Deep Thinking Function in AI-Mediated Travel Planning: How Reasoning Transparency Shapes Information Adoption and Destination Acceptance

Deep Thinking Function in AI-Mediated Travel Planning: How Reasoning Transparency Shapes Information Adoption and Destination Acceptance | Research Square window.SnipcartSettings = { analytics: { enabled: false } }; (function() { var accessVector = localStorage.getItem('access_vector') || ''; window.dataLayer = window.dataLayer || []; if (accessVector) { window.dataLayer.push({ user: { profile: { profileInfo: { snid: accessVector } } } }); } })(); (function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src='https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);})(window,document,'script','dataLayer','GTM-K279D39R'); Browse Preprints In Review Journals COVID-19 Preprints AJE Video Bytes Research Tools Research Promotion AJE Professional Editing AJE Rubriq About Preprint Platform In Review Editorial Policies Our Team Advisory Board Help Center Sign In Submit a Preprint Cite Share Download PDF Research Article Deep Thinking Function in AI-Mediated Travel Planning: How Reasoning Transparency Shapes Information Adoption and Destination Acceptance Song Chen, Fengbo Wang, Kalybek Zh Abdykadyrov, Junyu Long, Jiayi Xu, and 1 more This is a preprint; it has not been peer reviewed by a journal. https://doi.org/ 10.21203/rs.3.rs-8736638/v1 This work is licensed under a CC BY 4.0 License Status: Posted Version 1 posted You are reading this latest preprint version Abstract This study investigates how DeepSeek’s Deep Thinking Function, which introduces reasoning transparency into AI-mediated travel planning, shapes travelers’ psychological evaluation and adoption of AI-generated destination recommendations. Drawing on an extended Information Adoption Model (IAM), the study conceptualizes Deep Thinking Function not merely as a system design feature, but as a cognitive persuasion mechanism that reconstructs users’ judgment criteria through process-based reasoning disclosure. A purposive sampling strategy was applied to recruit tourism users who had recently interacted with AI travel assistants, yielding 260 valid responses collected via an online survey. Partial Least Squares Structural Equation Modeling (PLS-SEM) was employed to assess the research model. The results reveal that Deep Thinking Function significantly enhances perceived argument quality and source credibility, which subsequently strengthen perceived information usefulness, information adoption, and destination acceptance intention. The findings indicate that, in AI-generated information environments, travelers’ adoption decisions shift from outcome focused evaluation toward process-oriented cognitive appraisal, where analytical competence and methodological rigor become core foundations of trust. This study extends IAM into AI-mediated tourism contexts, conceptualizes transparency as an active persuasion mechanism rather than a static ethical attribute, and demonstrates the emergence of system centric cognitive trust in travel decision-making. Practical implications highlight the value of process aware AI interface design, transparency driven persuasion strategies, and the repositioning of destination branding toward decision logic compatibility in post-AIGC tourism ecosystems. AI-mediated travel planning Deep Thinking Function reasoning transparency Information Adoption Model Figures Figure 1 1. Introduction In recent years, the global online travel market has expanded rapidly alongside the widespread adoption of mobile internet technologies and platform-based tourism services. Reports indicate that the global online travel economy reached approximately USD 5667.4 billion in 2024 (IMARC, 2025 ). In China, although the tourism industry experienced unprecedented disruption during the COVID-19 pandemic, the sector has shown strong signs of recovery (Tang et al., 2025 ). With mobile-based booking becoming the dominant mode of travel reservation, the penetration and convenience of mobile devices have emerged as key drivers of market growth. According to Fastdata’s 2024 industry report, domestic tourist trips reached RMB 5.615 billion (a year-on-year increase of 14.8%), and total tourism expenditure grew by 17.1% to RMB 5.75 trillion (Fastdata, 2024 ). These figures suggest that both travel frequency and spending are approaching pre-pandemic levels, underscoring the increasing prominence of digitally mediated travel decision-making. Tourism is an information-intensive decision process in which travelers must evaluate complex and uncertain information relating to destination safety, attraction appeal, transportation accessibility, cultural experience, and potential constraints (Dai et al., 2022 ; Fang, 2023 ; Halkiopoulos et al., 2022 ). Traditionally, such information has been accessed through interpersonal communication, online reviews, travel blogs, guidebooks, and official destination marketing sources (Hu & Yang, 2021 ; Rehman Khan et al., 2021 ; Zelenka et al., 2021 ). The recent development of generative artificial intelligence, however, has introduced a new class of interactive decision-support agents into the tourism domain (Tuo et al., 2025 ). Large language model (LLM) based systems no longer function merely as background algorithms or automated service responders; instead, they increasingly act as active “decision participants,” assisting travelers in selecting destinations, generating itineraries, comparing alternatives, and assessing risks (Doborjeh et al., 2022 ; Doğan & Niyet, 2024 ; Samala et al., 2022 ). In particular, Dwivedi et al. ( 2024 ) pointed out that younger travelers are beginning to rely on AI tools during the early stages of travel planning, consulting them for personalized suggestions and itinerary recommendations. In the Chinese market, domestic platforms such as Doubao and DeepSeek have rapidly evolved into multimodal AI travel assistants capable of producing integrated text–image content, map-based planning outputs, and highly structured travel guidance (Xu et al., 2025 ; Zhu & Yu, 2025 ). Compared with conventional user-generated content, these AI-generated recommendations are often perceived as more conversational, personalized, and cognitively accessible, thereby reshaping how travelers acquire and evaluate tourism information (Alharbi et al.). At the same time, the increasing reliance on AI-generated content (AIGC) has raised important debates regarding reliability, authenticity, and epistemic trust (Li et al., 2025 ; Sable et al., 2025 ; Tan et al., 2025 ). LLMs may produce fabricated or inaccurate information (“hallucinations”), which poses particular risks in travel-related decision contexts (Jacob & Habibullah, 2025 ; Maleki et al., 2024 ; Ye et al., 2023 ). Although AI outputs often appear coherent and professional, their underlying data provenance and processing logic remain opaque, making it difficult for users to assess credibility or bias (Waqas & Naseem, 2025 ). Moreover, because AI systems do not possess lived experiential knowledge, their recommendations may lack the “situated authenticity” that travelers value when imagining or evaluating destination experiences (Ferhataj & Memaj, 2024 ; Selgas-Cors, 2025 ). Multimodal content may further amplify this tension by constructing hyper-realistic visualized travel scenarios that enhance persuasive appeal while obscuring uncertainty. Against this backdrop, a critical question emerges: through what psychological and cognitive mechanisms do travelers evaluate AI-generated travel information, and how do these evaluations shape their intention to accept AI-recommended destinations? While prior tourism research has extensively examined online reviews and user-generated content (Sánchez-Franco & Rey-Tienda, 2024 ; Yamagishi et al., 2024 ), relatively little is known about information adoption in contexts where content is algorithmically generated, conversational, and accompanied by dynamic system-generated reasoning traces. A particularly salient feature in this emerging landscape is DeepSeek’s Deep Thinking Function, which discloses the AI’s reasoning path by explicitly presenting its interpretive process, analytical logic, and content-generation strategy (Liu et al., 2025 ). This form of process transparency may not only alter travelers’ perceptions of argument quality and source credibility, but may also reshape their judgment of information usefulness when engaging with AI-mediated travel planning. To address this research gap, the present study draws upon the Information Adoption Model (IAM) to develop a theoretical framework explaining travelers’ responses to AI-generated travel recommendations in the context of DeepSeek’s Deep Thinking Function. We propose that the disclosure of AI reasoning enhances travelers’ cognitive evaluations of argument quality and source credibility, which in turn strengthen perceived information usefulness and subsequently promote destination acceptance intention. By empirically examining these relationships, this study advances theoretical understanding of information persuasion in AI-mediated tourism environments and offers practical insights for the design of transparent AI platforms and destination marketing strategies. 2. Literature Review 2.1 AI-generated Content and Information Processing in Tourism Existing research on digital tourism information has primarily focused on online reviews (Hu & Yang, 2021 ; Sun et al., 2024 ; Zelenka et al., 2021 ; Zhang et al., 2024 ), electronic word-of-mouth (eWOM) (Nguyen, 2025 ; Song & Song, 2025 ; Vu Dinh et al., 2025 ), and user-generated content (UGC) (Liang et al., 2024 ; Sujatmiko et al., 2025 ; Wijaya et al., 2025 ), highlighting their roles in shaping travelers’ attitude formation, risk perception, and behavioral intention. Prior studies indicate that message credibility, review valence, and argument relevance significantly influence travelers’ decision-making processes (Nguyen, 2025 ). However, these studies implicitly assume that information originates from human senders, whose experience, authenticity, and emotional engagement constitute core evaluative criteria. With the emergence of AIGC, this assumption is increasingly challenged. Unlike UGC, Cao et al. ( 2025 ) argued that AIGC is algorithmically constructed rather than experientially lived, and thus lacks embodied travel experience. While recent studies acknowledge the potential of AI tools as information mediators in tourism planning, most existing works treat AI outputs as functionally equivalent to textual reviews or recommendation messages (Guo et al., 2025 ; Zhang et al., 2025 ). This analytical equivalence obscures the fact that AI systems introduce an epistemically distinct information ecology, in which authorship, knowledge provenance, and agency are fundamentally reconfigured. More importantly, prior research has largely examined AIGC from an outcome-oriented perspective, such as usefulness perceptions (Li et al., 2025 ), trust formation (Zhou & Lu, 2025 ), or satisfaction (Tan et al., 2025 ), while overlooking the cognitive processes through which travelers interpret and validate AI-generated recommendations. As a result, we still know little about how travelers reconcile uncertainty, authenticity concerns, and credibility ambiguity when engaging with AI-mediated destination suggestions. 2.2 Argument Quality and Source Credibility in the Information Adoption Model The Information Adoption Model has been extensively used to explain how individuals evaluate information in online environments, emphasizing the effects of argument quality and source credibility on perceived usefulness and adoption intention (Gökerik, 2024 ; Horrich et al., 2024 ; Long et al., 2024 ). In tourism research, IAM has been widely applied to contexts such as social media reviews, travel blogs, and platform recommendations (Islam et al., 2025 ; Kumar & Lata, 2024 ; Shia et al.; Wei et al., 2025 ). These studies consistently demonstrate that logical coherence, informativeness, and message relevance promote users’ adoption of tourism information. However, the application of IAM in existing studies is characterized by two major limitations. First, most IAM-based tourism studies conceptualize source credibility in terms of reviewer expertise, experience, or identity cues, again presupposing a human information sender (Çelik & Aslan, 2025 ; Gökerik, 2024 ; Kumar & Lata, 2024 ). When the source becomes an artificial intelligence agent with unknown training data and opaque reasoning processes, the meaning of credibility can no longer be reduced to traditional notions of expertise or trustworthiness. Thus, prior IAM research provides limited explanatory power in contexts where agency is computational rather than human. Second, argument quality has been measured primarily through surface-level linguistic indicators such as clarity, completeness, or informativeness (Lata & Rana, 2025 ; Tseng & Wu, 2024 ). While these dimensions capture textual persuasiveness, they fail to address how arguments are constructed, justified, and rationalized within AI-generated recommendations. In other words, IAM research has focused on what the message says, but rarely on how the message is produced and justified a conceptual omission that becomes critical in the era of generative AI. 2.3 Transparency, Explainability, and AI Reasoning Disclosure A growing body of research in human–AI interaction suggests that system transparency and reasoning disclosure may enhance user trust and perceived legitimacy by revealing the internal logic behind AI outputs. However, empirical findings remain inconsistent. Some studies indicate that transparency improves users’ confidence and understanding (Radanliev, 2025 ; F. Wang et al., 2025 ), while others suggest that excessive disclosure may increase cognitive load or even expose uncertainty in AI reasoning (Aquilino et al., 2024 ; Lee & Cha, 2025 ; Ngo, 2025 ), thereby weakening perceived credibility. Despite these debates, very few tourism studies have examined process-level transparency in AI travel assistants. Existing tourism literature tends to conceptualize trust in AI tools as a static attribute, without considering how dynamic reasoning traces, such as DeepSeek’s Deep Thinking Function, actively reshape users’ evaluative criteria. This omission is particularly salient because transparency in AI contexts functions not merely as an informational supplement, but as a meta-argument that frames how travelers interpret argument quality and credibility itself (Afroogh et al., 2024 ). In this sense, Deep Thinking Function does not simply provide more information; rather, it discloses the epistemic structure behind AIGC, potentially transforming travelers’ perceptions from judging message content to evaluating the validity of the reasoning process. However, to date, no empirical tourism study has systematically examined whether the disclosure of AI reasoning processes strengthens or weakens perceived argument quality, nor whether source credibility shifts from human-based trust toward system-based cognitive trust as a result of such transparency. 3. Hypotheses and Research Framework 3.6. Deep Thinking Function as a Process-Level Epistemic Mechanism Unlike traditional online information environments, AI-generated travel recommendations are produced through computational reasoning processes that users are typically unable to observe. The Deep Thinking Function introduces a novel form of process-level transparency by revealing how the system interprets user instructions, organizes contextual constraints, and constructs its recommendation logic. Rather than simply adding more informational content, it operates as a meta-persuasive mechanism that shapes how travelers interpret and evaluate the cognitive validity of AI-generated outputs. From a cognitive perspective, the disclosure of reasoning steps may strengthen users’ perceptions that AI recommendations are grounded in explicit argumentation rather than opaque algorithmic inference. Such transparency, as noted by Aquilino et al. ( 2024 ), enables users to independently assess the generative pathway behind the content. When travelers are able to trace the logical trajectory underlying destination suggestions, for instance, trade-off comparisons, constraint filtering, or contextual matching, they may attribute greater coherence, rigor, and decision relevance to the final argument. Under this condition, where users actively participate in evaluating the content generation process, the skepticism toward the authenticity of AIGC outputs highlighted by prior scholars is likely to be reduced or even eliminated. In this sense, the way in which Deep Think enhances perceived argument quality does not lie in altering the information itself, but in strengthening users’ confidence in the reasoning logic that underpins it. Accordingly, we propose: H1. Deep Thinking Function positively influences travelers’ perception of argument quality in AI-generated travel information. In terms of source credibility, traditional conceptualizations of source credibility emphasize human-based attributes such as experience, identity cues, or reviewer expertise (Gökerik, 2024 ; Lata & Rana, 2025 ; Liu et al., 2025 ). In AI-mediated environments, however, Ngo ( 2025 ) pointed out that credibility shifts from social trust to epistemic trust. It means that the extent to which users have confidence in the system’s reasoning competence and decision logic becomes the primary determinant of credibility. Deep Thinking Function directly facilitates this shift by revealing how recommendations are generated, thereby reducing ambiguity surrounding the system’s knowledge provenance and analytical foundations. Process disclosure signals that the AI system is not arbitrarily producing destination suggestions, but is instead engaging in structured and explainable judgment. This visibility, as noted by Aquilino et al. ( 2024 ), may foster perceptions of competence, objectivity, and methodological soundness. Those attributes that collectively constitute a form of system-level credibility distinct from traditional human expertise. Conversely, when the reasoning process remains opaque, users may perceive AI recommendations as black-box outputs (Afroogh et al., 2024 ), weakening credibility perceptions even when the message content appears coherent. Therefore, when users develop a high level of cognitive trust in the AI system, they are more likely to perceive the content it generates as credible。 Therefore, we hypothesize: H2. Deep Thinking Function positively influences travelers’ perception of source credibility in AI-generated travel information. 3.2 Argument Quality and Perceived Information Usefulness Within the IAM, according to Sussman and Siegal ( 2003 ), argument quality refers to the extent to which information is perceived as logical, accurate, relevant, and well-supported in relation to users’ decision needs. Prior research in online tourism contexts has consistently shown that messages with coherent structure, rich informational content, and decision relevance are more likely to be perceived as useful during trip planning (Çelik & Aslan, 2025 ; Islam et al., 2025 ; Lata & Rana, 2025 ). These findings, however, have largely been established in human-generated information environments, where argument quality is implicitly associated with experiential authenticity and narrative credibility. In AI-mediated contexts, argument quality acquires a different meaning. AI-generated recommendations are constructed through computational reasoning processes rather than lived travel experiences. As such, travelers may evaluate argument quality less in terms of emotional authenticity and more in terms of logical justification, internal coherence, and contextual adequacy. When AI-generated outputs provide detailed explanations, rationalized recommendations, and structured reasoning, travelers may perceive them as cognitively robust and practically applicable to their planning needs. Conversely, when arguments appear generic, superficial, or weakly justified, travelers may interpret AI outputs as template-like or detached from real travel scenarios, thereby reducing perceived usefulness. Building on this perspective, argument quality in the AI setting functions as a cognitive cue that signals whether the recommendation is capable of supporting informed travel decision-making. Accordingly, we propose: H3. Argument quality of AI-generated travel information positively influences travelers’ perceived information usefulness. 3.3 Source Credibility and Perceived Information Usefulness Source credibility in IAM traditionally refers to perceptions of expertise and trustworthiness associated with human reviewers or information contributors (Lata & Rana, 2025 ; Sussman & Siegal, 2003 ). In tourism studies, credibility has typically been inferred from reviewer identity cues, travel experience indicators, or platform reputation (Lata & Rana, 2025 ; Shia et al.; Tseng & Wu, 2024 ). Such conceptualizations presume that the source of information is a human agent with experiential authority. In AI-generated environments, this assumption becomes problematic. The source of content shifts from identifiable human contributors to computational systems whose training data, reasoning mechanisms, and epistemic grounding are often opaque. As a result, credibility judgments are less about the personal reliability of a reviewer and more about system competence, transparency, and epistemic legitimacy (Afroogh et al., 2024 ; Schneier, 2025 ). Travelers may evaluate AI credibility based on whether the system demonstrates consistency, rational explanation, contextual sensitivity, and reliability across responses. When AI outputs appear well-reasoned, contextually aligned, and technically competent, travelers are more likely to perceive the system as a credible informational source, which subsequently enhances perceived usefulness. By contrast, perceived ambiguity, unexplained assertions, or generic recommendations may weaken credibility and reduce users’ confidence in applying the information to real decisions. Thus, we hypothesize: H4. Source credibility of AI-generated travel information positively influences travelers’ perceived information usefulness. 3.4 Perceived Information Usefulness and Information Adoption Perceived information usefulness, as highlighted by Horrich et al. ( 2024 ), represents a central cognitive mechanism through which online information influences behavioral intention in IAM. In tourism contexts, useful information not only facilitates cognitive evaluation of alternatives but also reduces uncertainty and decision effort, thereby increasing travelers’ willingness to adopt the information (Lata & Rana, 2025 ; Wei et al., 2025 ). In AI-mediated travel planning, usefulness plays an even more pivotal role because travelers must decide whether to rely on algorithmically constructed recommendations rather than human-experienced advice. When AI-generated information is perceived as practically relevant, contextually appropriate, and decision-supportive, travelers are more likely to accept the recommended information as a viable option. Conversely, when AI suggestions are perceived as abstract, generic, or insufficiently grounded, travelers may disengage or revert to conventional human-based information sources. Therefore, we propose: H5. Perceived information usefulness positively influences information adoption. 3.5 Information Adoption and Destination Acceptance Intention Within AI-mediated travel planning environments, travelers’ behavioral tendencies toward destinations are shaped not only by their cognitive judgments of information quality and credibility, but also by the extent to which they internalize and adopt the AI-generated recommendations. Prior IAM research has consistently demonstrated that once information is perceived as useful and epistemically trustworthy, individuals are more likely to integrate it into their decision-making processes, thereby translating cognitive evaluation into behavioral acceptance (Bao & Zhu, 2025 ; Çelik & Aslan, 2025 ; Lata & Rana, 2025 ). Information adoption reflects travelers’ willingness to rely on and incorporate AI-generated recommendations when forming travel expectations and planning choices. When travelers perceive that AI-generated content offers analytically sound reasoning, reliable justification, and context-appropriate guidance, they are more inclined to regard such recommendations as actionable decision inputs rather than merely informative references. This aligns with findings in technology-assisted tourism decision-making literature (Horrich et al., 2024 ; Islam et al., 2025 ), which indicate that the assimilation of mediated information can enhance perceived destination attractiveness, reduce cognitive risk, and strengthen commitment toward prospective travel choices. Accordingly, it is reasonable to expect that higher levels of information adoption will positively influence travelers’ intention to accept and consider the recommended destination. In other words, when travelers internalize AI-mediated recommendations as credible and useful, they are more likely to express willingness to visit, endorse, or further explore the suggested destination Thus, we hypothesize: H6: Information adoption positively influences traveler’s destination acceptance intention. 