0.001) showed a strong positive impact on cognitive image, while TGC (β = 0.171 and p < 0.05) and DSC (β = 0.307 and p < 0.001) significantly enhanced affective image.

The MGA results revealed statistically significant differences in path coefficients between DSC and cognitive image (p = 0.961 > 0.95) across Gen Y and Gen Z. Specifically, the path coefficient difference was 0.004, indicating that Gen Y’s cognitive image of a destination was more strongly influenced by DSC than Gen Z.

This study develops and empirically validates a conceptual framework based on HSM that elucidates the differential impacts of DGC, TGC and DSC on destination image. The findings advance theoretical discourse on the mechanisms through which social media information – both curated (DGC) and organic (TGC) – interacts with DSC to reconfigure tourists’ cognitive and affective perceptions of destinations.

First, the findings reveal a clear divergence in how DGC and TGC shape destination image. DGC significantly enhances the cognitive image but not the affective image, while TGC positively influences the affective image but not the cognitive dimension. This pattern, while contrasting with some previous studies (Kim et al., 2017), empirically substantiates the distinction between DMO-projected and tourist-perceived images (Geyer et al., 2024), highlighting the complementary roles of DGC and TGC. These results align with and extend the HSM. Although both DGC and TGC necessitate systematic processing, their distinct informational nature activates different subsystems. Factual DGC directs processing toward cognitive evaluation, fostering a cognitive image. In contrast, experiential TGC triggers affective processing, leading to an emotional image. This refines the HSM by demonstrating that systematic processing is not monolithic but can be cognitively or affectively oriented based on the cue’s characteristics.

Second, the results support the hypothesis that DSC has a significant direct and positive impact on cognitive and affective images. This finding seems consistent with previous studies, although they did not delve into the effects of DSC on cognitive and affective images (Veasna et al., 2013; Kani et al., 2017). This study also found that the impact of DSC on destination image was greater than that of both DGC and TGC. This finding suggests that, in an age of information overload, tourists are likelier to process information in a heuristic way to save time and cost. This might also be related to cultural background. China has the characteristics of collectivism and a high-power distance (Hofstede, 2011). The enhanced influence of DSC could be partly attributed to the high value placed on in-group harmony and shared opinions within collective cultures.

Third, the results reveal that perceived risk mediates the influence of DGC, TGC and DSC on cognitive image and affective image. The findings align with theoretical expectations that social media information and source credibility shape destination image by modulating tourists’ perceived risk. The findings also offer a more nuanced understanding of risk mediation mechanisms. Specifically, financial risk mediates the influence of DGC, TGC and DSC on cognitive image, while social-psychological risk mediates their effects on affective image. This aligns with the notion that cognitive evaluations are often tied to utilitarian, cost–benefit assessments (Quintal et al., 2010), whereas affective responses are more sensitive to identity and social belonging concerns (Williams and Soutar, 2009). By contrast, physical risk did not demonstrate a mediating effect, possibly because of the context of a rural tourism setting perceived as low-threat. These results refine and extend the proposition that social media serves as a critical risk-reduction mechanism (Noh and Vogt, 2013), highlighting that different risk dimensions operate through distinct cognitive and affective pathways.

Additionally, the results reveal distinct generational pathways in image formation: DGC significantly shapes cognitive image for Gen Y but not Gen Z, whereas TGC significantly influences affective image for Gen Z but not Gen Y. This indicates that each content type functions through different generational mechanisms. Furthermore, Gen Y’s cognitive image is more strongly influenced by DSC compared to Gen Z, which aligns with Ong et al. (2024), who found that Gen Z is more susceptible to broad social media influence, while Gen Y relies more on peripheral cues. These patterns are explained by the HSM and generational media socialization. As digital immigrants, Gen Y uses DSC as a heuristic cue for efficient cognitive validation, favoring peripheral processing. In contrast, Gen Z, as a digital native, excels in the systematic affective processing of TGC, engaging deeply with peer narratives to form emotional connections. Thus, generational identity moderates HSM pathways, specifying for whom and how different content types operate. Importantly, the moderating role of generational cohorts may be interpreted as not merely an age effect but also a proxy for embeddedness in different sociocultural media environments.

This study makes three pivotal contributions to the literature on tourism marketing and digital communication. First, it advances the HSM by applying and extending it within the novel context of destination image formation. While HSM has been used in persuasive communication (Xie et al., 2023), its application to disentangle the complex effects of organic (TGC) versus institutional (DGC) systematic cues alongside a heuristic source cue (DSC) in tourism is scarce. This study successfully integrates both heuristic (DSC) and systematic (DGC, TGC) cues into a unified HSM framework for examination, revealing a synergistic yet distinct relationship between them, overcoming limitations of prior isolated investigations (Huerta-Alvarez et al., 2020; Qiu et al., 2022). This study not only validate HSM’s core premise in this domain but also reveal a critical functional specialization: DGC, as a factual systematic cue, is primarily routed to shape cognitive image, whereas TGC, as an experiential systematic cue, predominantly shapes affective image. This finding moves beyond confirming HSM’s utility to refining its granularity, demonstrating that systematic processing is not monolithic but comprises distinct sub-pathways with attitudinal consequences.

