Generation Z preferences for cultural heritage landscapes using questionnaire and neuroscience ERP methods

Abstract

In the context of heritage tourism development, accurately identifying the preferences of new-generation tourists is critically important. Based on dual-process theory, this paper combines questionnaire surveys with neuroscience (ERP) techniques to construct an “explicit–implicit” dual-pathway model, using four types of heritage landscapes in Jingdezhen as examples to investigate the preference formation mechanisms of Generation Z. Research findings indicate that religious and traditional folk landscapes receive higher explicit preferences; “Place imagination-awe” constitutes a chain mediation, with visual aesthetics positively moderating its pathway and influence preferences; Preferred landscapes trigger stronger P200 and LPP responses, while non-preferred landscapes elicit higher N200, presenting neural processing features such as attention capture and conflict detection. The research innovatively introduces ERP technology to heritage perception studies, promoting the interdisciplinary integration of heritage science and neuroscience, and provides empirical support for heritage site planning and neuromarketing targeting youth groups.

Introduction

Cultural heritage landscapes, as material carriers of civilizational memory, not only embody historical value but also serve as core resources for cultural tourism¹. Statistics from the United Nations World Tourism Organization show that cultural tourism accounts for nearly 40% of global tourism revenue, making it one of the fastest-growing market segments²; the International Council on Monuments and Sites notes that it plays a significant role in promoting employment, income, heritage protection, and sustainable development³. Countries have successively introduced relevant policies, such as China’s “14th 5-Year Plan for Cultural and Tourism Development”⁴ and the European Union’s “Horizon Europe” program (2021–2027)⁵. Heritage-based cultural tourism responds to tourists’ demands for diverse cultural experiences while being viewed as an effective pathway for promoting heritage conservation and revitalization^1,6. As Generation Z, born between 1995 and 2012, gradually becomes the main force in tourism, their preferences for authenticity, emotional interaction, and personalized cultural expression are reshaping cultural heritage tourism trends^7,8. However, most heritage sites still show deficiencies in narrative renewal, responsiveness to youth needs, and taste compatibility, constraining their potential to stimulate cultural identity and consolidate consensus among young people. Furthermore, existing research primarily focuses on heritage conservation⁹, tourism development^10,11, and tourist experiences, exploring dual perspectives from both tourists¹² and residents¹³, while research targeting specific generational groups remains relatively limited. How to construct a methodological system that can identify and quantify Generation Z preferences has become a key issue for heritage site protection and revitalization.

Meanwhile, the heritage tourism experience is a dynamic, real-time interactive process, and traditional methods primarily based on questionnaires and self-reporting¹⁴, struggle to capture visitors’ immediate emotions and psychological changes, lacking mixed methods that combine subjective evaluations with objective measurements (such as neurophysiological indicators) that capture instantaneous reactions¹⁵. Furthermore, existing research primarily focuses on explicit preferences based on tourist satisfaction or travel intention assessments^16,17, while the implicit perceptions and underlying neural mechanisms of tourist preferences have not been sufficiently revealed.

Dual Process Theory (DPT) shows promise in providing a theoretical foundation for effectively explaining tourists’ complex preference processes, indicating that human cognition consists of two parallel and complementary information processing systems: System 1 (automatic processing) and System 2 (controlled processing). System 1 is characterized by rapid, automatic, and emotional features, relying on heuristic cues to make intuitive judgments; System 2 is relatively slow, controlled, and logical, requiring higher cognitive resources for deep reasoning^18,19,20, together forming the core mechanism of human judgment and decision-making²¹, and in the context of cultural heritage, tourists often exhibit complex and ineffable emotional responses, reflecting the interaction between System 1 and System 2. DPT was initially widely applied in the field of psychology to reveal biases and heuristic effects in human judgment²², and was also employed in social psychology to explain differences between implicit and explicit attitudes, as well as the potential impact of stereotypes on behavior^21,23,24. In economics, this theory helps understand behavioral anomalies that deviate from rational assumptions²⁵. In recent years, DPT has been introduced into fields such as consumer behavior²⁶ and industrial design²⁷, expanding the understanding of complex human cognitive behaviors. However, its application in cultural heritage tourism research remains relatively limited.

Within the DPT framework, this study considers “place imagination” as a key variable in the cognitive pathway. It integrates tourists’ impressions, associations, and emotions toward destinations, representing a process through which individuals attribute special meaning and emotional value to a place^28,29. Research indicates that tourists typically form specific expectations about destinations before travel, which serves as an important antecedent factor influencing tourist attitudes, visitation intentions, and recommendation behaviors³⁰. Place imagination not only serves as a cognitive bridge between tourist expectations and actual experiences, but also largely determines the depth and emotional value of the tourism experience^31,32. When it presents positive idealized characteristics, it particularly stimulates tourists’ enthusiasm for participation and emotional resonance³³. Through the ‘meaning-making’ process, abstract cues are transformed into individual meaning landscapes, activating emotions and promoting emotional connections, thereby becoming an important antecedent to the formation of place attachment³⁴. However, imagination may also “obscure reality,” causing tourists to indulge in preset imagery while neglecting the essence of the actual site³⁵. Existing research supports the mediating role of place imagination: Ji et al.²⁹ demonstrated that place imagination has a mediating effect between virtual tourism and tourist emotional attachment; Aboalganam et al.³⁶ found that destination images on social media first influence the imagery imagination in tourists’ minds, thereby indirectly affecting actual visitation intentions. In summary, place imagination is both a cognitive bond in the interaction between tourists and places, and a key driver of emotional attachment and behavioral intentions. Meanwhile, the differential influence of landscape types has been confirmed in relevant studies^37,38,39. The study proposes the following hypotheses:

H1: Different types of heritage landscapes have significant differences in Generation Z tourists’ preferences.