3.6 Research Framework Based on the above discussion, we constructed the following research framework (see Fig. 1 ). Drawing on the Information Adoption Model, this study conceptualizes argument quality and source credibility as the core antecedents shaping travelers’ perceptions of information usefulness, which subsequently influence their destination acceptance intention. Within this persuasion pathway, the Deep Thinking Function is incorporated as a key mechanism variable that operates at the process-transparency level rather than at the message-content level. 4.Methodology This study adopts a quantitative, cross-sectional research design to empirically test the proposed IAM-based conceptual model and the hypothesized relationships among Deep Thinking Function, argument quality, source credibility, perceived information usefulness, and traveler’s destination acceptance intention. A structured online questionnaire survey was employed as the primary data collection method, as it enables efficient access to digitally active respondents and aligns with the platform-mediated interaction context examined in this study. 4.1 Instrument Design and Measurement All constructs in this study were measured using multi-item scales adapted and adopted from validated instruments in prior literature, supplemented by self-developed items where necessary to capture the characteristics of the research context (see Table 1 ). The operational definitions of argument quality and source credibility follow prior studies grounded in the IAM. Argument quality focuses on the perceived clarity, logical consistency, and evidential support of the presented information, while source credibility reflects users’ evaluations of the trustworthiness and perceived expertise of the information source. Table 1 Measurement Items Indicators Items Sources Deep Thinking Function(DTF) DTF1: The AI travel assistant clearly explained the reasoning steps behind its destination recommendation Hoang and Nguyen ( 2025 ) DTF2: The Deep Thinking function helped me understand how the recommendation was generated Self-developed DTF3: The AI provided transparent justification for why certain options were selected or filtered out DTF4: The reasoning process disclosed by the AI made the recommendation appear more logical and trustworthy. Augment Quality (AQ) AQ1: The destination recommendation provided by the AI is logically structured and well justified S. Wang et al. ( 2025 ) AQ2: The information presented in the recommendation is detailed and well supported by relevant reasons AQ3: The arguments used by the AI to support the recommendation are strong and convincing AQ4: Overall, the content of the recommendation appears coherent and reasonable Source Credibility(SC) SC1: I believe the AI system is competent in analyzing travel-related information S. Wang et al. ( 2025 ) SC2: The AI appears knowledgeable and reliable when generating destination recommendations SC3: I trust the AI’s analytical process when evaluating travel options SC4: Overall, I consider the AI to be a credible source of travel planning information Perceived Information Usefulness(PIU) PIU1: The AI-generated recommendation is useful for my travel planning Hoang and Nguyen ( 2025 ) PIU2: The information provided by the AI helps me make better travel decisions Chiengkul et al. ( 2025 ) PIU3: The recommendation improves the efficiency of my travel planning process Hoang and Nguyen ( 2025 ) PIU4:Overall, I find the AI recommendation beneficial for evaluating travel destinations Chiengkul et al. ( 2025 ) Information Adoption(IA) IA1:I am willing to rely on the AI-generated recommendation when planning my trip Chiengkul et al. ( 2025 ) IA2: I would take the AI’s suggestion into account when choosing a travel destination. IA3: I consider the AI recommendation as an important reference for my decision-making IA4: I am likely to follow the recommendation provided by the AI travel assistant Destination Acceptance Intention (DAI) DAI1: I am willing to consider visiting the recommended destination Hoang and Nguyen ( 2025 ) DAI2:I would be interested in learning more about the recommended destination DAI3: I am likely to shortlist this destination as a potential travel option DAI4: Overall, I have a positive intention to accept the recommended destination The Deep Thinking Function construct was introduced as a newly added mechanism variable within the IAM framework, designed to assess the perceived impact of reasoning-process transparency on users’ evaluations of argument quality and source credibility. Destination acceptance intention was measured as a behavioral intention construct capturing users’ willingness to continue engaging with the AI-generated recommendation or to base future travel decisions on the presented information. To obtain respondents’ genuine attitudes as accurately as possible, Russo et al. ( 2021 ) had suggested that the five-point Likert scale is an appropriate and effective response format for social science surveys. Accordingly, all measurement items in this study were assessed using a five-point Likert scale ranging from 1 (“strongly disagree”) to 5 (“strongly agree”). A small-scale pilot test was conducted prior to the main survey to ensure content validity, wording clarity, and consistency of interpretation across respondents. Minor revisions were subsequently made to improve item comprehensibility and reduce potential ambiguity, as well as to mitigate the risk of common method bias associated with self-reported data. 4.2 Sampling Strategy and Data Collection This study adopted a non-probability purposive sampling strategy with the aim of recruiting tourism users who had recently engaged in AI-mediated information environments similar to the focal context of this research. Such an approach is theoretically appropriate because, as Stratton ( 2023 ) argued, purposive sampling prioritizes participants who possess direct experiential relevance and cognitive involvement in the phenomenon under investigation, rather than treating the general population as a homogeneous group of information recipients. Data were collected between May and July 2025 through an online questionnaire administered via the Wenjuanxing survey platform. Participation was entirely voluntary and anonymous, and all respondents were informed of the academic purpose of the study. Screening questions were embedded in the survey to verify the relevance of respondents’ prior experience and to exclude inattentive or ineligible participants. Following a rigorous data-cleaning procedure, including attention-check verification, response-time filtering, and the removal of patterned or incomplete submissions, a total of 260 valid responses were retained for analysis. The Table 2 shows details of the demographic profile. Table 2 Demographic Profile Variable Categories Frequency Percentage Gender Male 122 46.92 Female 138 53.08 Total 260 100 Age 18–28 80 30.77 29–40 85 32.69 Above 41 95 36.54 Total 260 100 Education Senior high school or below 81 31.15 Bachelor’s or associate degree 76 29.23 Master’s degree 62 23.85 PhD 41 15.77 Total 260 100 Frequency of using AI tools Multiple times a day 168 64.62 Everyday 67 25.77 3–4 days per week 23 8.84 Very rare 2 0.77 Total 260 100 Although the reliance on online purposive sampling may limit the statistical generalizability of the findings, as noted by Rahman ( 2023 ), it strengthens the study’s contextual validity by ensuring that the sample reflects active users embedded within the AI-driven digital persuasion environment under examination. This trade-off is acknowledged as a methodological limitation, yet it also represents a theoretically meaningful design choice that remains consistent with prior empirical studies employing the IAM. 5. Results Following the recommendations of Hair et al. ( 2019 ), this study employed partial least squares structural equation modeling (PLS-SEM) to validate the proposed research model using SmartPLS 4.0. The results are reported in two stages: (1) assessment of the measurement model and (2) evaluation of the structural model. 5.1 Assessment of Measurement Model 5.1.1 Reliability and Validity To ensure the reliability of the constructs, Cronbach’s alpha and composite reliability (CR) coefficients were examined. Jarupunphol et al. ( 2024 ) suggested that Cronbach’s alpha values exceeding 0.70 indicate satisfactory construct reliability. As shown in Table 3 , the Cronbach’s alpha values for all constructs in this study are above the recommended threshold. The lowest value is observed for Source Credibility (SC) at 0.727, which remains well above the minimum criterion, confirming adequate reliability across all constructs. Table 3 Results of Factor Loadings, Cronbach’s alpha, composite reliability and AVE Constructs Items Factor Loadings Cronbach's alpha Composite reliability AVE Deep Thinking Function (DTF) DTF1 0.811 0.820 0.881 0.650 DTF2 0.790 DTF3 0.830 DTF4 0.792 Argument Quality (AG) AG1 0.757 0.779 0.858 0.601 AG2 0.783 AG3 0.792 AG4 0.770 Source Credibility (SC) SC1 0.730 0.727 0.830 0.549 SC2 0.773 SC3 0.758 SC4 0.701 Perceived Information Usefulness (PIU) PIU1 0.809 0.771 0.853 0.593 PIU2 0.774 PIU3 0.706 PIU4 0.786 Information Adoption (IA) IA1 0.811 0.779 0.858 0.601 IA2 0.800 IA3 0.738 IA4 0.750 Destination Acceptance Intention (DAI) DAI1 0.829 0.789 0.863 0.613 DAI2 0.794 DAI3 0.781 DAI4 0.723 Composite reliability coefficients were also calculated to further verify internal consistency. Consistent with Hair et al. ( 2019 ), a CR value of 0.70 or higher was considered acceptable. As reported in Table 3 , all constructs exceed this threshold, indicating strong composite reliability and stable internal consistency. Convergent validity was assessed using the Average Variance Extracted (AVE). According to Shrestha ( 2021 ), AVE values above 0.50 indicate that a construct explains more than half of the variance in its indicators. In this study, all constructs exhibit AVE values greater than 0.50 (see Table 1 ), demonstrating satisfactory convergent validity and confirming that the indicators load appropriately on their respective constructs. 5.1.2 Discriminant Validity In addition to reliability and convergent validity, discriminant validity was evaluated using both the Fornell–Larcker criterion and the Heterotrait–Monotrait (HTMT) ratio. Table 4 presents the Fornell–Larcker results. A construct is considered to demonstrate discriminant validity when the square root of its AVE is greater than its correlations with other constructs (Afthanorhan et al., 2021 ; Rasoolimanesh, 2022 ). The results indicate that all constructs meet this criterion, suggesting the absence of discriminant validity concerns. Table 4 The results of Fornell–Larcker criterion AG DAI DTF IA PIU SC AG 0.775 DAI 0.655 0.783 DTF 0.674 0.702 0.806 IA 0.558 0.626 0.649 0.775 PIU 0.689 0.677 0.662 0.627 0.77 SC 0.629 0.651 0.671 0.566 0.675 0.741 To further strengthen the evaluation, the HTMT ratio was also examined. As noted by Dirgiatmo ( 2023 ), HTMT is a more rigorous and contemporary approach to assessing discriminant validity. Following Hair et al. ( 2019 ), an HTMT value below 0.90 was adopted as the threshold. As shown in Table 5 , all HTMT values fall below 0.90, thereby confirming that the constructs are empirically distinct from one another and that discriminant validity is satisfactorily established. Table 5 The results of HTMT AG DAI DTF IA PIU SC AG DAI 0.831 DTF 0.839 0.873 IA 0.714 0.792 0.811 PIU 0.884 0.860 0.830 0.804 SC 0.827 0.850 0.863 0.749 0.893 5.2 Assessment of Structural Model After establishing the adequacy of the measurement model, the structural model was assessed to examine the hypothesized relationships among the constructs. Prior to hypothesis testing, collinearity diagnostics were conducted using the variance inflation factor (VIF). All VIF values were below the recommended threshold of 5.0, indicating the absence of multicollinearity and confirming that the structural estimates were not biased by redundancy among predictors (Hair et al., 2019 ). The results are reported in Table 6 . Table 6 The results of VIF Indicators VIF Indicators VIF DTF1 1.778 PIU1 1.592 DTF2 1.746 PIU2 1.528 DTF3 1.920 PIU3 1.358 DTF4 1.710 PIU4 1.548 AG1 1.466 IA1 1.627 AG2 1.490 IA2 1.572 AG3 1.643 IA3 1.448 AG4 1.500 IA4 1.475 SC1 1.359 DAI1 1.767 SC2 1.382 DAI 2 1.580 SC3 1.424 DAI 3 1.628 SC4 1.315 DAI 4 1.377 Bootstrapping with 5,000 resamples was performed in SmartPLS 4.0 to obtain path coefficients, t-values, and significance levels for the hypothesized relationships. Table 7 reports the estimated path coefficients and statistical significance. Table 7 The results of path coefficients Path coefficients Standard deviation (STDEV) T statistics (|O/STDEV|) P values DTF -> AG 0.674 0.034 19.788 0.000 DTF -> SC 0.671 0.035 19.264 0.000 AG -> PIU 0.437 0.07 6.222 0.000 SC -> PIU 0.4 0.072 5.569 0.000 PIU -> IA 0.627 0.046 13.665 0.000 IA -> DAI 0.626 0.051 12.226 0.000 The results indicate that Deep Thinking Function exerts a significant positive effect on argument quality (β = 0.674, t = 19.788, p = 0.000) and source credibility (β = 0.671, t = 19.264, p = 0.000), supporting H1 and H2. This finding suggests that process-level transparency strengthens travelers’ cognitive evaluations of AI-generated content by enhancing both its perceived logical rigor and the epistemic trustworthiness of the AI system. Argument quality was found to have a significant positive influence on perceived information usefulness (β = 0.437, t = 6.222, p = 0.000, supporting H3. Similarly, source credibility also positively affected perceived information usefulness (β = 0.400, t = 5.569, p = 0.000), supporting H4. These results are consistent with prior IAM research, indicating that travelers assess AI-generated recommendations not only on message coherence, but also on confidence in the system’s reasoning competence. Furthermore, the results show that perceived information usefulness has an significant effect on information adoption (β = 0.627, t = 13.665, p = 0.000), supporting H5. Besides, information adoption demonstrated a significant positive effect on destination acceptance intention (β = 0.626, t = 12.226, p = 0.000), supporting H6. This confirms that when AI-generated travel information is perceived as useful and decision-relevant, travelers are more likely to accept and consider the recommended destination. Model explanatory power was assessed using R² values. The R² values for argument quality and source credibility indicate that a substantial proportion of variance is explained by Deep Thinking Function, while the R² value for information usefulness reflects the combined explanatory effects of argument quality and source credibility. Additionally, the R² value for destination acceptance intention demonstrates satisfactory predictive relevance of the model at the behavioral-intention level. Finally, the Q² values for the endogenous constructs was over 0, hence, predictive relevance was established The results are reported in Table 7 . Table 7 The results of R² and Q² R-square Q-square AG 0.454 0.445 DAT 0.391 0.259 IA 0.393 0.329 PIU 0.571 0.422 SC 0.45 0.438 6. Discussion This study investigates how DeepSeek’s Deep Thinking Function shapes travelers’ psychological evaluation and adoption of AI-generated destination recommendations in AI-mediated travel planning contexts. The findings show that, through an extension of the IAM, transparency enabled by Deep Thinking Function significantly enhances travelers’ perceptions of argument quality and source credibility, which subsequently strengthens perceived information usefulness and intention to accept the recommended destination. These results fill an important gap in existing research on the role of AI in tourism behavior, indicating that AI reasoning transparency is not merely an informational enhancement feature, but instead operates as a novel persuasive mechanism within human–AI interaction. Frist, at the overall structural level, the empirical results confirm the significant relationships among Deep Thinking Function, argument quality, source credibility, perceived information usefulness, information adoption, and destination acceptance intention, demonstrating strong explanatory power of the proposed pathway in the AI-mediated tourism context. On one hand, these findings reinforce the core assumption of the IAM that information adoption is jointly shaped by message attributes and individuals’ cognitive judgments regarding the underlying process through which information is generated (Sussman & Siegal, 2003 ). On the other hand, by conceptualizing Deep Thinking Function as a new AI-specific antecedent, this study refines the configuration and measurement of upstream cognitive variables and provides more concrete empirical evidence on the underlying mechanisms through which AI technologies influence tourism-related decision processes. Furthermore, the results reveal that travelers’ information adoption attitudes have evolved from a single rational-processing logic toward a dual cognitive–reflective structure. Prior tourism research has shown that travel decision-making is often characterized by contextual uncertainty, experiential risk, and reliance on mediated representations (Dai et al., 2022 ; Hu & Yang, 2021 ). Compared with routine consumption contexts, tourism choices, as noted by Zelenka et al. ( 2021 ), therefore involve higher involvement and stronger cognitive prudence. Unlike prior IAM studies that predominantly focus on online reviews (Bao & Zhu, 2025 ) or e-commerce platforms (Horrich et al., 2024 ), this study demonstrates that the presence of AI does not diminish travelers’ evaluative scrutiny. Instead, users strategically balance technological convenience against perceived credibility risks. In this sense, within contemporary AI-embedded tourism environments, AI functions not only as an information delivery instrument, but also as a cognitive agent that actively shapes interpretive frames, influencing how travelers understand, trust, and internalize tourism-related content. Finally, this study identifies the cognitive role of Deep Thinking Function in AI-mediated tourism information environments. The results show that Deep Thinking Function significantly enhances perceived argument quality and source credibility, suggesting that deeper cognitive engagement improves users’ evaluative precision in processing AI-generated recommendations. Prior studies have shown that applications such as intelligent itinerary planners, AI travel assistants, and automated recommendation systems produce information with characteristics such as personalized narratives, structured presentation, and implicit persuasive cues (Doğan & Niyet, 2024 ; Dwivedi et al., 2024 ; Ferhataj & Memaj, 2024 ). Although these features improve decision efficiency, they may also foster excessive dependence on algorithmic suggestions and obscure source transparency, thereby weakening users’ participatory reflection. Under such conditions, the appropriateness and reliability of AI-generated tourism recommendations may become uncertain. The present findings suggest that when users are able to cognitively engage with the reasoning process behind AI recommendations, they are better able to reflect on potential persuasive inducements, recognize boundaries between AI suggestions and embedded commercial intentions, and actively evaluate the authenticity and situational relevance of the generated content prior to adoption. This implies that Deep Thinking Function serves as a cognitive protection and corrective mechanism in AI-assisted tourism decision-making. In other words, technological convenience and cognitive vigilance must coexist. Although AI systems are designed to simplify decision processes, responsible tourism choices continue to rely on higher-order cognitive engagement and reflective judgment. 6.1 Theoretical Implications This study makes several important theoretical contributions to the intersection of AI cognition, persuasive communication, and technology-mediated tourism decision-making. First, the study extends the IAM into AI-mediated travel contexts. Prior IAM research has predominantly examined information evaluation in human-generated environments such as online reviews, peer communication, and social platforms (Bao & Zhu, 2025 ; Çelik & Aslan, 2025 ; Horrich et al., 2024 ). The present findings demonstrate that IAM mechanisms remain valid in AI-generated recommendation settings, but the evaluative focus of users shifts substantially. Rather than relying primarily on message clarity or human-based credibility cues, travelers begin to evaluate AI reasoning itself as an interpretive object. Through reasoning disclosure, Deep Thinking Function redirects users’ cognitive orientation from outcome-focused evaluation toward process-based rationality, indicating that perceived argument quality increasingly derives from assessments of logical coherence, constraint filtering, and contextual fit within travel scenarios. Second, this study reconceptualizes transparency disclosure as a mechanism variable, rather than a static technical attribute or ethical principle. Existing literature typically treats transparency as an interface feature or informational supplement. The findings of this research reveal that transparency actively reshapes users’ judgment criteria, reinforcing both argument quality and source credibility by functioning as a meta-argument that frames interpretations of content validity. In this regard, reasoning transparency does not merely increase awareness but establishes epistemic legitimacy, thereby enriching theoretical discussions in human–AI interaction and persuasive communication by evidencing how cognitive traceability can transform the meaning-construction process in AI-mediated environments. Third, the results advance theoretical understanding of trust formation in AI-mediated tourism decision contexts. Traditional tourism trust research has emphasized identity-based credibility, such as reviewer expertise, experiential similarity, or social proximity (Bao & Zhu, 2025 ; Long et al., 2024 ). In contrast, this study identifies a shift toward system-centric credibility, wherein user confidence is grounded in perceptions of analytical competence, inferential rigor, and decision-logic reliability. This suggests that trust in AI is not simply transplanted from interpersonal trust models, but instead reflects a qualitatively distinct form of cognitive trust in reasoning capacity, a form of trust particularly salient in high-involvement and uncertainty-laden contexts such as travel planning. 