While HSM provides a clear theoretical lens for classifying DGC and TGC as systematic cues because of their substantive content, this study acknowledges the contextual complexity of user engagement. As noted in recent literature (Rita et al., 2022; Liu et al., 2023), certain elements within the UGC ecosystem (such as aggregate ratings, summary scores or high-level popularity metrics) can themselves serve as potent heuristic cues, enabling quick judgments with minimal cognitive effort. The findings and model apply most directly to the elaborated, narrative content of DGC and TGC that requires systematic processing. The potential for simplified UGC elements to function heuristically represents a valuable content-level boundary condition. Moreover, platform affordances (such as algorithmic curation and visibility of peer endorsements) create digital contexts that may systematically moderate reliance on heuristic versus systematic processing. Thus, the applicability of the findings is delineated by both content complexity and platform architecture, outlining key boundary conditions for the framework.

Second, this study elucidates the “black box” by introducing and validating a nuanced risk-based mediation mechanism. This study transcends merely establishing direct effects between social media cues and destination image by uncovering how these effects occur. The differential mediation of financial risk (for cognitive image) and social-psychological risk (for affective image) provides a more precise psychological account of the underlying process. This answers calls for deeper mechanism-oriented research in tourism social media, moving from correlation to clearer causation.

Third, this study establishes generational cohort as a key theoretical boundary condition. The findings demonstrate that the proposed HSM pathways are not universal. The heightened reliance of Gen Y on heuristic credibility (DSC) versus Gen Z’s responsiveness to affective systematic processing of TGC contextualizes HSM within the critical framework of media socialization. This shifts the theoretical conversation from “do these cues work?” to “for whom do they work best, and why?”, significantly enhancing the model’s explanatory power and practical relevance.

The findings of this study yield actionable strategies for DMOs to optimize social media engagement and strategically cultivate destination image. First, DMOs must move beyond blanket social media posting. To build cognitive capital (e.g. for first-time visitors), investment should prioritize high-quality, detailed and utility-focused DGC. Concurrently, to fuel emotional appeal, resources must be directed toward curating and amplifying authentic TGC that showcases visceral experiences. This implies a dual strategy: producing official content that informs, while actively fostering a community where positive visitor narratives can thrive (e.g. through featured user campaigns or photo contests).

Second, as DSC exerts a foundational influence surpassing even content effects, managing perceived credibility must be a core organizational priority. This extends beyond consistent messaging to encompass platform-specific reputation management (e.g. responding professionally to all reviews), leveraging trusted endorsers (e.g. collaborations with credible influencers aligned with destination values) and transparent communication, especially during crises. DMOs should view every online interaction as an investment in their credibility stock. Moreover, when promoting tourist destinations, managers need to invest more in marketing communication and destination promotion, as well as strengthen the reputation of the destination, shape positive and firm brand beliefs and thereby enhance tourists’ perceptions of the credibility of the destination source (Baek et al., 2010).

Third, DMOs should audit their communications through a risk-mitigation lens. Information addressing financial security (e.g. clear pricing and refund policies) can directly enhance cognitive evaluations. Content that alleviates social-psychological concerns (e.g. showcasing welcoming locals and diverse visitor types) can bolster affective connection. Furthermore, marketing efforts must be generationally tailored: for Gen Y, emphasize credentials and expert endorsements (leveraging heuristic processing); and for Gen Z, prioritize authentic, peer-driven storytelling on visual platforms like TikTok or Instagram Reels (leveraging affective systematic processing).

This study has some limitations that should be addressed in future work. First, it was conducted in rural tourism destination in China. While this offers a valuable case-specific understanding, it limits the generalizability of the findings to other cultural settings. Future studies could replicate this research in different cultural contexts to allow for cross-cultural comparisons, which are crucial for explaining destination image evaluations (Lee and Park, 2023). Such work would validate the proposed model and explicitly test the moderating role of culture.

Second, this study did not focus on a specific social media platform, which precludes an examination of how platform-specific affordances moderate information processing. Subsequent research could examine how destination image is shaped within a particular platform’s context and could explicitly model such affordances as boundary conditions that moderate the pathways in our framework to advance a more context-sensitive theory. Moreover, although sufficient samples were obtained for Gen Y and Gen Z, the small sample size of Gen X (n = 43) limited meaningful analysis for this cohort, possibly reflecting recruitment challenges (e.g. coverage bias in online surveys). Future work should use targeted sampling strategies to better understand Gen X’s perceptions and their continuity with or divergence from younger generations.

Third, although demographic variables were controlled for, potential omitted variable bias (e.g. individual characteristics or travel motivations) cannot be ruled out. Because of this study’s cross-sectional design, the results should be interpreted as robust associations rather than causal relationships. Longitudinal designs or more comprehensive psychological measurements are recommended to strengthen causal inferences.