6.2 Practical Implications This study also yields several important practical implications for AI platform designers, tourism marketers, and destination management organizations. First, AI travel assistants should shift from output-only recommendation modes toward process-aware and reasoning-traceable presentation formats. The findings indicate that exposing reasoning logic can significantly enhance perceived argument quality, usefulness, and adoption confidence. Rather than merely delivering instant itineraries, AI systems may benefit from revealing constraint-filtering steps, trade-off comparisons among alternative destinations, and contextual justification based on traveler preferences or situational demands. Such design allows users to audit, interpret, and participate in the recommendation process, thereby strengthening engagement, transparency perception, and decision assurance in complex travel scenarios. Second, tourism platforms should treat transparency as a persuasion strategy rather than a compliance-oriented disclosure feature. Deep Think-type reasoning displays function as a cognitive scaffold that legitimizes AI-generated recommendations by signaling analytical rigor, internal coherence, and methodological soundness. In destination promotion contexts, this suggests that argument quality can be enhanced without intensifying promotional rhetoric, and credibility can be strengthened without relying on influencer identity cues or social endorsements. Accordingly, AI-based persuasion is likely to complement rather than replace human-generated travel narratives, forming a hybrid communication ecology in which algorithmic reasoning and experiential storytelling jointly shape travelers’ perception and trust. Third, the findings provide strategic insights for tourism marketing in the post-AIGC environment. As travelers increasingly consult AI systems during the early stages of trip planning, destinations are no longer competing only within social review ecosystems, but also within AI reasoning pipelines and decision-logic structures. This implies that content formats aligned with AI cognitive processing, such as clearly structured attribute descriptions, contextual relevance markers, and scenario-specific application cues, are more likely to be framed as high-quality recommendations. Consequently, destination branding strategies may need to evolve from narrative persuasion toward decision-logic compatibility, ensuring that tourism information not only attracts human audiences but is also interpretable, retrievable, and argumentatively meaningful within AI-mediated planning environments. 7. Limitations and Future Research Although this study offers meaningful empirical insights, several limitations should be acknowledged, which also provide avenues for future research. First, the research adopted a cross-sectional survey design, meaning that all data were collected at a single point in time. While this approach is appropriate for examining current perceptions and behavioral intentions, it limits the ability to infer causal relationships or capture dynamic changes in attitudes and behaviors. Future studies may employ longitudinal or experimental designs to better trace temporal variations and strengthen causal inferences. Second, the measurement of key constructs relied on self-reported data, which, despite the pilot testing and procedural remedies adopted to reduce response bias, may still be subject to social desirability effects or common method variance. Subsequent research could incorporate multi-source or behavioral data, such as platform usage logs, objective performance indicators, or peer evaluations, to enhance measurement robustness and triangulate findings. Third, the sample was drawn from a specific population and contextual setting, which may constrain the generalizability of the results. Although stratified sampling was used to ensure representation, cultural, institutional, or regional differences may influence respondents’ perceptions and decision-making processes. Future research is encouraged to replicate the study across different demographic groups, geographical contexts, or cultural environments, and to conduct cross-cultural comparative analyses to test the external validity of the model. Finally, this study concentrated on quantitative analysis, which, although valuable for hypothesis testing, may not fully capture the nuanced cognitive and emotional mechanisms underlying respondents’ attitudes and behavioral responses. Subsequent research could adopt mixed-methods or qualitative approaches, such as interviews, focus groups, or text mining of user-generated content, to provide deeper interpretive insights and complement the quantitative findings. Taken together, addressing these limitations in future research will not only enhance methodological rigor and contextual applicability, but also contribute to the ongoing refinement and extension of the theoretical framework examined in this study. Abbreviations IAM Information Adoption Model LLM Large Language Model AIGC AI-generated content UGC User-generated content DTF Deep Thinking Function AQ Augment Quality SC Source Credibility PIU Perceived Information Usefulness IA Information Adoption DAI Destination Acceptance Intention VIF Variance inflation factor AVE Average variance extracted Declarations Ethical approval and Consent to Participate The questionnaire was assessed and approved by Mianyang Polytechnic. An informed consent form signed by Mianyang Polytechnic was provided to every participant before the survey. All data remains confidential. Consent for Publication Not applicable Funding This work was supported by Open Project Fund of Key Laboratory of Digital Innovation of Tianfu Culture, Sichuan Provincial Department of Culture and Tourism (Project No. TFWH-2025-40) and Hainan provincial Natural Science Foundation of China (Project No. 623RC510). Author Contribution Conceptualization, S.C. and JY.L.; methodology, FB.W; software, JY.X. and XJ.M; validation, K.Z.A; data curation, K.Z.A.; writing—original draft preparation, S.C. and JY.L; writing—review and editing, FB.W. and JY.L.; visualization, K.Z.A. All authors have read and agreed to the published version of the manuscript. Data Availability The datasets used and analysed during the current study are available from the corresponding author on reasonable request References Afroogh, S., Akbari, A., Malone, E., Kargar, M., & Alambeigi, H. (2024). Trust in AI: progress, challenges, and future directions. Humanities and Social Sciences Communications , 11 (1), 1-30. Afthanorhan, A., Ghazali, P. L., & Rashid, N. (2021). Discriminant validity: A comparison of CBSEM and consistent PLS using Fornell & Larcker and HTMT approaches. Journal of Physics: Conference Series, Alharbi, D., Alharbi, R., Alafif, T., Jassas, M., Alfattni, G., Al-Luhaybi, M., Alotaibi, H., Gharawi, A., Zia, S., & Kazalah, F. DOSTE: Domain Specific LLMs for Saudi Tourism and Entertainment. Aquilino, L., Bisconti, P., & Marchetti, A. (2024). Trust in AI: Transparency, and uncertainty reduction. Development of a new theoretical framework. CEUR workshop proceedings, Bao, Z., & Zhu, Y. (2025). Understanding online reviews adoption in social network communities: an extension of the information adoption model. Information Technology & People , 38 (1), 48-69. Cao, Y., Li, S., Liu, Y., Yan, Z., Dai, Y., Yu, P., & Sun, L. (2025). A survey of ai-generated content (aigc). ACM Computing Surveys , 57 (5), 1-38. Çelik, K., & Aslan, A. (2025). The Impact of Electronic Word of Mouth (eWOM) on Visit Intention within the Framework of the Information Adoption Model: A Study on Instagram Users. International Journal of Marketing, Communication and New Media , 12 (23). Chiengkul, W., Kumjorn, P., Tantipanichkul, T., & Suphan, K. (2025). Engaging with AI in tourism: a key to enhancing smart experiences and emotional bonds. Asia-Pacific Journal of Business Administration . Dai, F., Wang, D., & Kirillova, K. (2022). Travel inspiration in tourist decision making. Tourism Management , 90 , 104484. Dirgiatmo, Y. (2023). Testing the discriminant validity and heterotrait–monotrait ratio of correlation (HTMT): A case in Indonesian SMEs. In Macroeconomic Risk and Growth in the Southeast Asian Countries: Insight from Indonesia (pp. 157-170). Emerald Publishing Limited. Doborjeh, Z., Hemmington, N., Doborjeh, M., & Kasabov, N. (2022). Artificial intelligence: a systematic review of methods and applications in hospitality and tourism. International journal of contemporary hospitality management , 34 (3), 1154-1176. Doğan, S., & Niyet, İ. Z. (2024). Artificial intelligence (AI) in tourism. In Future Tourism Trends Volume 2: Technology Advancement, Trends and Innovations for the Future in Tourism (pp. 3-21). Emerald Publishing Limited. Dwivedi, Y. K., Pandey, N., Currie, W., & Micu, A. (2024). Leveraging ChatGPT and other generative artificial intelligence (AI)-based applications in the hospitality and tourism industry: practices, challenges and research agenda. International journal of contemporary hospitality management , 36 (1), 1-12. Fang, R. (2023). Proposing a cyclic model of tourist decision making: A review and integration of behavioral and choice-set models. Journal of Hospitality & Tourism Research , 47 (7), 1161-1186. Fastdata. (2024). 中国旅游行业年度报告 . https://www.ifastdata.com/wp-content/uploads/2025/04/Fastdata%E6%9E%81%E6%95%B0%EF%BC%9A%E4%B8%AD%E5%9B%BD% E6%97%85%E6%B8%B8%E8%A1%8C%E4%B8%9A%E5%B9%B4%E5%BA%A6%E6%8A%A5%E5%91%8A2024.pdf Ferhataj, A., & Memaj, F. (2024). Challanges And Opportunities Of Ai Implementation In Tourism: An Ethical And Technological Perspective. STUDIJOS–VERSLAS–VISUOMENĖ: DABARTIS IR ATEITIES ĮŽVALGOS , 1 (IX), 217-231. Gökerik, M. (2024). The enchantment of social media influencers: Analysing consumer attitudes through the lens of the information adoption model. OPUS Journal of Society Research , 21 (3), 125-139. Guo, Y., Yu, F., Lai, J., & Yuan, X. (2025). How is AIGC shaping the world: an analysis of bibliometrics. Information Research an international electronic journal , 30 (iConf), 679-689. Hair, J. F., Risher, J. J., Sarstedt, M., & Ringle, C. M. (2019). When to use and how to report the results of PLS-SEM. European business review , 31 (1), 2-24. Halkiopoulos, C., Antonopoulou, H., Gkintoni, E., & Aroutzidis, A. (2022). Neuromarketing as an indicator of cognitive consumer behavior in decision-making process of tourism destination—An overview. Transcending Borders in Tourism Through Innovation and Cultural Heritage: 8th International Conference, IACuDiT, Hydra, Greece, 2021, Hoang, D. S., & Nguyen, D. T. A. (2025). The role of AI assistants in promoting sustainable Halal tourism in non-Muslim destinations. Discover Sustainability , 6 (1), 705. Horrich, A., Ertz, M., & Bekir, I. (2024). The effect of information adoption via social media on sustainable consumption intentions: The moderating influence of gender. Current Psychology , 43 (18), 16349-16362. Hu, X., & Yang, Y. (2021). What makes online reviews helpful in tourism and hospitality? A bare-bones meta-analysis. Journal of Hospitality Marketing & Management , 30 (2), 139-158. IMARC. (2025). Online Travel Market Size, Share, Trends and Forecast by Service Type, Platform, Mode of Booking, Age Group, and Region, 2026-2034 . transforming ideas into impact. Retrieved 26 December from https://www.imarcgroup.com/online-travel-market Islam, M. T., Herjanto, H., Kumar, J., & Amin, M. (2025). Online Travel Reviews and Tourist Destination Choices: An Extension of the Information Adoption Model. Tourism Review International , 29 (1), 17-32. Jacob, S. L., & Habibullah, P. S. (2025). A Systematic Analysis and Review on Intrusion Detection Systems Using Machine Learning and Deep Learning Algorithms. Journal of Computational and Cognitive Engineering , 4 (2), 108-120. Jarupunphol, P., Ikonnikov, O., Roncevic, I., Kapustina, S., Kataeva, A., Parfjonovs, M., & Tsarev, R. (2024). Applying Cronbach’s alpha to ensure reliable online testing in e-learning environments. In Proceedings of the Computational Methods in Systems and Software (pp. 120-139). Springer. Kumar, A., & Lata, S. (2024). Are travellers' destination visit intentions influenced by social media? An information adoption theory perspective. International Journal of Indian Culture and Business Management , 33 (1), 103-120. Lata, S., & Rana, K. (2025). Evaluating the Influence of YouTube Vlogs on Hotel Booking Decisions: An Information Adoption Model Approach. In Building Power, Safety, and Trust in Virtual Communities (pp. 257-280). IGI Global. Lee, C., & Cha, K. (2025). Toward the dynamic relationship between AI transparency and trust in AI: a case study on ChatGPT. International Journal of Human–Computer Interaction , 41 (13), 8086-8103. Li, C., Cao, Q., Hua, S., & Tao, C.-W. (2025). When AI takes the wheel: The effectiveness of AI versus human-generated content in tourism marketing. Journal of Vacation Marketing , 13567667251393512. Liang, K., Liu, H., Shan, M., Zhao, J., Li, X., & Zhou, L. (2024). Enhancing scenic recommendation and tour route personalization in tourism using UGC text mining. Applied Intelligence , 54 (1), 1063-1098. Liu, S., Lin, L., Han, X., Qiu, Y., Wang, Z., & Zhang, F. (2025). Evaluating the Impact of Multimodal Cognitive and Informational Interventions on Rumor Detection in Large Language Models: A DeepSeek Case Study. 2025 5th International Conference on Artificial Intelligence, Big Data and Algorithms (CAIBDA), Long, J., Zaidin, N., & Mai, X. (2024). Social media influencer streamers and live-streaming shopping: examining consumer behavioral intention through the lens of the theory of planned behavior. Future Business Journal , 10 (1), 80. Maleki, N., Padmanabhan, B., & Dutta, K. (2024). AI hallucinations: a misnomer worth clarifying. 2024 IEEE conference on artificial intelligence (CAI), Ngo, V. M. (2025). Balancing AI transparency: Trust, Certainty, and Adoption. Information Development , 02666669251346124. Nguyen, L. P. (2025). Value-driven environmental advocacy in tourism: understanding the drivers of negative eWOM among tourists visiting island destinations. Journal of Hospitality and Tourism Insights , 1-20. Radanliev, P. (2025). AI ethics: Integrating transparency, fairness, and privacy in AI development. Applied Artificial Intelligence , 39 (1), 2463722. Rahman, M. M. (2023). Sample size determination for survey research and non-probability sampling techniques: A review and set of recommendations. Journal of Entrepreneurship, Business and Economics , 11 (1), 42-62. Rasoolimanesh, S. M. (2022). Discriminant validity assessment in PLS-SEM: A comprehensive composite-based approach. Data Analysis Perspectives Journal , 3 (2), 1-8. Rehman Khan, H. U., Lim, C. K., Ahmed, M. F., Tan, K. L., & Bin Mokhtar, M. (2021). Systematic review of contextual suggestion and recommendation systems for sustainable e-tourism. Sustainability , 13 (15), 8141. Russo, G. M., Tomei, P. A., Serra, B., & Mello, S. (2021). Differences in the use of 5-or 7-point likert scale: an application in food safety culture. Organizational Cultures , 21 (2), 1. Sable, N., Mahalle, P., Kadam, K., Sule, B., Joshi, R., & Deore, M. (2025). Deep learning-based approach for monitoring and controlling fake reviews. Journal of Computational and Cognitive Engineering , 4 (3), 377-386. Samala, N., Katkam, B. S., Bellamkonda, R. S., & Rodriguez, R. V. (2022). Impact of AI and robotics in the tourism sector: a critical insight. Journal of tourism futures , 8 (1), 73-87. Sánchez-Franco, M. J., & Rey-Tienda, S. (2024). The role of user-generated content in tourism decision-making: an exemplary study of Andalusia, Spain. Management Decision , 62 (7), 2292-2328. Schneier, B. (2025). AI and Trust. Communications of the ACM , 68 (8), 29-33. Selgas-Cors, M. (2025). Sociotechnical Transformation: A Systematic Review on the Impact of Artificial Intelligence on Society and Organizations. FinTech and Sustainable Innovation , 1-16. Shia, Y., Bidina, R. B. H., & Mamata, R. B. A Review of Relevant Research on Tourism Information Adoption Model on Social Media. Shrestha, N. (2021). Factor analysis as a tool for survey analysis. American journal of Applied Mathematics and statistics , 9 (1), 4-11. Song, H., & Song, X. (2025). The effect of heritage tourism interpretation media type on tourists’ eWOM: the moderating role of travel group size. Journal of Sustainable Tourism , 33 (10), 2240-2260. Stratton, S. J. (2023). Population sampling: Probability and non-probability techniques. Prehospital and Disaster Medicine , 38 (2), 147-148. Sujatmiko, S., Ar, D. P., Hamdat, A., & Salam, K. N. (2025). User-Generated Content (UGC) and Its Impact on Tourism Marketing: A Systematic Literature Review. Golden Ratio of Mapping Idea and Literature Format , 5 (2), 97-105. Sun, X., Wang, Z., Zhou, M., Wang, T., & Li, H. (2024). Segmenting tourists’ motivations via online reviews: An exploration of the service strategies for enhancing tourist satisfaction. Heliyon , 10 (1). Sussman, S. W., & Siegal, W. S. (2003). Informational influence in organizations: An integrated approach to knowledge adoption. Information systems research , 14 (1), 47-65. Tan, Z., Yang, Y., Yang, X., Liu, X., Wei, Q., & Kong, L. H. (2025). AI in Tourism Education: A Review of AIGC Teaching and Evaluation Tools. 2025 11th International Conference on Education and Training Technologies (ICETT), Tang, Y., Weng, G., Qin, S., & Pan, Y. (2025). Spatial and temporal evolution of tourism flows among 296 Chinese cities in the context of COVID-19: a study based on Baidu Index. Humanities and Social Sciences Communications , 12 (1), 1-15. Tseng, T.-L., & Wu, C.-C. (2024). Application of the Information Adoption Model and Technology Acceptance Model in Electronic Word-of-Mouth. International Journal of Performance Measurement , 14 (1). Tuo, Y., Wu, J., Zhao, J., & Si, X. (2025). Artificial intelligence in tourism: insights and future research agenda. Tourism Review , 80 (4), 793-812. Vu Dinh, H., Tran Huu, T., Nguyen Thi Bich, N., Nguyen Thi Ngoc, A., & Doan Van, T. (2025). Effects of eWOM toward tourism destination: a bibliometric analysis and future research directions. Consumer Behavior in Tourism and Hospitality , 20 (3), 445-459. Wang, F., Lopez, C., & Okazaki, S. (2025). Signaling transparency in the era of artificial intelligence. Internet Research , 1-25. Wang, S., Peng, K.-L., Huang, Z., & Ma, L. (2025). AI-Generated Videos: Influencing Trustworthiness, Awe, and Behavioral Intention in Space Tourism E-Commerce. Journal of Theoretical and Applied Electronic Commerce Research , 20 (4), 307. Waqas, M., & Naseem, A. (2025). Artificial Intelligence in Sustainable Industrial Transformation: A Comparative Study of Industry 4.0 and Industry 5.0. FinTech and Sustainable Innovation , 1 , A2-A2. Wei, Y., Liu, H., Zhuo, W., & Park, K.-S. (2025). The influence of social media attributes on impulsive travel intentions: Integrating the Stimulus–Organism–Response theory and Information Adoption Model. Sustainability , 17 (10), 4404. Wijaya, C. O., Wijaya, S., & Jaolis, F. (2025). The influence of social media content on attitude, destination image and intention of female Muslim travelers to visit halal destinations: comparison between UGC and FGC. Journal of Islamic Marketing , 16 (2), 402-427. Xu, T., Gao, M., Liu, X., Ye, Q., Ma, L., & Zhu, B. (2025). Intelligent Travel Avatar: An LLM-based Tourism Quadrupedal Robot. Yamagishi, K., Canayong, D., Domingo, M., Maneja, K. N., Montolo, A., & Siton, A. (2024). User-generated content on Gen Z tourist visit intention: a stimulus-organism-response approach. Journal of Hospitality and Tourism Insights , 7 (4), 1949-1973. Ye, H., Liu, T., Zhang, A., Hua, W., & Jia, W. (2023). Cognitive mirage: A review of hallucinations in large language models. arXiv preprint arXiv:2309.06794 . Zelenka, J., Azubuike, T., & Pásková, M. (2021). Trust model for online reviews of tourism services and evaluation of destinations. Administrative Sciences , 11 (2), 34. Zhang, L., Yang, S., Wang, W., Gao, X., & Liu, J. (2025). AIGC-Based Image and Video Generation Method: A Review. Journal of Computer-Aided Design & Computer Graphics , 37 (3), 361-384. Zhang, X., Cheng, L., & Ma, G. (2024). Eliciting eudaimonic well-being in the tourism experiential space: Evidence from online reviews. Tourism Management , 105 , 104955. Zhou, T., & Lu, H. (2025). The effect of trust on user adoption of AI-generated content. The Electronic Library , 43 (1), 61-76. Zhu, J., & Yu, S. (2025). Generative AI for tourism and hospitality education. Current Issues in Tourism , 1-9. Additional Declarations No competing interests reported. Cite Share Download PDF Status: Posted Version 1 posted You are reading this latest preprint version Research Square lets you share your work early, gain feedback from the community, and start making changes to your manuscript prior to peer review in a journal. As a division of Research Square Company, we’re committed to making research communication faster, fairer, and more useful. We do this by developing innovative software and high quality services for the global research community. Our growing team is made up of researchers and industry professionals working together to solve the most critical problems facing scientific publishing. Also discoverable on Platform About Our Team In Review Editorial Policies Advisory Board Help Center Resources Author Services Accessibility API Access RSS feed Manage Cookie Preferences © Research Square 2026 | ISSN 2693-5015 (online) Privacy Policy Terms of Service Do Not Sell My Personal Information {"props":{"pageProps":{"initialData":{"identity":"rs-8736638","acceptedTermsAndConditions":true,"allowDirectSubmit":true,"archivedVersions":[],"articleType":"Research Article","associatedPublications":[],"authors":[{"id":604382892,"identity":"e633217c-5a34-47ef-98e6-6529367d33db","order_by":0,"name":"Song Chen","email":"","orcid":"","institution":"Kyrgyz State University named after. I. Arabaev","correspondingAuthor":false,"prefix":"","firstName":"Song","middleName":"","lastName":"Chen","suffix":""},{"id":604382893,"identity":"1cc2e602-8cb3-4ee9-8e23-15211ab5d04f","order_by":1,"name":"Fengbo Wang","email":"","orcid":"","institution":"Kyrgyz State University named after. I. Arabaev","correspondingAuthor":false,"prefix":"","firstName":"Fengbo","middleName":"","lastName":"Wang","suffix":""},{"id":604382894,"identity":"45b429bc-3c10-4397-a687-9c0af85f54c7","order_by":2,"name":"Kalybek Zh Abdykadyrov","email":"","orcid":"","institution":"Kyrgyz State University named after. I. Arabaev","correspondingAuthor":false,"prefix":"","firstName":"Kalybek","middleName":"Zh","lastName":"Abdykadyrov","suffix":""},{"id":604382895,"identity":"4dbbe420-766e-4e61-b74d-c3c6ec9ed068","order_by":3,"name":"Junyu Long","email":"data:image/png;base64,iVBORw0KGgoAAAANSUhEUgAAAZAAAAAyAQMAAABI0h/eAAAABlBMVEX///8AAABVwtN+AAAACXBIWXMAAA7EAAAOxAGVKw4bAAAAt0lEQVRIiWNgGAWjYBACPhDxAcqRIEoLGxAzzgASPCRpYeYhTQt77+HXtjmH8+wZmA/e5mGwyyOshedcmnXutsPFPAxsydY8DMnFhLVI5JgZA7Uk9jDwmEnzMBxIbCBKiyVYC/83orUYP2aE2MJGpBaeM2aMvdvSi3kOsxlbzjFIJqyFn73H+MPPbdZ57O3ND2+8qbAjrAXsNgaG5gQGZhDbgAj1QMAMTC91CcSpHQWjYBSMghEJAKqKMzkGejMsAAAAAElFTkSuQmCC","orcid":"","institution":"Mianyang Polytechnic","correspondingAuthor":true,"prefix":"","firstName":"Junyu","middleName":"","lastName":"Long","suffix":""},{"id":604382896,"identity":"9b188f0e-6428-46e6-81f1-309912f4614d","order_by":4,"name":"Jiayi Xu","email":"","orcid":"","institution":"Southwest University","correspondingAuthor":false,"prefix":"","firstName":"Jiayi","middleName":"","lastName":"Xu","suffix":""},{"id":604382897,"identity":"922ff02c-3192-43be-80d8-74761a05bd27","order_by":5,"name":"Xiaojun Mai","email":"","orcid":"","institution":"Anhui Xinhua University","correspondingAuthor":false,"prefix":"","firstName":"Xiaojun","middleName":"","lastName":"Mai","suffix":""}],"badges":[],"createdAt":"2026-01-30 03:39:54","currentVersionCode":1,"declarations":"","doi":"10.21203/rs.3.rs-8736638/v1","doiUrl":"https://doi.org/10.21203/rs.3.rs-8736638/v1","draftVersion":[],"editorialEvents":[],"editorialNote":"","failedWorkflow":false,"files":[{"id":104488429,"identity":"533e0e07-3b07-42c1-8bc5-ac16c337deda","added_by":"auto","created_at":"2026-03-12 10:57:09","extension":"png","order_by":1,"title":"Figure 1","display":"","copyAsset":false,"role":"figure","size":56578,"visible":true,"origin":"","legend":"\u003cp\u003eResearch Framework.\u003c/p\u003e","description":"","filename":"floatimage1.png","url":"https://assets-eu.researchsquare.com/files/rs-8736638/v1/a588da5023f20d6af70f9b78.png"},{"id":104488485,"identity":"a8835c3c-d093-4e6d-9c66-4df70fb6bb19","added_by":"auto","created_at":"2026-03-12 10:57:26","extension":"pdf","order_by":0,"title":"","display":"","copyAsset":false,"role":"manuscript-pdf","size":1401813,"visible":true,"origin":"","legend":"","description":"","filename":"manuscript.pdf","url":"https://assets-eu.researchsquare.com/files/rs-8736638/v1/53f8df7e-d032-4e01-adf5-eccba7cc786a.pdf"}],"financialInterests":"No competing interests reported.","formattedTitle":"Deep Thinking Function in AI-Mediated Travel Planning: How Reasoning Transparency Shapes Information Adoption and Destination Acceptance","fulltext":[{"header":"1. Introduction","content":"\u003cp\u003eIn recent years, the global online travel market has expanded rapidly alongside the widespread adoption of mobile internet technologies and platform-based tourism services. Reports indicate that the global online travel economy reached approximately USD 5667.4\u0026nbsp;billion in 2024 (IMARC, \u003cspan citationid=\"CR24\" class=\"CitationRef\"\u003e2025\u003c/span\u003e). In China, although the tourism industry experienced unprecedented disruption during the COVID-19 pandemic, the sector has shown strong signs of recovery (Tang et al., \u003cspan citationid=\"CR56\" class=\"CitationRef\"\u003e2025\u003c/span\u003e). With mobile-based booking becoming the dominant mode of travel reservation, the penetration and convenience of mobile devices have emerged as key drivers of market growth. According to Fastdata\u0026rsquo;s \u003cspan citationid=\"CR15\" class=\"CitationRef\"\u003e2024\u003c/span\u003e industry report, domestic tourist trips reached RMB 5.615\u0026nbsp;billion (a year-on-year increase of 14.8%), and total tourism expenditure grew by 17.1% to RMB 5.75 trillion (Fastdata, \u003cspan citationid=\"CR15\" class=\"CitationRef\"\u003e2024\u003c/span\u003e). These figures suggest that both travel frequency and spending are approaching pre-pandemic levels, underscoring the increasing prominence of digitally mediated travel decision-making.\u003c/p\u003e \u003cp\u003eTourism is an information-intensive decision process in which travelers must evaluate complex and uncertain information relating to destination safety, attraction appeal, transportation accessibility, cultural experience, and potential constraints (Dai et al., \u003cspan citationid=\"CR9\" class=\"CitationRef\"\u003e2022\u003c/span\u003e; Fang, \u003cspan citationid=\"CR14\" class=\"CitationRef\"\u003e2023\u003c/span\u003e; Halkiopoulos et al., \u003cspan citationid=\"CR20\" class=\"CitationRef\"\u003e2022\u003c/span\u003e). Traditionally, such information has been accessed through interpersonal communication, online reviews, travel blogs, guidebooks, and official destination marketing sources (Hu \u0026amp; Yang, \u003cspan citationid=\"CR23\" class=\"CitationRef\"\u003e2021\u003c/span\u003e; Rehman Khan et al., \u003cspan citationid=\"CR41\" class=\"CitationRef\"\u003e2021\u003c/span\u003e; Zelenka et al., \u003cspan citationid=\"CR68\" class=\"CitationRef\"\u003e2021\u003c/span\u003e). The recent development of generative artificial intelligence, however, has introduced a new class of interactive decision-support agents into the tourism domain (Tuo et al., \u003cspan citationid=\"CR58\" class=\"CitationRef\"\u003e2025\u003c/span\u003e). Large language model (LLM) based systems no longer function merely as background algorithms or automated service responders; instead, they increasingly act as active \u0026ldquo;decision participants,\u0026rdquo; assisting travelers in selecting destinations, generating itineraries, comparing alternatives, and assessing risks (Doborjeh et al., \u003cspan citationid=\"CR11\" class=\"CitationRef\"\u003e2022\u003c/span\u003e; Doğan \u0026amp; Niyet, \u003cspan citationid=\"CR12\" class=\"CitationRef\"\u003e2024\u003c/span\u003e; Samala et al., \u003cspan citationid=\"CR44\" class=\"CitationRef\"\u003e2022\u003c/span\u003e). In particular, Dwivedi et al. (\u003cspan citationid=\"CR13\" class=\"CitationRef\"\u003e2024\u003c/span\u003e) pointed out that younger travelers are beginning to rely on AI tools during the early stages of travel planning, consulting them for personalized suggestions and itinerary recommendations. In the Chinese market, domestic platforms such as Doubao and DeepSeek have rapidly evolved into multimodal AI travel assistants capable of producing integrated text\u0026ndash;image content, map-based planning outputs, and highly structured travel guidance (Xu et al., \u003cspan citationid=\"CR65\" class=\"CitationRef\"\u003e2025\u003c/span\u003e; Zhu \u0026amp; Yu, \u003cspan citationid=\"CR72\" class=\"CitationRef\"\u003e2025\u003c/span\u003e). Compared with conventional user-generated content, these AI-generated recommendations are often perceived as more conversational, personalized, and cognitively accessible, thereby reshaping how travelers acquire and evaluate tourism information (Alharbi et al.).\u003c/p\u003e \u003cp\u003eAt the same time, the increasing reliance on AI-generated content (AIGC) has raised important debates regarding reliability, authenticity, and epistemic trust (Li et al., \u003cspan citationid=\"CR31\" class=\"CitationRef\"\u003e2025\u003c/span\u003e; Sable et al., \u003cspan citationid=\"CR43\" class=\"CitationRef\"\u003e2025\u003c/span\u003e; Tan et al., \u003cspan citationid=\"CR55\" class=\"CitationRef\"\u003e2025\u003c/span\u003e). LLMs may produce fabricated or inaccurate information (\u0026ldquo;hallucinations\u0026rdquo;), which poses particular risks in travel-related decision contexts (Jacob \u0026amp; Habibullah, \u003cspan citationid=\"CR26\" class=\"CitationRef\"\u003e2025\u003c/span\u003e; Maleki et al., \u003cspan citationid=\"CR35\" class=\"CitationRef\"\u003e2024\u003c/span\u003e; Ye et al., \u003cspan citationid=\"CR67\" class=\"CitationRef\"\u003e2023\u003c/span\u003e). Although AI outputs often appear coherent and professional, their underlying data provenance and processing logic remain opaque, making it difficult for users to assess credibility or bias (Waqas \u0026amp; Naseem, \u003cspan citationid=\"CR62\" class=\"CitationRef\"\u003e2025\u003c/span\u003e). Moreover, because AI systems do not possess lived experiential knowledge, their recommendations may lack the \u0026ldquo;situated authenticity\u0026rdquo; that travelers value when imagining or evaluating destination experiences (Ferhataj \u0026amp; Memaj, \u003cspan citationid=\"CR16\" class=\"CitationRef\"\u003e2024\u003c/span\u003e; Selgas-Cors, \u003cspan citationid=\"CR47\" class=\"CitationRef\"\u003e2025\u003c/span\u003e). Multimodal content may further amplify this tension by constructing hyper-realistic visualized travel scenarios that enhance persuasive appeal while obscuring uncertainty.\u003c/p\u003e \u003cp\u003eAgainst this backdrop, a critical question emerges: through what psychological and cognitive mechanisms do travelers evaluate AI-generated travel information, and how do these evaluations shape their intention to accept AI-recommended destinations? While prior tourism research has extensively examined online reviews and user-generated content (S\u0026aacute;nchez-Franco \u0026amp; Rey-Tienda, \u003cspan citationid=\"CR45\" class=\"CitationRef\"\u003e2024\u003c/span\u003e; Yamagishi et al., \u003cspan citationid=\"CR66\" class=\"CitationRef\"\u003e2024\u003c/span\u003e), relatively little is known about information adoption in contexts where content is algorithmically generated, conversational, and accompanied by dynamic system-generated reasoning traces.\u003c/p\u003e \u003cp\u003eA particularly salient feature in this emerging landscape is DeepSeek\u0026rsquo;s Deep Thinking Function, which discloses the AI\u0026rsquo;s reasoning path by explicitly presenting its interpretive process, analytical logic, and content-generation strategy (Liu et al., \u003cspan citationid=\"CR33\" class=\"CitationRef\"\u003e2025\u003c/span\u003e). This form of process transparency may not only alter travelers\u0026rsquo; perceptions of argument quality and source credibility, but may also reshape their judgment of information usefulness when engaging with AI-mediated travel planning.\u003c/p\u003e \u003cp\u003eTo address this research gap, the present study draws upon the Information Adoption Model (IAM) to develop a theoretical framework explaining travelers\u0026rsquo; responses to AI-generated travel recommendations in the context of DeepSeek\u0026rsquo;s Deep Thinking Function. We propose that the disclosure of AI reasoning enhances travelers\u0026rsquo; cognitive evaluations of argument quality and source credibility, which in turn strengthen perceived information usefulness and subsequently promote destination acceptance intention. By empirically examining these relationships, this study advances theoretical understanding of information persuasion in AI-mediated tourism environments and offers practical insights for the design of transparent AI platforms and destination marketing strategies.\u003c/p\u003e"},{"header":"2. Literature Review","content":"\u003cdiv id=\"Sec3\" class=\"Section2\"\u003e \u003ch2\u003e2.1 AI-generated Content and Information Processing in Tourism\u003c/h2\u003e \u003cp\u003eExisting research on digital tourism information has primarily focused on online reviews (Hu \u0026amp; Yang, \u003cspan citationid=\"CR23\" class=\"CitationRef\"\u003e2021\u003c/span\u003e; Sun et al., \u003cspan citationid=\"CR53\" class=\"CitationRef\"\u003e2024\u003c/span\u003e; Zelenka et al., \u003cspan citationid=\"CR68\" class=\"CitationRef\"\u003e2021\u003c/span\u003e; Zhang et al., \u003cspan citationid=\"CR70\" class=\"CitationRef\"\u003e2024\u003c/span\u003e), electronic word-of-mouth (eWOM) (Nguyen, \u003cspan citationid=\"CR37\" class=\"CitationRef\"\u003e2025\u003c/span\u003e; Song \u0026amp; Song, \u003cspan citationid=\"CR50\" class=\"CitationRef\"\u003e2025\u003c/span\u003e; Vu Dinh et al., \u003cspan citationid=\"CR59\" class=\"CitationRef\"\u003e2025\u003c/span\u003e), and user-generated content (UGC) (Liang et al., \u003cspan citationid=\"CR32\" class=\"CitationRef\"\u003e2024\u003c/span\u003e; Sujatmiko et al., \u003cspan citationid=\"CR52\" class=\"CitationRef\"\u003e2025\u003c/span\u003e; Wijaya et al., \u003cspan citationid=\"CR64\" class=\"CitationRef\"\u003e2025\u003c/span\u003e), highlighting their roles in shaping travelers\u0026rsquo; attitude formation, risk perception, and behavioral intention. Prior studies indicate that message credibility, review valence, and argument relevance significantly influence travelers\u0026rsquo; decision-making processes (Nguyen, \u003cspan citationid=\"CR37\" class=\"CitationRef\"\u003e2025\u003c/span\u003e). However, these studies implicitly assume that information originates from human senders, whose experience, authenticity, and emotional engagement constitute core evaluative criteria.\u003c/p\u003e \u003cp\u003eWith the emergence of AIGC, this assumption is increasingly challenged. Unlike UGC, Cao et al. (\u003cspan citationid=\"CR6\" class=\"CitationRef\"\u003e2025\u003c/span\u003e) argued that AIGC is algorithmically constructed rather than experientially lived, and thus lacks embodied travel experience. While recent studies acknowledge the potential of AI tools as information mediators in tourism planning, most existing works treat AI outputs as functionally equivalent to textual reviews or recommendation messages (Guo et al., \u003cspan citationid=\"CR18\" class=\"CitationRef\"\u003e2025\u003c/span\u003e; Zhang et al., \u003cspan citationid=\"CR69\" class=\"CitationRef\"\u003e2025\u003c/span\u003e). This analytical equivalence obscures the fact that AI systems introduce an epistemically distinct information ecology, in which authorship, knowledge provenance, and agency are fundamentally reconfigured.\u003c/p\u003e \u003cp\u003eMore importantly, prior research has largely examined AIGC from an outcome-oriented perspective, such as usefulness perceptions (Li et al., \u003cspan citationid=\"CR31\" class=\"CitationRef\"\u003e2025\u003c/span\u003e), trust formation (Zhou \u0026amp; Lu, \u003cspan citationid=\"CR71\" class=\"CitationRef\"\u003e2025\u003c/span\u003e), or satisfaction (Tan et al., \u003cspan citationid=\"CR55\" class=\"CitationRef\"\u003e2025\u003c/span\u003e), while overlooking the cognitive processes through which travelers interpret and validate AI-generated recommendations. As a result, we still know little about how travelers reconcile uncertainty, authenticity concerns, and credibility ambiguity when engaging with AI-mediated destination suggestions.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec4\" class=\"Section2\"\u003e \u003ch2\u003e2.2 Argument Quality and Source Credibility in the Information Adoption Model\u003c/h2\u003e \u003cp\u003eThe Information Adoption Model has been extensively used to explain how individuals evaluate information in online environments, emphasizing the effects of argument quality and source credibility on perceived usefulness and adoption intention (G\u0026ouml;kerik, \u003cspan citationid=\"CR17\" class=\"CitationRef\"\u003e2024\u003c/span\u003e; Horrich et al., \u003cspan citationid=\"CR22\" class=\"CitationRef\"\u003e2024\u003c/span\u003e; Long et al., \u003cspan citationid=\"CR34\" class=\"CitationRef\"\u003e2024\u003c/span\u003e). In tourism research, IAM has been widely applied to contexts such as social media reviews, travel blogs, and platform recommendations (Islam et al., \u003cspan citationid=\"CR25\" class=\"CitationRef\"\u003e2025\u003c/span\u003e; Kumar \u0026amp; Lata, \u003cspan citationid=\"CR28\" class=\"CitationRef\"\u003e2024\u003c/span\u003e; Shia et al.; Wei et al., \u003cspan citationid=\"CR63\" class=\"CitationRef\"\u003e2025\u003c/span\u003e). These studies consistently demonstrate that logical coherence, informativeness, and message relevance promote users\u0026rsquo; adoption of tourism information.\u003c/p\u003e \u003cp\u003eHowever, the application of IAM in existing studies is characterized by two major limitations. First, most IAM-based tourism studies conceptualize source credibility in terms of reviewer expertise, experience, or identity cues, again presupposing a human information sender (\u0026Ccedil;elik \u0026amp; Aslan, \u003cspan citationid=\"CR7\" class=\"CitationRef\"\u003e2025\u003c/span\u003e; G\u0026ouml;kerik, \u003cspan citationid=\"CR17\" class=\"CitationRef\"\u003e2024\u003c/span\u003e; Kumar \u0026amp; Lata, \u003cspan citationid=\"CR28\" class=\"CitationRef\"\u003e2024\u003c/span\u003e). When the source becomes an artificial intelligence agent with unknown training data and opaque reasoning processes, the meaning of credibility can no longer be reduced to traditional notions of expertise or trustworthiness. Thus, prior IAM research provides limited explanatory power in contexts where agency is computational rather than human.\u003c/p\u003e \u003cp\u003eSecond, argument quality has been measured primarily through surface-level linguistic indicators such as clarity, completeness, or informativeness (Lata \u0026amp; Rana, \u003cspan citationid=\"CR29\" class=\"CitationRef\"\u003e2025\u003c/span\u003e; Tseng \u0026amp; Wu, \u003cspan citationid=\"CR57\" class=\"CitationRef\"\u003e2024\u003c/span\u003e). While these dimensions capture textual persuasiveness, they fail to address how arguments are constructed, justified, and rationalized within AI-generated recommendations. In other words, IAM research has focused on what the message says, but rarely on how the message is produced and justified a conceptual omission that becomes critical in the era of generative AI.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec5\" class=\"Section2\"\u003e \u003ch2\u003e2.3 Transparency, Explainability, and AI Reasoning Disclosure\u003c/h2\u003e \u003cp\u003eA growing body of research in human\u0026ndash;AI interaction suggests that system transparency and reasoning disclosure may enhance user trust and perceived legitimacy by revealing the internal logic behind AI outputs. However, empirical findings remain inconsistent. Some studies indicate that transparency improves users\u0026rsquo; confidence and understanding (Radanliev, \u003cspan citationid=\"CR38\" class=\"CitationRef\"\u003e2025\u003c/span\u003e; F. Wang et al., \u003cspan citationid=\"CR61\" class=\"CitationRef\"\u003e2025\u003c/span\u003e), while others suggest that excessive disclosure may increase cognitive load or even expose uncertainty in AI reasoning (Aquilino et al., \u003cspan citationid=\"CR4\" class=\"CitationRef\"\u003e2024\u003c/span\u003e; Lee \u0026amp; Cha, \u003cspan citationid=\"CR30\" class=\"CitationRef\"\u003e2025\u003c/span\u003e; Ngo, \u003cspan citationid=\"CR36\" class=\"CitationRef\"\u003e2025\u003c/span\u003e), thereby weakening perceived credibility.\u003c/p\u003e \u003cp\u003eDespite these debates, very few tourism studies have examined process-level transparency in AI travel assistants. Existing tourism literature tends to conceptualize trust in AI tools as a static attribute, without considering how dynamic reasoning traces, such as DeepSeek\u0026rsquo;s Deep Thinking Function, actively reshape users\u0026rsquo; evaluative criteria. This omission is particularly salient because transparency in AI contexts functions not merely as an informational supplement, but as a meta-argument that frames how travelers interpret argument quality and credibility itself (Afroogh et al., \u003cspan citationid=\"CR1\" class=\"CitationRef\"\u003e2024\u003c/span\u003e).\u003c/p\u003e \u003cp\u003eIn this sense, Deep Thinking Function does not simply provide more information; rather, it discloses the epistemic structure behind AIGC, potentially transforming travelers\u0026rsquo; perceptions from judging message content to evaluating the validity of the reasoning process. However, to date, no empirical tourism study has systematically examined whether the disclosure of AI reasoning processes strengthens or weakens perceived argument quality, nor whether source credibility shifts from human-based trust toward system-based cognitive trust as a result of such transparency.\u003c/p\u003e \u003c/div\u003e"},{"header":"3. Hypotheses and Research Framework","content":"\u003cdiv id=\"Sec7\" class=\"Section2\"\u003e \u003ch2\u003e3.6. Deep Thinking Function as a Process-Level Epistemic Mechanism\u003c/h2\u003e \u003cp\u003eUnlike traditional online information environments, AI-generated travel recommendations are produced through computational reasoning processes that users are typically unable to observe. The Deep Thinking Function introduces a novel form of process-level transparency by revealing how the system interprets user instructions, organizes contextual constraints, and constructs its recommendation logic. Rather than simply adding more informational content, it operates as a meta-persuasive mechanism that shapes how travelers interpret and evaluate the cognitive validity of AI-generated outputs.\u003c/p\u003e \u003cp\u003eFrom a cognitive perspective, the disclosure of reasoning steps may strengthen users\u0026rsquo; perceptions that AI recommendations are grounded in explicit argumentation rather than opaque algorithmic inference. Such transparency, as noted by Aquilino et al. (\u003cspan citationid=\"CR4\" class=\"CitationRef\"\u003e2024\u003c/span\u003e), enables users to independently assess the generative pathway behind the content. When travelers are able to trace the logical trajectory underlying destination suggestions, for instance, trade-off comparisons, constraint filtering, or contextual matching, they may attribute greater coherence, rigor, and decision relevance to the final argument. Under this condition, where users actively participate in evaluating the content generation process, the skepticism toward the authenticity of AIGC outputs highlighted by prior scholars is likely to be reduced or even eliminated. In this sense, the way in which Deep Think enhances perceived argument quality does not lie in altering the information itself, but in strengthening users\u0026rsquo; confidence in the reasoning logic that underpins it.\u003c/p\u003e \u003cp\u003eAccordingly, we propose:\u003c/p\u003e \u003cp\u003eH1. Deep Thinking Function positively influences travelers\u0026rsquo; perception of argument quality in AI-generated travel information.\u003c/p\u003e \u003cp\u003eIn terms of source credibility, traditional conceptualizations of source credibility emphasize human-based attributes such as experience, identity cues, or reviewer expertise (G\u0026ouml;kerik, \u003cspan citationid=\"CR17\" class=\"CitationRef\"\u003e2024\u003c/span\u003e; Lata \u0026amp; Rana, \u003cspan citationid=\"CR29\" class=\"CitationRef\"\u003e2025\u003c/span\u003e; Liu et al., \u003cspan citationid=\"CR33\" class=\"CitationRef\"\u003e2025\u003c/span\u003e). In AI-mediated environments, however, Ngo (\u003cspan citationid=\"CR36\" class=\"CitationRef\"\u003e2025\u003c/span\u003e) pointed out that credibility shifts from social trust to epistemic trust. It means that the extent to which users have confidence in the system\u0026rsquo;s reasoning competence and decision logic becomes the primary determinant of credibility. Deep Thinking Function directly facilitates this shift by revealing how recommendations are generated, thereby reducing ambiguity surrounding the system\u0026rsquo;s knowledge provenance and analytical foundations.\u003c/p\u003e \u003cp\u003eProcess disclosure signals that the AI system is not arbitrarily producing destination suggestions, but is instead engaging in structured and explainable judgment. This visibility, as noted by Aquilino et al. (\u003cspan citationid=\"CR4\" class=\"CitationRef\"\u003e2024\u003c/span\u003e), may foster perceptions of competence, objectivity, and methodological soundness. Those attributes that collectively constitute a form of system-level credibility distinct from traditional human expertise. Conversely, when the reasoning process remains opaque, users may perceive AI recommendations as black-box outputs (Afroogh et al., \u003cspan citationid=\"CR1\" class=\"CitationRef\"\u003e2024\u003c/span\u003e), weakening credibility perceptions even when the message content appears coherent. Therefore, when users develop a high level of cognitive trust in the AI system, they are more likely to perceive the content it generates as credible。\u003c/p\u003e \u003cp\u003eTherefore, we hypothesize:\u003c/p\u003e \u003cp\u003eH2. Deep Thinking Function positively influences travelers\u0026rsquo; perception of source credibility in AI-generated travel information.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec8\" class=\"Section2\"\u003e \u003ch2\u003e3.2 Argument Quality and Perceived Information Usefulness\u003c/h2\u003e \u003cp\u003eWithin the IAM, according to Sussman and Siegal (\u003cspan citationid=\"CR54\" class=\"CitationRef\"\u003e2003\u003c/span\u003e), argument quality refers to the extent to which information is perceived as logical, accurate, relevant, and well-supported in relation to users\u0026rsquo; decision needs. Prior research in online tourism contexts has consistently shown that messages with coherent structure, rich informational content, and decision relevance are more likely to be perceived as useful during trip planning (\u0026Ccedil;elik \u0026amp; Aslan, \u003cspan citationid=\"CR7\" class=\"CitationRef\"\u003e2025\u003c/span\u003e; Islam et al., \u003cspan citationid=\"CR25\" class=\"CitationRef\"\u003e2025\u003c/span\u003e; Lata \u0026amp; Rana, \u003cspan citationid=\"CR29\" class=\"CitationRef\"\u003e2025\u003c/span\u003e). These findings, however, have largely been established in human-generated information environments, where argument quality is implicitly associated with experiential authenticity and narrative credibility.\u003c/p\u003e \u003cp\u003eIn AI-mediated contexts, argument quality acquires a different meaning. AI-generated recommendations are constructed through computational reasoning processes rather than lived travel experiences. As such, travelers may evaluate argument quality less in terms of emotional authenticity and more in terms of logical justification, internal coherence, and contextual adequacy. When AI-generated outputs provide detailed explanations, rationalized recommendations, and structured reasoning, travelers may perceive them as cognitively robust and practically applicable to their planning needs.\u003c/p\u003e \u003cp\u003eConversely, when arguments appear generic, superficial, or weakly justified, travelers may interpret AI outputs as template-like or detached from real travel scenarios, thereby reducing perceived usefulness. Building on this perspective, argument quality in the AI setting functions as a cognitive cue that signals whether the recommendation is capable of supporting informed travel decision-making.\u003c/p\u003e \u003cp\u003eAccordingly, we propose:\u003c/p\u003e \u003cp\u003eH3. Argument quality of AI-generated travel information positively influences travelers\u0026rsquo; perceived information usefulness.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec9\" class=\"Section2\"\u003e \u003ch2\u003e3.3 Source Credibility and Perceived Information Usefulness\u003c/h2\u003e \u003cp\u003eSource credibility in IAM traditionally refers to perceptions of expertise and trustworthiness associated with human reviewers or information contributors (Lata \u0026amp; Rana, \u003cspan citationid=\"CR29\" class=\"CitationRef\"\u003e2025\u003c/span\u003e; Sussman \u0026amp; Siegal, \u003cspan citationid=\"CR54\" class=\"CitationRef\"\u003e2003\u003c/span\u003e). In tourism studies, credibility has typically been inferred from reviewer identity cues, travel experience indicators, or platform reputation (Lata \u0026amp; Rana, \u003cspan citationid=\"CR29\" class=\"CitationRef\"\u003e2025\u003c/span\u003e; Shia et al.; Tseng \u0026amp; Wu, \u003cspan citationid=\"CR57\" class=\"CitationRef\"\u003e2024\u003c/span\u003e). Such conceptualizations presume that the source of information is a human agent with experiential authority.\u003c/p\u003e \u003cp\u003eIn AI-generated environments, this assumption becomes problematic. The source of content shifts from identifiable human contributors to computational systems whose training data, reasoning mechanisms, and epistemic grounding are often opaque. As a result, credibility judgments are less about the personal reliability of a reviewer and more about system competence, transparency, and epistemic legitimacy (Afroogh et al., \u003cspan citationid=\"CR1\" class=\"CitationRef\"\u003e2024\u003c/span\u003e; Schneier, \u003cspan citationid=\"CR46\" class=\"CitationRef\"\u003e2025\u003c/span\u003e).\u003c/p\u003e \u003cp\u003eTravelers may evaluate AI credibility based on whether the system demonstrates consistency, rational explanation, contextual sensitivity, and reliability across responses. When AI outputs appear well-reasoned, contextually aligned, and technically competent, travelers are more likely to perceive the system as a credible informational source, which subsequently enhances perceived usefulness. By contrast, perceived ambiguity, unexplained assertions, or generic recommendations may weaken credibility and reduce users\u0026rsquo; confidence in applying the information to real decisions.\u003c/p\u003e \u003cp\u003eThus, we hypothesize:\u003c/p\u003e \u003cp\u003eH4. Source credibility of AI-generated travel information positively influences travelers\u0026rsquo; perceived information usefulness.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec10\" class=\"Section2\"\u003e \u003ch2\u003e3.4 Perceived Information Usefulness and Information Adoption\u003c/h2\u003e \u003cp\u003ePerceived information usefulness, as highlighted by Horrich et al. (\u003cspan citationid=\"CR22\" class=\"CitationRef\"\u003e2024\u003c/span\u003e), represents a central cognitive mechanism through which online information influences behavioral intention in IAM. In tourism contexts, useful information not only facilitates cognitive evaluation of alternatives but also reduces uncertainty and decision effort, thereby increasing travelers\u0026rsquo; willingness to adopt the information (Lata \u0026amp; Rana, \u003cspan citationid=\"CR29\" class=\"CitationRef\"\u003e2025\u003c/span\u003e; Wei et al., \u003cspan citationid=\"CR63\" class=\"CitationRef\"\u003e2025\u003c/span\u003e).\u003c/p\u003e \u003cp\u003eIn AI-mediated travel planning, usefulness plays an even more pivotal role because travelers must decide whether to rely on algorithmically constructed recommendations rather than human-experienced advice. When AI-generated information is perceived as practically relevant, contextually appropriate, and decision-supportive, travelers are more likely to accept the recommended information as a viable option. Conversely, when AI suggestions are perceived as abstract, generic, or insufficiently grounded, travelers may disengage or revert to conventional human-based information sources.\u003c/p\u003e \u003cp\u003eTherefore, we propose:\u003c/p\u003e \u003cp\u003eH5. Perceived information usefulness positively influences information adoption.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec11\" class=\"Section2\"\u003e \u003ch2\u003e3.5 Information Adoption and Destination Acceptance Intention\u003c/h2\u003e \u003cp\u003eWithin AI-mediated travel planning environments, travelers\u0026rsquo; behavioral tendencies toward destinations are shaped not only by their cognitive judgments of information quality and credibility, but also by the extent to which they internalize and adopt the AI-generated recommendations. Prior IAM research has consistently demonstrated that once information is perceived as useful and epistemically trustworthy, individuals are more likely to integrate it into their decision-making processes, thereby translating cognitive evaluation into behavioral acceptance (Bao \u0026amp; Zhu, \u003cspan citationid=\"CR5\" class=\"CitationRef\"\u003e2025\u003c/span\u003e; \u0026Ccedil;elik \u0026amp; Aslan, \u003cspan citationid=\"CR7\" class=\"CitationRef\"\u003e2025\u003c/span\u003e; Lata \u0026amp; Rana, \u003cspan citationid=\"CR29\" class=\"CitationRef\"\u003e2025\u003c/span\u003e).\u003c/p\u003e \u003cp\u003eInformation adoption reflects travelers\u0026rsquo; willingness to rely on and incorporate AI-generated recommendations when forming travel expectations and planning choices. When travelers perceive that AI-generated content offers analytically sound reasoning, reliable justification, and context-appropriate guidance, they are more inclined to regard such recommendations as actionable decision inputs rather than merely informative references. This aligns with findings in technology-assisted tourism decision-making literature (Horrich et al., \u003cspan citationid=\"CR22\" class=\"CitationRef\"\u003e2024\u003c/span\u003e; Islam et al., \u003cspan citationid=\"CR25\" class=\"CitationRef\"\u003e2025\u003c/span\u003e), which indicate that the assimilation of mediated information can enhance perceived destination attractiveness, reduce cognitive risk, and strengthen commitment toward prospective travel choices.\u003c/p\u003e \u003cp\u003eAccordingly, it is reasonable to expect that higher levels of information adoption will positively influence travelers\u0026rsquo; intention to accept and consider the recommended destination. In other words, when travelers internalize AI-mediated recommendations as credible and useful, they are more likely to express willingness to visit, endorse, or further explore the suggested destination\u003c/p\u003e \u003cp\u003eThus, we hypothesize:\u003c/p\u003e \u003cp\u003eH6: Information adoption positively influences traveler\u0026rsquo;s destination acceptance intention.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec12\" class=\"Section2\"\u003e \u003ch2\u003e3.6 Research Framework\u003c/h2\u003e \u003cp\u003eBased on the above discussion, we constructed the following research framework (see Fig.\u0026nbsp;\u003cspan refid=\"Fig1\" class=\"InternalRef\"\u003e1\u003c/span\u003e). Drawing on the Information Adoption Model, this study conceptualizes argument quality and source credibility as the core antecedents shaping travelers\u0026rsquo; perceptions of information usefulness, which subsequently influence their destination acceptance intention. Within this persuasion pathway, the Deep Thinking Function is incorporated as a key mechanism variable that operates at the process-transparency level rather than at the message-content level.\u003c/p\u003e \u003cp\u003e \u003c/p\u003e \u003c/div\u003e"},{"header":"4.Methodology","content":"\u003cp\u003eThis study adopts a quantitative, cross-sectional research design to empirically test the proposed IAM-based conceptual model and the hypothesized relationships among Deep Thinking Function, argument quality, source credibility, perceived information usefulness, and traveler\u0026rsquo;s destination acceptance intention. A structured online questionnaire survey was employed as the primary data collection method, as it enables efficient access to digitally active respondents and aligns with the platform-mediated interaction context examined in this study.\u003c/p\u003e \u003cdiv id=\"Sec14\" class=\"Section2\"\u003e \u003ch2\u003e4.1 Instrument Design and Measurement\u003c/h2\u003e \u003cp\u003eAll constructs in this study were measured using multi-item scales adapted and adopted from validated instruments in prior literature, supplemented by self-developed items where necessary to capture the characteristics of the research context (see Table\u0026nbsp;\u003cspan refid=\"Tab1\" class=\"InternalRef\"\u003e1\u003c/span\u003e). The operational definitions of argument quality and source credibility follow prior studies grounded in the IAM. Argument quality focuses on the perceived clarity, logical consistency, and evidential support of the presented information, while source credibility reflects users\u0026rsquo; evaluations of the trustworthiness and perceived expertise of the information source.\u003c/p\u003e \u003cp\u003e \u003cdiv class=\"gridtable\"\u003e\u003ctable float=\"Yes\" id=\"Tab1\" border=\"1\"\u003e \u003ccaption language=\"En\"\u003e \u003cdiv class=\"CaptionNumber\"\u003eTable 1\u003c/div\u003e \u003cdiv class=\"CaptionContent\"\u003e \u003cp\u003eMeasurement Items\u003c/p\u003e \u003c/div\u003e \u003c/caption\u003e \u003ccolgroup cols=\"3\"\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c1\" colnum=\"1\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c2\" colnum=\"2\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c3\" colnum=\"3\"\u003e\u003c/div\u003e \u003cthead\u003e \u003ctr\u003e \u003cth align=\"left\" colname=\"c1\"\u003e \u003cp\u003eIndicators\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c2\"\u003e \u003cp\u003eItems\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c3\"\u003e \u003cp\u003eSources\u003c/p\u003e \u003c/th\u003e \u003c/tr\u003e \u003c/thead\u003e \u003ctbody\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\" morerows=\"3\" rowspan=\"4\"\u003e \u003cp\u003eDeep Thinking Function(DTF)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eDTF1: The AI travel assistant clearly explained the reasoning steps behind its destination recommendation\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003eHoang and Nguyen (\u003cspan citationid=\"CR21\" class=\"CitationRef\"\u003e2025\u003c/span\u003e)\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eDTF2: The Deep Thinking function helped me understand how the recommendation was generated\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\" morerows=\"2\" rowspan=\"3\"\u003e \u003cp\u003eSelf-developed\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eDTF3: The AI provided transparent justification for why certain options were selected or filtered out\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eDTF4: The reasoning process disclosed by the AI made the recommendation appear more logical and trustworthy.\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\" morerows=\"3\" rowspan=\"4\"\u003e \u003cp\u003eAugment Quality (AQ)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eAQ1: The destination recommendation provided by the AI is logically structured and well justified\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\" morerows=\"3\" rowspan=\"4\"\u003e \u003cp\u003eS. Wang et al. (\u003cspan citationid=\"CR61\" class=\"CitationRef\"\u003e2025\u003c/span\u003e)\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eAQ2: The information presented in the recommendation is detailed and well supported by relevant reasons\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eAQ3: The arguments used by the AI to support the recommendation are strong and convincing\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eAQ4: Overall, the content of the recommendation appears coherent and reasonable\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\" morerows=\"3\" rowspan=\"4\"\u003e \u003cp\u003eSource Credibility(SC)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eSC1: I believe the AI system is competent in analyzing travel-related information\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\" morerows=\"3\" rowspan=\"4\"\u003e \u003cp\u003eS. Wang et al. (\u003cspan citationid=\"CR61\" class=\"CitationRef\"\u003e2025\u003c/span\u003e)\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eSC2: The AI appears knowledgeable and reliable when generating destination recommendations\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eSC3: I trust the AI\u0026rsquo;s analytical process when evaluating travel options\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eSC4: Overall, I consider the AI to be a credible source of travel planning information\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\" morerows=\"3\" rowspan=\"4\"\u003e \u003cp\u003ePerceived Information Usefulness(PIU)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003ePIU1: The AI-generated recommendation is useful for my travel planning\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003eHoang and Nguyen (\u003cspan citationid=\"CR21\" class=\"CitationRef\"\u003e2025\u003c/span\u003e)\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003ePIU2: The information provided by the AI helps me make better travel decisions\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003eChiengkul et al. (\u003cspan citationid=\"CR8\" class=\"CitationRef\"\u003e2025\u003c/span\u003e)\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003ePIU3: The recommendation improves the efficiency of my travel planning process\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003eHoang and Nguyen (\u003cspan citationid=\"CR21\" class=\"CitationRef\"\u003e2025\u003c/span\u003e)\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003ePIU4:Overall, I find the AI recommendation beneficial for evaluating travel destinations\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003eChiengkul et al. (\u003cspan citationid=\"CR8\" class=\"CitationRef\"\u003e2025\u003c/span\u003e)\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\" morerows=\"3\" rowspan=\"4\"\u003e \u003cp\u003eInformation Adoption(IA)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eIA1:I am willing to rely on the AI-generated recommendation when planning my trip\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\" morerows=\"3\" rowspan=\"4\"\u003e \u003cp\u003eChiengkul et al. (\u003cspan citationid=\"CR8\" class=\"CitationRef\"\u003e2025\u003c/span\u003e)\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eIA2: I would take the AI\u0026rsquo;s suggestion into account when choosing a travel destination.\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eIA3: I consider the AI recommendation as an important reference for my decision-making\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eIA4: I am likely to follow the recommendation provided by the AI travel assistant\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\" morerows=\"3\" rowspan=\"4\"\u003e \u003cp\u003eDestination Acceptance Intention (DAI)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eDAI1: I am willing to consider visiting the recommended destination\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\" morerows=\"3\" rowspan=\"4\"\u003e \u003cp\u003eHoang and Nguyen (\u003cspan citationid=\"CR21\" class=\"CitationRef\"\u003e2025\u003c/span\u003e)\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eDAI2:I would be interested in learning more about the recommended destination\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eDAI3: I am likely to shortlist this destination as a potential travel option\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eDAI4: Overall, I have a positive intention to accept the recommended destination\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003c/tbody\u003e \u003c/colgroup\u003e \u003c/table\u003e\u003c/div\u003e \u003c/p\u003e \u003cp\u003eThe Deep Thinking Function construct was introduced as a newly added mechanism variable within the IAM framework, designed to assess the perceived impact of reasoning-process transparency on users\u0026rsquo; evaluations of argument quality and source credibility. Destination acceptance intention was measured as a behavioral intention construct capturing users\u0026rsquo; willingness to continue engaging with the AI-generated recommendation or to base future travel decisions on the presented information.\u003c/p\u003e \u003cp\u003eTo obtain respondents\u0026rsquo; genuine attitudes as accurately as possible, Russo et al. (\u003cspan citationid=\"CR42\" class=\"CitationRef\"\u003e2021\u003c/span\u003e) had suggested that the five-point Likert scale is an appropriate and effective response format for social science surveys. Accordingly, all measurement items in this study were assessed using a five-point Likert scale ranging from 1 (\u0026ldquo;strongly disagree\u0026rdquo;) to 5 (\u0026ldquo;strongly agree\u0026rdquo;). A small-scale pilot test was conducted prior to the main survey to ensure content validity, wording clarity, and consistency of interpretation across respondents. Minor revisions were subsequently made to improve item comprehensibility and reduce potential ambiguity, as well as to mitigate the risk of common method bias associated with self-reported data.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec15\" class=\"Section2\"\u003e \u003ch2\u003e4.2 Sampling Strategy and Data Collection\u003c/h2\u003e \u003cp\u003eThis study adopted a non-probability purposive sampling strategy with the aim of recruiting tourism users who had recently engaged in AI-mediated information environments similar to the focal context of this research. Such an approach is theoretically appropriate because, as Stratton (\u003cspan citationid=\"CR51\" class=\"CitationRef\"\u003e2023\u003c/span\u003e) argued, purposive sampling prioritizes participants who possess direct experiential relevance and cognitive involvement in the phenomenon under investigation, rather than treating the general population as a homogeneous group of information recipients.\u003c/p\u003e \u003cp\u003eData were collected between May and July 2025 through an online questionnaire administered via the Wenjuanxing survey platform. Participation was entirely voluntary and anonymous, and all respondents were informed of the academic purpose of the study. Screening questions were embedded in the survey to verify the relevance of respondents\u0026rsquo; prior experience and to exclude inattentive or ineligible participants.\u003c/p\u003e \u003cp\u003eFollowing a rigorous data-cleaning procedure, including attention-check verification, response-time filtering, and the removal of patterned or incomplete submissions, a total of 260 valid responses were retained for analysis. The Table\u0026nbsp;\u003cspan refid=\"Tab2\" class=\"InternalRef\"\u003e2\u003c/span\u003e shows details of the demographic profile.\u003c/p\u003e \u003cp\u003e \u003cdiv class=\"gridtable\"\u003e\u003ctable float=\"Yes\" id=\"Tab2\" border=\"1\"\u003e \u003ccaption language=\"En\"\u003e \u003cdiv class=\"CaptionNumber\"\u003eTable 2\u003c/div\u003e \u003cdiv class=\"CaptionContent\"\u003e \u003cp\u003eDemographic Profile\u003c/p\u003e \u003c/div\u003e \u003c/caption\u003e \u003ccolgroup cols=\"4\"\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c1\" colnum=\"1\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c2\" colnum=\"2\"\u003e\u003c/div\u003e \u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c3\" colnum=\"3\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c4\" colnum=\"4\"\u003e\u003c/div\u003e \u003cthead\u003e \u003ctr\u003e \u003cth align=\"left\" colname=\"c1\"\u003e \u003cp\u003eVariable\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c2\"\u003e \u003cp\u003eCategories\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c3\"\u003e \u003cp\u003eFrequency\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c4\"\u003e \u003cp\u003ePercentage\u003c/p\u003e \u003c/th\u003e \u003c/tr\u003e \u003c/thead\u003e \u003ctbody\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\" morerows=\"2\" rowspan=\"3\"\u003e \u003cp\u003eGender\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eMale\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e122\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e46.92\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eFemale\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e138\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e53.08\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eTotal\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e260\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e100\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\" morerows=\"3\" rowspan=\"4\"\u003e \u003cp\u003eAge\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e18\u0026ndash;28\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e80\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e30.77\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e29\u0026ndash;40\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e85\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e32.69\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eAbove 41\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e95\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e36.54\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eTotal\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e260\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e100\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\" morerows=\"4\" rowspan=\"5\"\u003e \u003cp\u003eEducation\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eSenior high school or below\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e81\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e31.15\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eBachelor\u0026rsquo;s or associate degree\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e76\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e29.23\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eMaster\u0026rsquo;s degree\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e62\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e23.85\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003ePhD\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e41\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e15.77\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eTotal\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e260\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e100\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\" morerows=\"4\" rowspan=\"5\"\u003e \u003cp\u003eFrequency of using AI tools\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eMultiple times a day\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e168\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e64.62\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eEveryday\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e67\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e25.77\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e3\u0026ndash;4 days per week\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e23\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e8.84\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eVery rare\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e2\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e0.77\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eTotal\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e260\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e100\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003c/tbody\u003e \u003c/colgroup\u003e \u003c/table\u003e\u003c/div\u003e \u003c/p\u003e \u003cp\u003eAlthough the reliance on online purposive sampling may limit the statistical generalizability of the findings, as noted by Rahman (\u003cspan citationid=\"CR39\" class=\"CitationRef\"\u003e2023\u003c/span\u003e), it strengthens the study\u0026rsquo;s contextual validity by ensuring that the sample reflects active users embedded within the AI-driven digital persuasion environment under examination. This trade-off is acknowledged as a methodological limitation, yet it also represents a theoretically meaningful design choice that remains consistent with prior empirical studies employing the IAM.\u003c/p\u003e \u003c/div\u003e"},{"header":"5. Results","content":"\u003cp\u003eFollowing the recommendations of Hair et al. (\u003cspan citationid=\"CR19\" class=\"CitationRef\"\u003e2019\u003c/span\u003e), this study employed partial least squares structural equation modeling (PLS-SEM) to validate the proposed research model using SmartPLS 4.0. The results are reported in two stages: (1) assessment of the measurement model and (2) evaluation of the structural model.\u003c/p\u003e \u003cdiv id=\"Sec17\" class=\"Section2\"\u003e \u003ch2\u003e5.1 Assessment of Measurement Model\u003c/h2\u003e \u003cdiv id=\"Sec18\" class=\"Section3\"\u003e \u003ch2\u003e5.1.1 Reliability and Validity\u003c/h2\u003e \u003cp\u003eTo ensure the reliability of the constructs, Cronbach\u0026rsquo;s alpha and composite reliability (CR) coefficients were examined. Jarupunphol et al. (\u003cspan citationid=\"CR27\" class=\"CitationRef\"\u003e2024\u003c/span\u003e) suggested that Cronbach\u0026rsquo;s alpha values exceeding 0.70 indicate satisfactory construct reliability. As shown in Table\u0026nbsp;\u003cspan refid=\"Tab3\" class=\"InternalRef\"\u003e3\u003c/span\u003e, the Cronbach\u0026rsquo;s alpha values for all constructs in this study are above the recommended threshold. The lowest value is observed for Source Credibility (SC) at 0.727, which remains well above the minimum criterion, confirming adequate reliability across all constructs.\u003c/p\u003e \u003cp\u003e \u003cdiv class=\"gridtable\"\u003e\u003ctable float=\"Yes\" id=\"Tab3\" border=\"1\"\u003e \u003ccaption language=\"En\"\u003e \u003cdiv class=\"CaptionNumber\"\u003eTable 3\u003c/div\u003e \u003cdiv class=\"CaptionContent\"\u003e \u003cp\u003eResults of Factor Loadings, Cronbach\u0026rsquo;s alpha, composite reliability and AVE\u003c/p\u003e \u003c/div\u003e \u003c/caption\u003e \u003ccolgroup cols=\"6\"\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c1\" colnum=\"1\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c2\" colnum=\"2\"\u003e\u003c/div\u003e \u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c3\" colnum=\"3\"\u003e\u003c/div\u003e \u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c4\" colnum=\"4\"\u003e\u003c/div\u003e \u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c5\" colnum=\"5\"\u003e\u003c/div\u003e \u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c6\" colnum=\"6\"\u003e\u003c/div\u003e \u003cthead\u003e \u003ctr\u003e \u003cth align=\"left\" colname=\"c1\"\u003e \u003cp\u003eConstructs\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c2\"\u003e \u003cp\u003eItems\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c3\"\u003e \u003cp\u003eFactor Loadings\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c4\"\u003e \u003cp\u003eCronbach's alpha\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c5\"\u003e \u003cp\u003eComposite reliability\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c6\"\u003e \u003cp\u003eAVE\u003c/p\u003e \u003c/th\u003e \u003c/tr\u003e \u003c/thead\u003e \u003ctbody\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eDeep Thinking Function (DTF)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eDTF1\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e0.811\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e0.820\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e0.881\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e0.650\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eDTF2\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e0.790\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eDTF3\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e0.830\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eDTF4\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e0.792\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eArgument Quality (AG)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eAG1\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e0.757\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e0.779\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e0.858\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e0.601\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eAG2\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e0.783\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eAG3\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e0.792\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eAG4\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e0.770\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eSource Credibility (SC)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eSC1\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e0.730\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e0.727\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e0.830\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e0.549\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eSC2\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e0.773\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eSC3\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e0.758\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eSC4\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e0.701\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003ePerceived Information Usefulness (PIU)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003ePIU1\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e0.809\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e0.771\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e0.853\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e0.593\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003ePIU2\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e0.774\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003ePIU3\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e0.706\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003ePIU4\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e0.786\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eInformation Adoption (IA)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eIA1\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e0.811\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e0.779\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e0.858\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e0.601\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eIA2\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e0.800\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eIA3\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e0.738\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eIA4\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e0.750\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eDestination Acceptance Intention (DAI)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eDAI1\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e0.829\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e0.789\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e0.863\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e0.613\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eDAI2\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e0.794\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eDAI3\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e0.781\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eDAI4\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e0.723\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003c/tbody\u003e \u003c/colgroup\u003e \u003c/table\u003e\u003c/div\u003e \u003c/p\u003e \u003cp\u003eComposite reliability coefficients were also calculated to further verify internal consistency. Consistent with Hair et al. (\u003cspan citationid=\"CR19\" class=\"CitationRef\"\u003e2019\u003c/span\u003e), a CR value of 0.70 or higher was considered acceptable. As reported in Table\u0026nbsp;\u003cspan refid=\"Tab3\" class=\"InternalRef\"\u003e3\u003c/span\u003e, all constructs exceed this threshold, indicating strong composite reliability and stable internal consistency.\u003c/p\u003e \u003cp\u003eConvergent validity was assessed using the Average Variance Extracted (AVE). According to Shrestha (\u003cspan citationid=\"CR49\" class=\"CitationRef\"\u003e2021\u003c/span\u003e), AVE values above 0.50 indicate that a construct explains more than half of the variance in its indicators. In this study, all constructs exhibit AVE values greater than 0.50 (see Table\u0026nbsp;\u003cspan refid=\"Tab1\" class=\"InternalRef\"\u003e1\u003c/span\u003e), demonstrating satisfactory convergent validity and confirming that the indicators load appropriately on their respective constructs.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec19\" class=\"Section3\"\u003e \u003ch2\u003e5.1.2 Discriminant Validity\u003c/h2\u003e \u003cp\u003eIn addition to reliability and convergent validity, discriminant validity was evaluated using both the Fornell\u0026ndash;Larcker criterion and the Heterotrait\u0026ndash;Monotrait (HTMT) ratio.\u003c/p\u003e \u003cp\u003eTable\u0026nbsp;\u003cspan refid=\"Tab4\" class=\"InternalRef\"\u003e4\u003c/span\u003e presents the Fornell\u0026ndash;Larcker results. A construct is considered to demonstrate discriminant validity when the square root of its AVE is greater than its correlations with other constructs (Afthanorhan et al., \u003cspan citationid=\"CR2\" class=\"CitationRef\"\u003e2021\u003c/span\u003e; Rasoolimanesh, \u003cspan citationid=\"CR40\" class=\"CitationRef\"\u003e2022\u003c/span\u003e). The results indicate that all constructs meet this criterion, suggesting the absence of discriminant validity concerns.\u003c/p\u003e \u003cp\u003e \u003cdiv class=\"gridtable\"\u003e\u003ctable float=\"Yes\" id=\"Tab4\" border=\"1\"\u003e \u003ccaption language=\"En\"\u003e \u003cdiv class=\"CaptionNumber\"\u003eTable 4\u003c/div\u003e \u003cdiv class=\"CaptionContent\"\u003e \u003cp\u003eThe results of Fornell\u0026ndash;Larcker criterion\u003c/p\u003e \u003c/div\u003e \u003c/caption\u003e \u003ccolgroup cols=\"7\"\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c1\" colnum=\"1\"\u003e\u003c/div\u003e \u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c2\" colnum=\"2\"\u003e\u003c/div\u003e \u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c3\" colnum=\"3\"\u003e\u003c/div\u003e \u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c4\" colnum=\"4\"\u003e\u003c/div\u003e \u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c5\" colnum=\"5\"\u003e\u003c/div\u003e \u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c6\" colnum=\"6\"\u003e\u003c/div\u003e \u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c7\" colnum=\"7\"\u003e\u003c/div\u003e \u003cthead\u003e \u003ctr\u003e \u003cth align=\"left\" colname=\"c1\"\u003e\u0026nbsp;\u003c/th\u003e \u003cth align=\"left\" colname=\"c2\"\u003e \u003cp\u003eAG\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c3\"\u003e \u003cp\u003eDAI\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c4\"\u003e \u003cp\u003eDTF\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c5\"\u003e \u003cp\u003eIA\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c6\"\u003e \u003cp\u003ePIU\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c7\"\u003e \u003cp\u003eSC\u003c/p\u003e \u003c/th\u003e \u003c/tr\u003e \u003c/thead\u003e \u003ctbody\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eAG\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e0.775\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c7\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eDAI\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e0.655\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e0.783\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c7\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eDTF\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e0.674\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e0.702\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e0.806\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c7\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eIA\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e0.558\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e0.626\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e0.649\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e0.775\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c7\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003ePIU\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e0.689\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e0.677\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e0.662\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e0.627\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e0.77\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c7\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eSC\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e0.629\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e0.651\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e0.671\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e0.566\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e0.675\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e \u003cp\u003e0.741\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003c/tbody\u003e \u003c/colgroup\u003e \u003c/table\u003e\u003c/div\u003e \u003c/p\u003e \u003cp\u003eTo further strengthen the evaluation, the HTMT ratio was also examined. As noted by Dirgiatmo (\u003cspan citationid=\"CR10\" class=\"CitationRef\"\u003e2023\u003c/span\u003e), HTMT is a more rigorous and contemporary approach to assessing discriminant validity. Following Hair et al. (\u003cspan citationid=\"CR19\" class=\"CitationRef\"\u003e2019\u003c/span\u003e), an HTMT value below 0.90 was adopted as the threshold. As shown in Table\u0026nbsp;\u003cspan refid=\"Tab5\" class=\"InternalRef\"\u003e5\u003c/span\u003e, all HTMT values fall below 0.90, thereby confirming that the constructs are empirically distinct from one another and that discriminant validity is satisfactorily established.\u003c/p\u003e \u003cp\u003e \u003cdiv class=\"gridtable\"\u003e\u003ctable float=\"Yes\" id=\"Tab5\" border=\"1\"\u003e \u003ccaption language=\"En\"\u003e \u003cdiv class=\"CaptionNumber\"\u003eTable 5\u003c/div\u003e \u003cdiv class=\"CaptionContent\"\u003e \u003cp\u003eThe results of HTMT\u003c/p\u003e \u003c/div\u003e \u003c/caption\u003e \u003ccolgroup cols=\"7\"\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c1\" colnum=\"1\"\u003e\u003c/div\u003e \u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c2\" colnum=\"2\"\u003e\u003c/div\u003e \u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c3\" colnum=\"3\"\u003e\u003c/div\u003e \u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c4\" colnum=\"4\"\u003e\u003c/div\u003e \u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c5\" colnum=\"5\"\u003e\u003c/div\u003e \u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c6\" colnum=\"6\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c7\" colnum=\"7\"\u003e\u003c/div\u003e \u003cthead\u003e \u003ctr\u003e \u003cth align=\"left\" colname=\"c1\"\u003e\u0026nbsp;\u003c/th\u003e \u003cth align=\"left\" colname=\"c2\"\u003e \u003cp\u003eAG\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c3\"\u003e \u003cp\u003eDAI\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c4\"\u003e \u003cp\u003eDTF\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c5\"\u003e \u003cp\u003eIA\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c6\"\u003e \u003cp\u003ePIU\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c7\"\u003e \u003cp\u003eSC\u003c/p\u003e \u003c/th\u003e \u003c/tr\u003e \u003c/thead\u003e \u003ctbody\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eAG\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c7\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eDAI\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e0.831\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c7\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eDTF\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e0.839\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e0.873\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c7\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eIA\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e0.714\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e0.792\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e0.811\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c7\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003ePIU\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e0.884\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e0.860\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e0.830\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e0.804\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c7\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eSC\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e0.827\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e0.850\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e0.863\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e0.749\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e0.893\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c7\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003c/tbody\u003e \u003c/colgroup\u003e \u003c/table\u003e\u003c/div\u003e \u003c/p\u003e \u003c/div\u003e \u003c/div\u003e \u003cdiv id=\"Sec20\" class=\"Section2\"\u003e \u003ch2\u003e5.2 Assessment of Structural Model\u003c/h2\u003e \u003cp\u003eAfter establishing the adequacy of the measurement model, the structural model was assessed to examine the hypothesized relationships among the constructs. Prior to hypothesis testing, collinearity diagnostics were conducted using the variance inflation factor (VIF). All VIF values were below the recommended threshold of 5.0, indicating the absence of multicollinearity and confirming that the structural estimates were not biased by redundancy among predictors (Hair et al., \u003cspan citationid=\"CR19\" class=\"CitationRef\"\u003e2019\u003c/span\u003e). The results are reported in Table\u0026nbsp;\u003cspan refid=\"Tab6\" class=\"InternalRef\"\u003e6\u003c/span\u003e.\u003c/p\u003e \u003cp\u003e \u003cdiv class=\"gridtable\"\u003e\u003ctable float=\"Yes\" id=\"Tab6\" border=\"1\"\u003e \u003ccaption language=\"En\"\u003e \u003cdiv class=\"CaptionNumber\"\u003eTable 6\u003c/div\u003e \u003cdiv class=\"CaptionContent\"\u003e \u003cp\u003eThe results of VIF\u003c/p\u003e \u003c/div\u003e \u003c/caption\u003e \u003ccolgroup cols=\"4\"\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c1\" colnum=\"1\"\u003e\u003c/div\u003e \u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c2\" colnum=\"2\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c3\" colnum=\"3\"\u003e\u003c/div\u003e \u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c4\" colnum=\"4\"\u003e\u003c/div\u003e \u003cthead\u003e \u003ctr\u003e \u003cth align=\"left\" colname=\"c1\"\u003e \u003cp\u003eIndicators\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c2\"\u003e \u003cp\u003eVIF\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c3\"\u003e \u003cp\u003eIndicators\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c4\"\u003e \u003cp\u003eVIF\u003c/p\u003e \u003c/th\u003e \u003c/tr\u003e \u003c/thead\u003e \u003ctbody\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eDTF1\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e1.778\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003ePIU1\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e1.592\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eDTF2\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e1.746\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003ePIU2\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e1.528\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eDTF3\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e1.920\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003ePIU3\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e1.358\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eDTF4\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e1.710\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003ePIU4\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e1.548\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eAG1\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e1.466\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003eIA1\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e1.627\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eAG2\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e1.490\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003eIA2\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e1.572\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eAG3\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e1.643\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003eIA3\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e1.448\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eAG4\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e1.500\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003eIA4\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e1.475\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eSC1\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e1.359\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003eDAI1\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e1.767\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eSC2\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e1.382\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003eDAI 2\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e1.580\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eSC3\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e1.424\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003eDAI 3\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e1.628\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eSC4\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e1.315\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003eDAI 4\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e1.377\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003c/tbody\u003e \u003c/colgroup\u003e \u003c/table\u003e\u003c/div\u003e \u003c/p\u003e \u003cp\u003eBootstrapping with 5,000 resamples was performed in SmartPLS 4.0 to obtain path coefficients, t-values, and significance levels for the hypothesized relationships. Table\u0026nbsp;\u003cspan refid=\"Tab8\" class=\"InternalRef\"\u003e7\u003c/span\u003e reports the estimated path coefficients and statistical significance.\u003c/p\u003e \u003cp\u003e \u003cdiv class=\"gridtable\"\u003e\u003ctable float=\"Yes\" id=\"Tab7\" border=\"1\"\u003e \u003ccaption language=\"En\"\u003e \u003cdiv class=\"CaptionNumber\"\u003eTable 7\u003c/div\u003e \u003cdiv class=\"CaptionContent\"\u003e \u003cp\u003eThe results of path coefficients\u003c/p\u003e \u003c/div\u003e \u003c/caption\u003e \u003ccolgroup cols=\"5\"\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c1\" colnum=\"1\"\u003e\u003c/div\u003e \u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c2\" colnum=\"2\"\u003e\u003c/div\u003e \u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c3\" colnum=\"3\"\u003e\u003c/div\u003e \u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c4\" colnum=\"4\"\u003e\u003c/div\u003e \u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c5\" colnum=\"5\"\u003e\u003c/div\u003e \u003cthead\u003e \u003ctr\u003e \u003cth align=\"left\" colname=\"c1\"\u003e\u0026nbsp;\u003c/th\u003e \u003cth align=\"left\" colname=\"c2\"\u003e \u003cp\u003ePath coefficients\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c3\"\u003e \u003cp\u003eStandard deviation (STDEV)\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c4\"\u003e \u003cp\u003eT statistics (|O/STDEV|)\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c5\"\u003e \u003cp\u003eP values\u003c/p\u003e \u003c/th\u003e \u003c/tr\u003e \u003c/thead\u003e \u003ctbody\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eDTF -\u0026gt; AG\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e0.674\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e0.034\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e19.788\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e0.000\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eDTF -\u0026gt; SC\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e0.671\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e0.035\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e19.264\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e0.000\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eAG -\u0026gt; PIU\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e0.437\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e0.07\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e6.222\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e0.000\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eSC -\u0026gt; PIU\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e0.4\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e0.072\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e5.569\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e0.000\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003ePIU -\u0026gt; IA\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e0.627\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e0.046\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e13.665\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e0.000\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eIA -\u0026gt; DAI\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e0.626\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e0.051\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e12.226\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e0.000\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003c/tbody\u003e \u003c/colgroup\u003e \u003c/table\u003e\u003c/div\u003e \u003c/p\u003e \u003cp\u003eThe results indicate that Deep Thinking Function exerts a significant positive effect on argument quality (β\u0026thinsp;=\u0026thinsp;0.674, t\u0026thinsp;=\u0026thinsp;19.788, p\u0026thinsp;=\u0026thinsp;0.000) and source credibility (β\u0026thinsp;=\u0026thinsp;0.671, t\u0026thinsp;=\u0026thinsp;19.264, p\u0026thinsp;=\u0026thinsp;0.000), supporting H1 and H2. This finding suggests that process-level transparency strengthens travelers\u0026rsquo; cognitive evaluations of AI-generated content by enhancing both its perceived logical rigor and the epistemic trustworthiness of the AI system.\u003c/p\u003e \u003cp\u003eArgument quality was found to have a significant positive influence on perceived information usefulness (β\u0026thinsp;=\u0026thinsp;0.437, t\u0026thinsp;=\u0026thinsp;6.222, p\u0026thinsp;=\u0026thinsp;0.000, supporting H3. Similarly, source credibility also positively affected perceived information usefulness (β\u0026thinsp;=\u0026thinsp;0.400, t\u0026thinsp;=\u0026thinsp;5.569, p\u0026thinsp;=\u0026thinsp;0.000), supporting H4. These results are consistent with prior IAM research, indicating that travelers assess AI-generated recommendations not only on message coherence, but also on confidence in the system\u0026rsquo;s reasoning competence.\u003c/p\u003e \u003cp\u003eFurthermore, the results show that perceived information usefulness has an significant effect on information adoption (β\u0026thinsp;=\u0026thinsp;0.627, t\u0026thinsp;=\u0026thinsp;13.665, p\u0026thinsp;=\u0026thinsp;0.000), supporting H5. Besides, information adoption demonstrated a significant positive effect on destination acceptance intention (β\u0026thinsp;=\u0026thinsp;0.626, t\u0026thinsp;=\u0026thinsp;12.226, p\u0026thinsp;=\u0026thinsp;0.000), supporting H6. This confirms that when AI-generated travel information is perceived as useful and decision-relevant, travelers are more likely to accept and consider the recommended destination.\u003c/p\u003e \u003cp\u003eModel explanatory power was assessed using R\u0026sup2; values. The R\u0026sup2; values for argument quality and source credibility indicate that a substantial proportion of variance is explained by Deep Thinking Function, while the R\u0026sup2; value for information usefulness reflects the combined explanatory effects of argument quality and source credibility. Additionally, the R\u0026sup2; value for destination acceptance intention demonstrates satisfactory predictive relevance of the model at the behavioral-intention level. Finally, the Q\u0026sup2; values for the endogenous constructs was over 0, hence, predictive relevance was established The results are reported in Table\u0026nbsp;\u003cspan refid=\"Tab8\" class=\"InternalRef\"\u003e7\u003c/span\u003e.\u003c/p\u003e \u003cp\u003e \u003cdiv class=\"gridtable\"\u003e\u003ctable float=\"Yes\" id=\"Tab8\" border=\"1\"\u003e \u003ccaption language=\"En\"\u003e \u003cdiv class=\"CaptionNumber\"\u003eTable 7\u003c/div\u003e \u003cdiv class=\"CaptionContent\"\u003e \u003cp\u003eThe results of R\u0026sup2; and Q\u0026sup2;\u003c/p\u003e \u003c/div\u003e \u003c/caption\u003e \u003ccolgroup cols=\"3\"\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c1\" colnum=\"1\"\u003e\u003c/div\u003e \u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c2\" colnum=\"2\"\u003e\u003c/div\u003e \u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c3\" colnum=\"3\"\u003e\u003c/div\u003e \u003cthead\u003e \u003ctr\u003e \u003cth align=\"left\" colname=\"c1\"\u003e\u0026nbsp;\u003c/th\u003e \u003cth align=\"left\" colname=\"c2\"\u003e \u003cp\u003eR-square\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c3\"\u003e \u003cp\u003eQ-square\u003c/p\u003e \u003c/th\u003e \u003c/tr\u003e \u003c/thead\u003e \u003ctbody\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eAG\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e0.454\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e0.445\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eDAT\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e0.391\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e0.259\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eIA\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e0.393\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e0.329\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003ePIU\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e0.571\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e0.422\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eSC\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e0.45\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e0.438\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003c/tbody\u003e \u003c/colgroup\u003e \u003c/table\u003e\u003c/div\u003e \u003c/p\u003e \u003c/div\u003e"},{"header":"6. Discussion","content":"\u003cp\u003eThis study investigates how DeepSeek\u0026rsquo;s Deep Thinking Function shapes travelers\u0026rsquo; psychological evaluation and adoption of AI-generated destination recommendations in AI-mediated travel planning contexts. The findings show that, through an extension of the IAM, transparency enabled by Deep Thinking Function significantly enhances travelers\u0026rsquo; perceptions of argument quality and source credibility, which subsequently strengthens perceived information usefulness and intention to accept the recommended destination. These results fill an important gap in existing research on the role of AI in tourism behavior, indicating that AI reasoning transparency is not merely an informational enhancement feature, but instead operates as a novel persuasive mechanism within human\u0026ndash;AI interaction.\u003c/p\u003e \u003cp\u003eFrist, at the overall structural level, the empirical results confirm the significant relationships among Deep Thinking Function, argument quality, source credibility, perceived information usefulness, information adoption, and destination acceptance intention, demonstrating strong explanatory power of the proposed pathway in the AI-mediated tourism context. On one hand, these findings reinforce the core assumption of the IAM that information adoption is jointly shaped by message attributes and individuals\u0026rsquo; cognitive judgments regarding the underlying process through which information is generated (Sussman \u0026amp; Siegal, \u003cspan citationid=\"CR54\" class=\"CitationRef\"\u003e2003\u003c/span\u003e). On the other hand, by conceptualizing Deep Thinking Function as a new AI-specific antecedent, this study refines the configuration and measurement of upstream cognitive variables and provides more concrete empirical evidence on the underlying mechanisms through which AI technologies influence tourism-related decision processes.\u003c/p\u003e \u003cp\u003eFurthermore, the results reveal that travelers\u0026rsquo; information adoption attitudes have evolved from a single rational-processing logic toward a dual cognitive\u0026ndash;reflective structure. Prior tourism research has shown that travel decision-making is often characterized by contextual uncertainty, experiential risk, and reliance on mediated representations (Dai et al., \u003cspan citationid=\"CR9\" class=\"CitationRef\"\u003e2022\u003c/span\u003e; Hu \u0026amp; Yang, \u003cspan citationid=\"CR23\" class=\"CitationRef\"\u003e2021\u003c/span\u003e). Compared with routine consumption contexts, tourism choices, as noted by Zelenka et al. (\u003cspan citationid=\"CR68\" class=\"CitationRef\"\u003e2021\u003c/span\u003e), therefore involve higher involvement and stronger cognitive prudence. Unlike prior IAM studies that predominantly focus on online reviews (Bao \u0026amp; Zhu, \u003cspan citationid=\"CR5\" class=\"CitationRef\"\u003e2025\u003c/span\u003e) or e-commerce platforms (Horrich et al., \u003cspan citationid=\"CR22\" class=\"CitationRef\"\u003e2024\u003c/span\u003e), this study demonstrates that the presence of AI does not diminish travelers\u0026rsquo; evaluative scrutiny. Instead, users strategically balance technological convenience against perceived credibility risks. In this sense, within contemporary AI-embedded tourism environments, AI functions not only as an information delivery instrument, but also as a cognitive agent that actively shapes interpretive frames, influencing how travelers understand, trust, and internalize tourism-related content.\u003c/p\u003e \u003cp\u003eFinally, this study identifies the cognitive role of Deep Thinking Function in AI-mediated tourism information environments. The results show that Deep Thinking Function significantly enhances perceived argument quality and source credibility, suggesting that deeper cognitive engagement improves users\u0026rsquo; evaluative precision in processing AI-generated recommendations. Prior studies have shown that applications such as intelligent itinerary planners, AI travel assistants, and automated recommendation systems produce information with characteristics such as personalized narratives, structured presentation, and implicit persuasive cues (Doğan \u0026amp; Niyet, \u003cspan citationid=\"CR12\" class=\"CitationRef\"\u003e2024\u003c/span\u003e; Dwivedi et al., \u003cspan citationid=\"CR13\" class=\"CitationRef\"\u003e2024\u003c/span\u003e; Ferhataj \u0026amp; Memaj, \u003cspan citationid=\"CR16\" class=\"CitationRef\"\u003e2024\u003c/span\u003e). Although these features improve decision efficiency, they may also foster excessive dependence on algorithmic suggestions and obscure source transparency, thereby weakening users\u0026rsquo; participatory reflection.\u003c/p\u003e \u003cp\u003eUnder such conditions, the appropriateness and reliability of AI-generated tourism recommendations may become uncertain. The present findings suggest that when users are able to cognitively engage with the reasoning process behind AI recommendations, they are better able to reflect on potential persuasive inducements, recognize boundaries between AI suggestions and embedded commercial intentions, and actively evaluate the authenticity and situational relevance of the generated content prior to adoption. This implies that Deep Thinking Function serves as a cognitive protection and corrective mechanism in AI-assisted tourism decision-making. In other words, technological convenience and cognitive vigilance must coexist. Although AI systems are designed to simplify decision processes, responsible tourism choices continue to rely on higher-order cognitive engagement and reflective judgment.\u003c/p\u003e \u003cdiv id=\"Sec22\" class=\"Section2\"\u003e \u003ch2\u003e6.1 Theoretical Implications\u003c/h2\u003e \u003cp\u003eThis study makes several important theoretical contributions to the intersection of AI cognition, persuasive communication, and technology-mediated tourism decision-making.\u003c/p\u003e \u003cp\u003eFirst, the study extends the IAM into AI-mediated travel contexts. Prior IAM research has predominantly examined information evaluation in human-generated environments such as online reviews, peer communication, and social platforms (Bao \u0026amp; Zhu, \u003cspan citationid=\"CR5\" class=\"CitationRef\"\u003e2025\u003c/span\u003e; \u0026Ccedil;elik \u0026amp; Aslan, \u003cspan citationid=\"CR7\" class=\"CitationRef\"\u003e2025\u003c/span\u003e; Horrich et al., \u003cspan citationid=\"CR22\" class=\"CitationRef\"\u003e2024\u003c/span\u003e). The present findings demonstrate that IAM mechanisms remain valid in AI-generated recommendation settings, but the evaluative focus of users shifts substantially. Rather than relying primarily on message clarity or human-based credibility cues, travelers begin to evaluate AI reasoning itself as an interpretive object. Through reasoning disclosure, Deep Thinking Function redirects users\u0026rsquo; cognitive orientation from outcome-focused evaluation toward process-based rationality, indicating that perceived argument quality increasingly derives from assessments of logical coherence, constraint filtering, and contextual fit within travel scenarios.\u003c/p\u003e \u003cp\u003e Second, this study reconceptualizes transparency disclosure as a mechanism variable, rather than a static technical attribute or ethical principle. Existing literature typically treats transparency as an interface feature or informational supplement. The findings of this research reveal that transparency actively reshapes users\u0026rsquo; judgment criteria, reinforcing both argument quality and source credibility by functioning as a meta-argument that frames interpretations of content validity. In this regard, reasoning transparency does not merely increase awareness but establishes epistemic legitimacy, thereby enriching theoretical discussions in human\u0026ndash;AI interaction and persuasive communication by evidencing how cognitive traceability can transform the meaning-construction process in AI-mediated environments.\u003c/p\u003e \u003cp\u003eThird, the results advance theoretical understanding of trust formation in AI-mediated tourism decision contexts. Traditional tourism trust research has emphasized identity-based credibility, such as reviewer expertise, experiential similarity, or social proximity (Bao \u0026amp; Zhu, \u003cspan citationid=\"CR5\" class=\"CitationRef\"\u003e2025\u003c/span\u003e; Long et al., \u003cspan citationid=\"CR34\" class=\"CitationRef\"\u003e2024\u003c/span\u003e). In contrast, this study identifies a shift toward system-centric credibility, wherein user confidence is grounded in perceptions of analytical competence, inferential rigor, and decision-logic reliability. This suggests that trust in AI is not simply transplanted from interpersonal trust models, but instead reflects a qualitatively distinct form of cognitive trust in reasoning capacity, a form of trust particularly salient in high-involvement and uncertainty-laden contexts such as travel planning.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec23\" class=\"Section2\"\u003e \u003ch2\u003e6.2 Practical Implications\u003c/h2\u003e \u003cp\u003eThis study also yields several important practical implications for AI platform designers, tourism marketers, and destination management organizations.\u003c/p\u003e \u003cp\u003eFirst, AI travel assistants should shift from output-only recommendation modes toward process-aware and reasoning-traceable presentation formats. The findings indicate that exposing reasoning logic can significantly enhance perceived argument quality, usefulness, and adoption confidence. Rather than merely delivering instant itineraries, AI systems may benefit from revealing constraint-filtering steps, trade-off comparisons among alternative destinations, and contextual justification based on traveler preferences or situational demands. Such design allows users to audit, interpret, and participate in the recommendation process, thereby strengthening engagement, transparency perception, and decision assurance in complex travel scenarios.\u003c/p\u003e \u003cp\u003eSecond, tourism platforms should treat transparency as a persuasion strategy rather than a compliance-oriented disclosure feature. Deep Think-type reasoning displays function as a cognitive scaffold that legitimizes AI-generated recommendations by signaling analytical rigor, internal coherence, and methodological soundness. In destination promotion contexts, this suggests that argument quality can be enhanced without intensifying promotional rhetoric, and credibility can be strengthened without relying on influencer identity cues or social endorsements. Accordingly, AI-based persuasion is likely to complement rather than replace human-generated travel narratives, forming a hybrid communication ecology in which algorithmic reasoning and experiential storytelling jointly shape travelers\u0026rsquo; perception and trust.\u003c/p\u003e \u003cp\u003eThird, the findings provide strategic insights for tourism marketing in the post-AIGC environment. As travelers increasingly consult AI systems during the early stages of trip planning, destinations are no longer competing only within social review ecosystems, but also within AI reasoning pipelines and decision-logic structures. This implies that content formats aligned with AI cognitive processing, such as clearly structured attribute descriptions, contextual relevance markers, and scenario-specific application cues, are more likely to be framed as high-quality recommendations. Consequently, destination branding strategies may need to evolve from narrative persuasion toward decision-logic compatibility, ensuring that tourism information not only attracts human audiences but is also interpretable, retrievable, and argumentatively meaningful within AI-mediated planning environments.\u003c/p\u003e \u003c/div\u003e"},{"header":"7. Limitations and Future Research","content":"\u003cp\u003eAlthough this study offers meaningful empirical insights, several limitations should be acknowledged, which also provide avenues for future research. First, the research adopted a cross-sectional survey design, meaning that all data were collected at a single point in time. While this approach is appropriate for examining current perceptions and behavioral intentions, it limits the ability to infer causal relationships or capture dynamic changes in attitudes and behaviors. Future studies may employ longitudinal or experimental designs to better trace temporal variations and strengthen causal inferences.\u003c/p\u003e \u003cp\u003eSecond, the measurement of key constructs relied on self-reported data, which, despite the pilot testing and procedural remedies adopted to reduce response bias, may still be subject to social desirability effects or common method variance. Subsequent research could incorporate multi-source or behavioral data, such as platform usage logs, objective performance indicators, or peer evaluations, to enhance measurement robustness and triangulate findings.\u003c/p\u003e \u003cp\u003eThird, the sample was drawn from a specific population and contextual setting, which may constrain the generalizability of the results. Although stratified sampling was used to ensure representation, cultural, institutional, or regional differences may influence respondents\u0026rsquo; perceptions and decision-making processes. Future research is encouraged to replicate the study across different demographic groups, geographical contexts, or cultural environments, and to conduct cross-cultural comparative analyses to test the external validity of the model.\u003c/p\u003e \u003cp\u003eFinally, this study concentrated on quantitative analysis, which, although valuable for hypothesis testing, may not fully capture the nuanced cognitive and emotional mechanisms underlying respondents\u0026rsquo; attitudes and behavioral responses. Subsequent research could adopt mixed-methods or qualitative approaches, such as interviews, focus groups, or text mining of user-generated content, to provide deeper interpretive insights and complement the quantitative findings.\u003c/p\u003e \u003cp\u003eTaken together, addressing these limitations in future research will not only enhance methodological rigor and contextual applicability, but also contribute to the ongoing refinement and extension of the theoretical framework examined in this study.\u003c/p\u003e"},{"header":"Abbreviations","content":"\u003cdiv class=\"DefinitionList\"\u003e \u003cdiv class=\"DefinitionListEntry\"\u003e \u003cdiv class=\"Term\"\u003eIAM\u003c/div\u003e \u003cdiv class=\"Description\"\u003e \u003cp\u003eInformation Adoption Model\u003c/p\u003e \u003c/div\u003e \u003c/div\u003e \u003cdiv class=\"DefinitionListEntry\"\u003e \u003cdiv class=\"Term\"\u003eLLM\u003c/div\u003e \u003cdiv class=\"Description\"\u003e \u003cp\u003eLarge Language Model\u003c/p\u003e \u003c/div\u003e \u003c/div\u003e \u003cdiv class=\"DefinitionListEntry\"\u003e \u003cdiv class=\"Term\"\u003eAIGC\u003c/div\u003e \u003cdiv class=\"Description\"\u003e \u003cp\u003eAI-generated content\u003c/p\u003e \u003c/div\u003e \u003c/div\u003e \u003cdiv class=\"DefinitionListEntry\"\u003e \u003cdiv class=\"Term\"\u003eUGC\u003c/div\u003e \u003cdiv class=\"Description\"\u003e \u003cp\u003eUser-generated content\u003c/p\u003e \u003c/div\u003e \u003c/div\u003e \u003cdiv class=\"DefinitionListEntry\"\u003e \u003cdiv class=\"Term\"\u003eDTF\u003c/div\u003e \u003cdiv class=\"Description\"\u003e \u003cp\u003eDeep Thinking Function\u003c/p\u003e \u003c/div\u003e \u003c/div\u003e \u003cdiv class=\"DefinitionListEntry\"\u003e \u003cdiv class=\"Term\"\u003eAQ\u003c/div\u003e \u003cdiv class=\"Description\"\u003e \u003cp\u003eAugment Quality\u003c/p\u003e \u003c/div\u003e \u003c/div\u003e \u003cdiv class=\"DefinitionListEntry\"\u003e \u003cdiv class=\"Term\"\u003eSC\u003c/div\u003e \u003cdiv class=\"Description\"\u003e \u003cp\u003eSource Credibility\u003c/p\u003e \u003c/div\u003e \u003c/div\u003e \u003cdiv class=\"DefinitionListEntry\"\u003e \u003cdiv class=\"Term\"\u003ePIU\u003c/div\u003e \u003cdiv class=\"Description\"\u003e \u003cp\u003ePerceived Information Usefulness\u003c/p\u003e \u003c/div\u003e \u003c/div\u003e \u003cdiv class=\"DefinitionListEntry\"\u003e \u003cdiv class=\"Term\"\u003eIA\u003c/div\u003e \u003cdiv class=\"Description\"\u003e \u003cp\u003eInformation Adoption\u003c/p\u003e \u003c/div\u003e \u003c/div\u003e \u003cdiv class=\"DefinitionListEntry\"\u003e \u003cdiv class=\"Term\"\u003eDAI\u003c/div\u003e \u003cdiv class=\"Description\"\u003e \u003cp\u003eDestination Acceptance Intention\u003c/p\u003e \u003c/div\u003e \u003c/div\u003e \u003cdiv class=\"DefinitionListEntry\"\u003e \u003cdiv class=\"Term\"\u003eVIF\u003c/div\u003e \u003cdiv class=\"Description\"\u003e \u003cp\u003eVariance inflation factor\u003c/p\u003e \u003c/div\u003e \u003c/div\u003e \u003cdiv class=\"DefinitionListEntry\"\u003e \u003cdiv class=\"Term\"\u003eAVE\u003c/div\u003e \u003cdiv class=\"Description\"\u003e \u003cp\u003eAverage variance extracted\u003c/p\u003e \u003c/div\u003e \u003c/div\u003e \u003c/div\u003e"},{"header":"Declarations","content":"\u003cp\u003e\u003cstrong\u003eEthical approval and Consent to Participate\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eThe questionnaire was assessed and approved by Mianyang Polytechnic. An informed consent form signed by Mianyang Polytechnic was provided to every participant before the survey. All data remains confidential.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eConsent for Publication\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eNot applicable\u003c/p\u003e\n\u003ch2\u003eFunding\u003c/h2\u003e\n\u003cp\u003eThis work was supported by Open Project Fund of Key Laboratory of Digital Innovation of Tianfu Culture, Sichuan Provincial Department of Culture and Tourism (Project No. TFWH-2025-40) and Hainan provincial Natural Science Foundation of China (Project No. 623RC510).\u003c/p\u003e\n\u003ch2\u003eAuthor Contribution\u003c/h2\u003e\n\u003cp\u003eConceptualization, S.C. and JY.L.; methodology, FB.W; software, JY.X. and XJ.M; validation, K.Z.A; data curation, K.Z.A.; writing\u0026mdash;original draft preparation, S.C. and JY.L; writing\u0026mdash;review and editing, FB.W. and JY.L.; visualization, K.Z.A. All authors have read and agreed to the published version of the manuscript.\u003c/p\u003e\n\u003ch2\u003eData Availability\u003c/h2\u003e\n\u003cp\u003eThe datasets used and analysed during the current study are available from the corresponding author on reasonable request\u003c/p\u003e"},{"header":"References","content":"\u003col\u003e\n \u003cli\u003eAfroogh, S., Akbari, A., Malone, E., Kargar, M., \u0026amp; Alambeigi, H. (2024). Trust in AI: progress, challenges, and future directions. \u003cem\u003eHumanities and Social Sciences Communications\u003c/em\u003e,\u003cem\u003e\u0026nbsp;11\u003c/em\u003e(1), 1-30.\u003c/li\u003e\n \u003cli\u003eAfthanorhan, A., Ghazali, P. L., \u0026amp; Rashid, N. (2021). Discriminant validity: A comparison of CBSEM and consistent PLS using Fornell \u0026amp; Larcker and HTMT approaches. Journal of Physics: Conference Series,\u003c/li\u003e\n \u003cli\u003eAlharbi, D., Alharbi, R., Alafif, T., Jassas, M., Alfattni, G., Al-Luhaybi, M., Alotaibi, H., Gharawi, A., Zia, S., \u0026amp; Kazalah, F. DOSTE: Domain Specific LLMs for Saudi Tourism and Entertainment.\u003c/li\u003e\n \u003cli\u003eAquilino, L., Bisconti, P., \u0026amp; Marchetti, A. (2024). Trust in AI: Transparency, and uncertainty reduction. Development of a new theoretical framework. CEUR workshop proceedings,\u003c/li\u003e\n \u003cli\u003eBao, Z., \u0026amp; Zhu, Y. (2025). Understanding online reviews adoption in social network communities: an extension of the information adoption model. \u003cem\u003eInformation Technology \u0026amp; People\u003c/em\u003e,\u003cem\u003e\u0026nbsp;38\u003c/em\u003e(1), 48-69.\u003c/li\u003e\n \u003cli\u003eCao, Y., Li, S., Liu, Y., Yan, Z., Dai, Y., Yu, P., \u0026amp; Sun, L. (2025). A survey of ai-generated content (aigc). \u003cem\u003eACM Computing Surveys\u003c/em\u003e,\u003cem\u003e\u0026nbsp;57\u003c/em\u003e(5), 1-38.\u003c/li\u003e\n \u003cli\u003e\u0026Ccedil;elik, K., \u0026amp; Aslan, A. (2025). The Impact of Electronic Word of Mouth (eWOM) on Visit Intention within the Framework of the Information Adoption Model: A Study on Instagram Users. \u003cem\u003eInternational Journal of Marketing, Communication and New Media\u003c/em\u003e,\u003cem\u003e\u0026nbsp;12\u003c/em\u003e(23).\u003c/li\u003e\n \u003cli\u003eChiengkul, W., Kumjorn, P., Tantipanichkul, T., \u0026amp; Suphan, K. (2025). Engaging with AI in tourism: a key to enhancing smart experiences and emotional bonds. \u003cem\u003eAsia-Pacific Journal of Business Administration\u003c/em\u003e.\u003c/li\u003e\n \u003cli\u003eDai, F., Wang, D., \u0026amp; Kirillova, K. (2022). Travel inspiration in tourist decision making. \u003cem\u003eTourism Management\u003c/em\u003e,\u003cem\u003e\u0026nbsp;90\u003c/em\u003e, 104484.\u003c/li\u003e\n \u003cli\u003eDirgiatmo, Y. (2023). Testing the discriminant validity and heterotrait\u0026ndash;monotrait ratio of correlation (HTMT): A case in Indonesian SMEs. In \u003cem\u003eMacroeconomic Risk and Growth in the Southeast Asian Countries: Insight from Indonesia\u003c/em\u003e (pp. 157-170). Emerald Publishing Limited.\u003c/li\u003e\n \u003cli\u003eDoborjeh, Z., Hemmington, N., Doborjeh, M., \u0026amp; Kasabov, N. (2022). Artificial intelligence: a systematic review of methods and applications in hospitality and tourism. \u003cem\u003eInternational journal of contemporary hospitality management\u003c/em\u003e,\u003cem\u003e\u0026nbsp;34\u003c/em\u003e(3), 1154-1176.\u003c/li\u003e\n \u003cli\u003eDoğan, S., \u0026amp; Niyet, İ. Z. (2024). Artificial intelligence (AI) in tourism. In \u003cem\u003eFuture Tourism Trends Volume 2: Technology Advancement, Trends and Innovations for the Future in Tourism\u003c/em\u003e (pp. 3-21). Emerald Publishing Limited.\u003c/li\u003e\n \u003cli\u003eDwivedi, Y. K., Pandey, N., Currie, W., \u0026amp; Micu, A. (2024). Leveraging ChatGPT and other generative artificial intelligence (AI)-based applications in the hospitality and tourism industry: practices, challenges and research agenda. \u003cem\u003eInternational journal of contemporary hospitality management\u003c/em\u003e,\u003cem\u003e\u0026nbsp;36\u003c/em\u003e(1), 1-12.\u003c/li\u003e\n \u003cli\u003eFang, R. (2023). Proposing a cyclic model of tourist decision making: A review and integration of behavioral and choice-set models. \u003cem\u003eJournal of Hospitality \u0026amp; Tourism Research\u003c/em\u003e,\u003cem\u003e\u0026nbsp;47\u003c/em\u003e(7), 1161-1186.\u003c/li\u003e\n \u003cli\u003eFastdata. (2024). \u003cem\u003e中国旅游行业年度报告\u003c/em\u003e. https://www.ifastdata.com/wp-content/uploads/2025/04/Fastdata%E6%9E%81%E6%95%B0%EF%BC%9A%E4%B8%AD%E5%9B%BD%\u003cbr\u003eE6%97%85%E6%B8%B8%E8%A1%8C%E4%B8%9A%E5%B9%B4%E5%BA%A6%E6%8A%A5%E5%91%8A2024.pdf\u003c/li\u003e\n \u003cli\u003eFerhataj, A., \u0026amp; Memaj, F. (2024). Challanges And Opportunities Of Ai Implementation In Tourism: An Ethical And Technological Perspective. \u003cem\u003eSTUDIJOS\u0026ndash;VERSLAS\u0026ndash;VISUOMENĖ: DABARTIS IR ATEITIES ĮŽVALGOS\u003c/em\u003e,\u003cem\u003e\u0026nbsp;1\u003c/em\u003e(IX), 217-231.\u003c/li\u003e\n \u003cli\u003eG\u0026ouml;kerik, M. (2024). The enchantment of social media influencers: Analysing consumer attitudes through the lens of the information adoption model. \u003cem\u003eOPUS Journal of Society Research\u003c/em\u003e,\u003cem\u003e\u0026nbsp;21\u003c/em\u003e(3), 125-139.\u003c/li\u003e\n \u003cli\u003eGuo, Y., Yu, F., Lai, J., \u0026amp; Yuan, X. (2025). How is AIGC shaping the world: an analysis of bibliometrics. \u003cem\u003eInformation Research an international electronic journal\u003c/em\u003e,\u003cem\u003e\u0026nbsp;30\u003c/em\u003e(iConf), 679-689.\u003c/li\u003e\n \u003cli\u003eHair, J. F., Risher, J. J., Sarstedt, M., \u0026amp; Ringle, C. M. (2019). When to use and how to report the results of PLS-SEM. \u003cem\u003eEuropean business review\u003c/em\u003e,\u003cem\u003e\u0026nbsp;31\u003c/em\u003e(1), 2-24.\u003c/li\u003e\n \u003cli\u003eHalkiopoulos, C., Antonopoulou, H., Gkintoni, E., \u0026amp; Aroutzidis, A. (2022). Neuromarketing as an indicator of cognitive consumer behavior in decision-making process of tourism destination\u0026mdash;An overview. Transcending Borders in Tourism Through Innovation and Cultural Heritage: 8th International Conference, IACuDiT, Hydra, Greece, 2021,\u003c/li\u003e\n \u003cli\u003eHoang, D. S., \u0026amp; Nguyen, D. T. A. (2025). The role of AI assistants in promoting sustainable Halal tourism in non-Muslim destinations. \u003cem\u003eDiscover Sustainability\u003c/em\u003e,\u003cem\u003e\u0026nbsp;6\u003c/em\u003e(1), 705.\u003c/li\u003e\n \u003cli\u003eHorrich, A., Ertz, M., \u0026amp; Bekir, I. (2024). The effect of information adoption via social media on sustainable consumption intentions: The moderating influence of gender. \u003cem\u003eCurrent Psychology\u003c/em\u003e,\u003cem\u003e\u0026nbsp;43\u003c/em\u003e(18), 16349-16362.\u003c/li\u003e\n \u003cli\u003eHu, X., \u0026amp; Yang, Y. (2021). What makes online reviews helpful in tourism and hospitality? A bare-bones meta-analysis. \u003cem\u003eJournal of Hospitality Marketing \u0026amp; Management\u003c/em\u003e,\u003cem\u003e\u0026nbsp;30\u003c/em\u003e(2), 139-158.\u003c/li\u003e\n \u003cli\u003eIMARC. (2025). \u003cem\u003eOnline Travel Market Size, Share, Trends and Forecast by Service Type, Platform, Mode of Booking, Age Group, and Region, 2026-2034\u003c/em\u003e. transforming ideas into impact. Retrieved 26 December from https://www.imarcgroup.com/online-travel-market\u003c/li\u003e\n \u003cli\u003eIslam, M. T., Herjanto, H., Kumar, J., \u0026amp; Amin, M. (2025). Online Travel Reviews and Tourist Destination Choices: An Extension of the Information Adoption Model. \u003cem\u003eTourism Review International\u003c/em\u003e,\u003cem\u003e\u0026nbsp;29\u003c/em\u003e(1), 17-32.\u003c/li\u003e\n \u003cli\u003eJacob, S. L., \u0026amp; Habibullah, P. S. (2025). A Systematic Analysis and Review on Intrusion Detection Systems Using Machine Learning and Deep Learning Algorithms. \u003cem\u003eJournal of Computational and Cognitive Engineering\u003c/em\u003e,\u003cem\u003e\u0026nbsp;4\u003c/em\u003e(2), 108-120.\u003c/li\u003e\n \u003cli\u003eJarupunphol, P., Ikonnikov, O., Roncevic, I., Kapustina, S., Kataeva, A., Parfjonovs, M., \u0026amp; Tsarev, R. (2024). Applying Cronbach\u0026rsquo;s alpha to ensure reliable online testing in e-learning environments. In \u003cem\u003eProceedings of the Computational Methods in Systems and Software\u003c/em\u003e (pp. 120-139). Springer.\u003c/li\u003e\n \u003cli\u003eKumar, A., \u0026amp; Lata, S. (2024). Are travellers\u0026apos; destination visit intentions influenced by social media? An information adoption theory perspective. \u003cem\u003eInternational Journal of Indian Culture and Business Management\u003c/em\u003e,\u003cem\u003e\u0026nbsp;33\u003c/em\u003e(1), 103-120.\u003c/li\u003e\n \u003cli\u003eLata, S., \u0026amp; Rana, K. (2025). Evaluating the Influence of YouTube Vlogs on Hotel Booking Decisions: An Information Adoption Model Approach. In \u003cem\u003eBuilding Power, Safety, and Trust in Virtual Communities\u003c/em\u003e (pp. 257-280). IGI Global.\u003c/li\u003e\n \u003cli\u003eLee, C., \u0026amp; Cha, K. (2025). Toward the dynamic relationship between AI transparency and trust in AI: a case study on ChatGPT. \u003cem\u003eInternational Journal of Human\u0026ndash;Computer Interaction\u003c/em\u003e,\u003cem\u003e\u0026nbsp;41\u003c/em\u003e(13), 8086-8103.\u003c/li\u003e\n \u003cli\u003eLi, C., Cao, Q., Hua, S., \u0026amp; Tao, C.-W. (2025). When AI takes the wheel: The effectiveness of AI versus human-generated content in tourism marketing. \u003cem\u003eJournal of Vacation Marketing\u003c/em\u003e, 13567667251393512.\u003c/li\u003e\n \u003cli\u003eLiang, K., Liu, H., Shan, M., Zhao, J., Li, X., \u0026amp; Zhou, L. (2024). Enhancing scenic recommendation and tour route personalization in tourism using UGC text mining. \u003cem\u003eApplied Intelligence\u003c/em\u003e,\u003cem\u003e\u0026nbsp;54\u003c/em\u003e(1), 1063-1098.\u003c/li\u003e\n \u003cli\u003eLiu, S., Lin, L., Han, X., Qiu, Y., Wang, Z., \u0026amp; Zhang, F. (2025). Evaluating the Impact of Multimodal Cognitive and Informational Interventions on Rumor Detection in Large Language Models: A DeepSeek Case Study. 2025 5th International Conference on Artificial Intelligence, Big Data and Algorithms (CAIBDA),\u003c/li\u003e\n \u003cli\u003eLong, J., Zaidin, N., \u0026amp; Mai, X. (2024). Social media influencer streamers and live-streaming shopping: examining consumer behavioral intention through the lens of the theory of planned behavior. \u003cem\u003eFuture Business Journal\u003c/em\u003e,\u003cem\u003e\u0026nbsp;10\u003c/em\u003e(1), 80.\u003c/li\u003e\n \u003cli\u003eMaleki, N., Padmanabhan, B., \u0026amp; Dutta, K. (2024). AI hallucinations: a misnomer worth clarifying. 2024 IEEE conference on artificial intelligence (CAI),\u003c/li\u003e\n \u003cli\u003eNgo, V. M. (2025). Balancing AI transparency: Trust, Certainty, and Adoption. \u003cem\u003eInformation Development\u003c/em\u003e, 02666669251346124.\u003c/li\u003e\n \u003cli\u003eNguyen, L. P. (2025). Value-driven environmental advocacy in tourism: understanding the drivers of negative eWOM among tourists visiting island destinations. \u003cem\u003eJournal of Hospitality and Tourism Insights\u003c/em\u003e, 1-20.\u003c/li\u003e\n \u003cli\u003eRadanliev, P. (2025). AI ethics: Integrating transparency, fairness, and privacy in AI development. \u003cem\u003eApplied Artificial Intelligence\u003c/em\u003e,\u003cem\u003e\u0026nbsp;39\u003c/em\u003e(1), 2463722.\u003c/li\u003e\n \u003cli\u003eRahman, M. M. (2023). Sample size determination for survey research and non-probability sampling techniques: A review and set of recommendations. \u003cem\u003eJournal of Entrepreneurship, Business and Economics\u003c/em\u003e,\u003cem\u003e\u0026nbsp;11\u003c/em\u003e(1), 42-62.\u003c/li\u003e\n \u003cli\u003eRasoolimanesh, S. M. (2022). Discriminant validity assessment in PLS-SEM: A comprehensive composite-based approach. \u003cem\u003eData Analysis Perspectives Journal\u003c/em\u003e,\u003cem\u003e\u0026nbsp;3\u003c/em\u003e(2), 1-8.\u003c/li\u003e\n \u003cli\u003eRehman Khan, H. U., Lim, C. K., Ahmed, M. F., Tan, K. L., \u0026amp; Bin Mokhtar, M. (2021). Systematic review of contextual suggestion and recommendation systems for sustainable e-tourism. \u003cem\u003eSustainability\u003c/em\u003e,\u003cem\u003e\u0026nbsp;13\u003c/em\u003e(15), 8141.\u003c/li\u003e\n \u003cli\u003eRusso, G. M., Tomei, P. A., Serra, B., \u0026amp; Mello, S. (2021). Differences in the use of 5-or 7-point likert scale: an application in food safety culture. \u003cem\u003eOrganizational Cultures\u003c/em\u003e,\u003cem\u003e\u0026nbsp;21\u003c/em\u003e(2), 1.\u003c/li\u003e\n \u003cli\u003eSable, N., Mahalle, P., Kadam, K., Sule, B., Joshi, R., \u0026amp; Deore, M. (2025). Deep learning-based approach for monitoring and controlling fake reviews. \u003cem\u003eJournal of Computational and Cognitive Engineering\u003c/em\u003e,\u003cem\u003e\u0026nbsp;4\u003c/em\u003e(3), 377-386.\u003c/li\u003e\n \u003cli\u003eSamala, N., Katkam, B. S., Bellamkonda, R. S., \u0026amp; Rodriguez, R. V. (2022). Impact of AI and robotics in the tourism sector: a critical insight. \u003cem\u003eJournal of tourism futures\u003c/em\u003e,\u003cem\u003e\u0026nbsp;8\u003c/em\u003e(1), 73-87.\u003c/li\u003e\n \u003cli\u003eS\u0026aacute;nchez-Franco, M. J., \u0026amp; Rey-Tienda, S. (2024). The role of user-generated content in tourism decision-making: an exemplary study of Andalusia, Spain. \u003cem\u003eManagement Decision\u003c/em\u003e,\u003cem\u003e\u0026nbsp;62\u003c/em\u003e(7), 2292-2328.\u003c/li\u003e\n \u003cli\u003eSchneier, B. (2025). AI and Trust. \u003cem\u003eCommunications of the ACM\u003c/em\u003e,\u003cem\u003e\u0026nbsp;68\u003c/em\u003e(8), 29-33.\u003c/li\u003e\n \u003cli\u003eSelgas-Cors, M. (2025). Sociotechnical Transformation: A Systematic Review on the Impact of Artificial Intelligence on Society and Organizations. \u003cem\u003eFinTech and Sustainable Innovation\u003c/em\u003e, 1-16.\u003c/li\u003e\n \u003cli\u003eShia, Y., Bidina, R. B. H., \u0026amp; Mamata, R. B. A Review of Relevant Research on Tourism Information Adoption Model on Social Media.\u003c/li\u003e\n \u003cli\u003eShrestha, N. (2021). Factor analysis as a tool for survey analysis. \u003cem\u003eAmerican journal of Applied Mathematics and statistics\u003c/em\u003e,\u003cem\u003e\u0026nbsp;9\u003c/em\u003e(1), 4-11.\u003c/li\u003e\n \u003cli\u003eSong, H., \u0026amp; Song, X. (2025). The effect of heritage tourism interpretation media type on tourists\u0026rsquo; eWOM: the moderating role of travel group size. \u003cem\u003eJournal of Sustainable Tourism\u003c/em\u003e,\u003cem\u003e\u0026nbsp;33\u003c/em\u003e(10), 2240-2260.\u003c/li\u003e\n \u003cli\u003eStratton, S. J. (2023). Population sampling: Probability and non-probability techniques. \u003cem\u003ePrehospital and Disaster Medicine\u003c/em\u003e,\u003cem\u003e\u0026nbsp;38\u003c/em\u003e(2), 147-148.\u003c/li\u003e\n \u003cli\u003eSujatmiko, S., Ar, D. P., Hamdat, A., \u0026amp; Salam, K. N. (2025). User-Generated Content (UGC) and Its Impact on Tourism Marketing: A Systematic Literature Review. \u003cem\u003eGolden Ratio of Mapping Idea and Literature Format\u003c/em\u003e,\u003cem\u003e\u0026nbsp;5\u003c/em\u003e(2), 97-105.\u003c/li\u003e\n \u003cli\u003eSun, X., Wang, Z., Zhou, M., Wang, T., \u0026amp; Li, H. (2024). Segmenting tourists\u0026rsquo; motivations via online reviews: An exploration of the service strategies for enhancing tourist satisfaction. \u003cem\u003eHeliyon\u003c/em\u003e,\u003cem\u003e\u0026nbsp;10\u003c/em\u003e(1).\u003c/li\u003e\n \u003cli\u003eSussman, S. W., \u0026amp; Siegal, W. S. (2003). Informational influence in organizations: An integrated approach to knowledge adoption. \u003cem\u003eInformation systems research\u003c/em\u003e,\u003cem\u003e\u0026nbsp;14\u003c/em\u003e(1), 47-65.\u003c/li\u003e\n \u003cli\u003eTan, Z., Yang, Y., Yang, X., Liu, X., Wei, Q., \u0026amp; Kong, L. H. (2025). AI in Tourism Education: A Review of AIGC Teaching and Evaluation Tools. 2025 11th International Conference on Education and Training Technologies (ICETT),\u003c/li\u003e\n \u003cli\u003eTang, Y., Weng, G., Qin, S., \u0026amp; Pan, Y. (2025). Spatial and temporal evolution of tourism flows among 296 Chinese cities in the context of COVID-19: a study based on Baidu Index. \u003cem\u003eHumanities and Social Sciences Communications\u003c/em\u003e,\u003cem\u003e\u0026nbsp;12\u003c/em\u003e(1), 1-15.\u003c/li\u003e\n \u003cli\u003eTseng, T.-L., \u0026amp; Wu, C.-C. (2024). Application of the Information Adoption Model and Technology Acceptance Model in Electronic Word-of-Mouth. \u003cem\u003eInternational Journal of Performance Measurement\u003c/em\u003e,\u003cem\u003e\u0026nbsp;14\u003c/em\u003e(1).\u003c/li\u003e\n \u003cli\u003eTuo, Y., Wu, J., Zhao, J., \u0026amp; Si, X. (2025). Artificial intelligence in tourism: insights and future research agenda. \u003cem\u003eTourism Review\u003c/em\u003e,\u003cem\u003e\u0026nbsp;80\u003c/em\u003e(4), 793-812.\u003c/li\u003e\n \u003cli\u003eVu Dinh, H., Tran Huu, T., Nguyen Thi Bich, N., Nguyen Thi Ngoc, A., \u0026amp; Doan Van, T. (2025). Effects of eWOM toward tourism destination: a bibliometric analysis and future research directions. \u003cem\u003eConsumer Behavior in Tourism and Hospitality\u003c/em\u003e,\u003cem\u003e\u0026nbsp;20\u003c/em\u003e(3), 445-459.\u003c/li\u003e\n \u003cli\u003eWang, F., Lopez, C., \u0026amp; Okazaki, S. (2025). Signaling transparency in the era of artificial intelligence. \u003cem\u003eInternet Research\u003c/em\u003e, 1-25.\u003c/li\u003e\n \u003cli\u003eWang, S., Peng, K.-L., Huang, Z., \u0026amp; Ma, L. (2025). AI-Generated Videos: Influencing Trustworthiness, Awe, and Behavioral Intention in Space Tourism E-Commerce. \u003cem\u003eJournal of Theoretical and Applied Electronic Commerce Research\u003c/em\u003e,\u003cem\u003e\u0026nbsp;20\u003c/em\u003e(4), 307.\u003c/li\u003e\n \u003cli\u003eWaqas, M., \u0026amp; Naseem, A. (2025). Artificial Intelligence in Sustainable Industrial Transformation: A Comparative Study of Industry 4.0 and Industry 5.0. \u003cem\u003eFinTech and Sustainable Innovation\u003c/em\u003e,\u003cem\u003e\u0026nbsp;1\u003c/em\u003e, A2-A2.\u003c/li\u003e\n \u003cli\u003eWei, Y., Liu, H., Zhuo, W., \u0026amp; Park, K.-S. (2025). The influence of social media attributes on impulsive travel intentions: Integrating the Stimulus\u0026ndash;Organism\u0026ndash;Response theory and Information Adoption Model. \u003cem\u003eSustainability\u003c/em\u003e,\u003cem\u003e\u0026nbsp;17\u003c/em\u003e(10), 4404.\u003c/li\u003e\n \u003cli\u003eWijaya, C. O., Wijaya, S., \u0026amp; Jaolis, F. (2025). The influence of social media content on attitude, destination image and intention of female Muslim travelers to visit halal destinations: comparison between UGC and FGC. \u003cem\u003eJournal of Islamic Marketing\u003c/em\u003e,\u003cem\u003e\u0026nbsp;16\u003c/em\u003e(2), 402-427.\u003c/li\u003e\n \u003cli\u003eXu, T., Gao, M., Liu, X., Ye, Q., Ma, L., \u0026amp; Zhu, B. (2025). Intelligent Travel Avatar: An LLM-based Tourism Quadrupedal Robot.\u003c/li\u003e\n \u003cli\u003eYamagishi, K., Canayong, D., Domingo, M., Maneja, K. N., Montolo, A., \u0026amp; Siton, A. (2024). User-generated content on Gen Z tourist visit intention: a stimulus-organism-response approach. \u003cem\u003eJournal of Hospitality and Tourism Insights\u003c/em\u003e,\u003cem\u003e\u0026nbsp;7\u003c/em\u003e(4), 1949-1973.\u003c/li\u003e\n \u003cli\u003eYe, H., Liu, T., Zhang, A., Hua, W., \u0026amp; Jia, W. (2023). Cognitive mirage: A review of hallucinations in large language models. \u003cem\u003earXiv preprint arXiv:2309.06794\u003c/em\u003e.\u003c/li\u003e\n \u003cli\u003eZelenka, J., Azubuike, T., \u0026amp; P\u0026aacute;skov\u0026aacute;, M. (2021). Trust model for online reviews of tourism services and evaluation of destinations. \u003cem\u003eAdministrative Sciences\u003c/em\u003e,\u003cem\u003e\u0026nbsp;11\u003c/em\u003e(2), 34.\u003c/li\u003e\n \u003cli\u003eZhang, L., Yang, S., Wang, W., Gao, X., \u0026amp; Liu, J. (2025). AIGC-Based Image and Video Generation Method: A Review. \u003cem\u003eJournal of Computer-Aided Design \u0026amp; Computer Graphics\u003c/em\u003e,\u003cem\u003e\u0026nbsp;37\u003c/em\u003e(3), 361-384.\u003c/li\u003e\n \u003cli\u003eZhang, X., Cheng, L., \u0026amp; Ma, G. (2024). Eliciting eudaimonic well-being in the tourism experiential space: Evidence from online reviews. \u003cem\u003eTourism Management\u003c/em\u003e,\u003cem\u003e\u0026nbsp;105\u003c/em\u003e, 104955.\u003c/li\u003e\n \u003cli\u003eZhou, T., \u0026amp; Lu, H. (2025). The effect of trust on user adoption of AI-generated content. \u003cem\u003eThe Electronic Library\u003c/em\u003e,\u003cem\u003e\u0026nbsp;43\u003c/em\u003e(1), 61-76.\u003c/li\u003e\n \u003cli\u003eZhu, J., \u0026amp; Yu, S. (2025). Generative AI for tourism and hospitality education. \u003cem\u003eCurrent Issues in Tourism\u003c/em\u003e, 1-9.\u003c/li\u003e\n\u003c/ol\u003e"}],"fulltextSource":"","fullText":"","funders":[],"hasAdminPriorityOnWorkflow":false,"hasManuscriptDocX":true,"hasOptedInToPreprint":true,"hasPassedJournalQc":"","hasAnyPriority":false,"hideJournal":true,"highlight":"","institution":"","isAcceptedByJournal":false,"isAuthorSuppliedPdf":false,"isDeskRejected":"","isHiddenFromSearch":false,"isInQc":false,"isInWorkflow":false,"isPdf":false,"isPdfUpToDate":true,"isWithdrawnOrRetracted":false,"journal":{"display":true,"email":"[email protected]","identity":"researchsquare","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":true,"externalIdentity":"","sideBox":"","snPcode":"","submissionUrl":"/submission","title":"Research Square","twitterHandle":"researchsquare","acdcEnabled":true,"dfaEnabled":false,"editorialSystem":"","reportingPortfolio":"","inReviewEnabled":false,"inReviewRevisionsEnabled":true},"keywords":"AI-mediated travel planning, Deep Thinking Function, reasoning transparency, Information Adoption Model","lastPublishedDoi":"10.21203/rs.3.rs-8736638/v1","lastPublishedDoiUrl":"https://doi.org/10.21203/rs.3.rs-8736638/v1","license":{"name":"CC BY 4.0","url":"https://creativecommons.org/licenses/by/4.0/"},"manuscriptAbstract":"\u003cp\u003eThis study investigates how DeepSeek\u0026rsquo;s Deep Thinking Function, which introduces reasoning transparency into AI-mediated travel planning, shapes travelers\u0026rsquo; psychological evaluation and adoption of AI-generated destination recommendations. Drawing on an extended Information Adoption Model (IAM), the study conceptualizes Deep Thinking Function not merely as a system design feature, but as a cognitive persuasion mechanism that reconstructs users\u0026rsquo; judgment criteria through process-based reasoning disclosure. A purposive sampling strategy was applied to recruit tourism users who had recently interacted with AI travel assistants, yielding 260 valid responses collected via an online survey. Partial Least Squares Structural Equation Modeling (PLS-SEM) was employed to assess the research model. The results reveal that Deep Thinking Function significantly enhances perceived argument quality and source credibility, which subsequently strengthen perceived information usefulness, information adoption, and destination acceptance intention. The findings indicate that, in AI-generated information environments, travelers\u0026rsquo; adoption decisions shift from outcome focused evaluation toward process-oriented cognitive appraisal, where analytical competence and methodological rigor become core foundations of trust. This study extends IAM into AI-mediated tourism contexts, conceptualizes transparency as an active persuasion mechanism rather than a static ethical attribute, and demonstrates the emergence of system centric cognitive trust in travel decision-making. Practical implications highlight the value of process aware AI interface design, transparency driven persuasion strategies, and the repositioning of destination branding toward decision logic compatibility in post-AIGC tourism ecosystems.\u003c/p\u003e","manuscriptTitle":"Deep Thinking Function in AI-Mediated Travel Planning: How Reasoning Transparency Shapes Information Adoption and Destination Acceptance","msid":"","msnumber":"","nonDraftVersions":[{"code":1,"date":"2026-03-12 10:54:59","doi":"10.21203/rs.3.rs-8736638/v1","editorialEvents":[{"type":"communityComments","content":0}],"status":"published","journal":{"display":true,"email":"[email protected]","identity":"researchsquare","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":true,"externalIdentity":"","sideBox":"","snPcode":"","submissionUrl":"/submission","title":"Research Square","twitterHandle":"researchsquare","acdcEnabled":true,"dfaEnabled":false,"editorialSystem":"","reportingPortfolio":"","inReviewEnabled":false,"inReviewRevisionsEnabled":true}}],"origin":"","ownerIdentity":"45d4a5f5-ae97-49ce-9019-ef45069fa10e","owner":[],"postedDate":"March 12th, 2026","published":true,"recentEditorialEvents":[],"rejectedJournal":[],"revision":"","amendment":"","status":"posted","subjectAreas":[],"tags":[],"updatedAt":"2026-05-03T07:23:27+00:00","versionOfRecord":[],"versionCreatedAt":"2026-03-12 10:54:59","video":"","vorDoi":"","vorDoiUrl":"","workflowStages":[]},"version":"v1","identity":"rs-8736638","journalConfig":"researchsquare"},"__N_SSP":true},"page":"/article/[identity]/[[...version]]","query":{"redirect":"/article/rs-8736638","identity":"rs-8736638","version":["v1"]},"buildId":"XKTyCvWXoU3ODBz1xrDgd","isFallback":false,"isExperimentalCompile":false,"dynamicIds":[84888],"gssp":true,"scriptLoader":[]}