H2: Place imagination plays a mediating role between heritage landscapes and tourist preferences.

Awe is also a key mediating variable in this study, referring to the profound emotional response individuals experience after perceiving intense external stimuli, with its core being the cognition of vast forces and the epiphany of self-diminishment⁴⁰. Based on their triggers, tourists’ awe primarily stems from three types of stimuli: social (such as religious ceremonies), physical (such as magnificent mountains and rivers), and cognitive (such as art or theories that break cognitive frameworks)⁴¹. Keltner and Haidt⁴² point out that the awe experience must simultaneously satisfy “perceptual vastness” and “need for cognitive accommodation,” meaning that people actively adjust their existing schemas to integrate new information when facing extraordinary landscapes. In recent years, scholars have conducted relevant research: Powell et al.⁴³ proposed a five-dimensional model, while Coghlan et al.⁴⁴ constructed a tripartite framework of “physiological impact—uniqueness—pattern alteration,” highlighting its dual cognitive-emotional attributes. The sense of awe can not only enhance satisfaction, preference, and revisit intentions⁴⁵, but also promote prosocial and pro-environmental behaviors in ecological and religious tourism^46,47. The mediating role of awe in tourist experiences has been confirmed⁴⁸, particularly in heritage tourism, where “vastness” evoked by place imagination often triggers awe, which then deepens tourists’ identification with and investment in heritage through self-diminishment and emotional appreciation⁴⁶. That is, cognitive constructs activate awe, and awe enhances positive intentions. Therefore, the study proposes the following hypotheses:

H3: Awe plays a positive mediating role in the influence of heritage landscapes on tourist preferences.

H4: Place imagination and awe form a chain mediating pathway between heritage landscapes and explicit preferences.

Visual esthetics as an important dimension of tourism experience has received extensive empirical support^49,50, and represents a core pathway through which humans perceive and interpret their environment⁵¹. As the predominant sensory pathway, vision exerts a determinative influence on the development of landscape preferences⁵². It encompasses both basic feature judgments, such as color, form, and symmetry, as well as higher-order esthetic evaluations, including complexity, coherence, and mystery⁵³. In heritage landscapes, visual esthetics forms the foundation of visitors’ first impressions and sustained interest^54,55. Porteous⁵⁶ points out that the esthetic characteristics of heritage landscapes modulate the psychological distance between tourists and destinations, affecting their level of engagement. Cognitive processing fluency theory and Leder’s model of esthetic experience suggest that easily processed visual information not only reduces cognitive load but also continuously influences experience throughout multi-stage perceptual processes^57,58. Zhang and Xu⁵⁹ further note that high-quality landscapes strengthen place imagination by stimulating associations, while tourists’ prior knowledge reciprocally shapes esthetic perception. Zhuang et al.⁶⁰ also found that geographical imagination and esthetic emotions formed under musical backgrounds can enhance emotional connections to places. Moreover, existing research has demonstrated a connection between esthetic experience and the emotion of awe⁶¹. Based on this, we propose Hypothesis 5:

H5: Visual esthetics positively moderate the effect of place-based imagination on awe; specifically, higher visual esthetics allow place imagination to elicit greater tourist preferences.

To detect instantaneous neural reactions, this research incorporates event-related potential (ERP) methodology. Numerous studies have demonstrated the influence of emotions (affective states) on tourists’ motivations, behaviors, choices, and satisfaction with tourism and hospitality⁶². Although subjective measurement tools such as survey questionnaires have significant value in capturing tourist perceptions, they still have certain limitations in capturing instantaneous emotions and subconscious reactions: on one hand, tourists often find it difficult to accurately recall or express momentary feelings; on the other hand, certain implicit responses are also difficult to directly present through subjective methods¹⁵. Among numerous neurophysiological measurement techniques, ERP have received significant attention due to their high temporal resolution in capturing brain cognitive and emotional processes. ERP extracts brain electrical signals evoked by specific events through electroencephalography (EEG), reflecting processes such as attention allocation, emotional arousal, and semantic cognition within 10–100 ms after stimulus presentation⁶³. ERP typically contains three core parameters: amplitude, latency, and scalp distribution. Mean amplitude can be used to analyze response differences under various conditions; latency (measured in milliseconds) refers to the time from stimulus onset to peak potential; scalp distribution reflects brain activation regions⁶⁴. For instance, Zuckerman et al.⁶⁵ in their research on attachment styles, emotional feedback, and neural feedback discovered relevant ERP components, P200/P400, and selected relevant brain region electrodes, frontal and prefrontal electrodes (Fp1, F7, Fp2, F8, F3, and F7) for analysis. This characteristic makes ERP particularly suitable for studying visual, emotional, and cognitive responses in tourism contexts. ERP methods have been applied in fields such as industrial design⁶⁶, consumer decision-making⁶⁷, and language research⁶⁸, with further exploration needed in the field of heritage science studies.

P100 is an ERP component associated with early visual cortical processing, typically peaking within 100–150 ms after stimulus onset. It reflects the individual’s selection and recognition of perceptual features such as physical attributes of images, including color and brightness^69,70. The P200 component, which peaks around 200 ms post-stimulus, is associated with the allocation of selective attentional resources driven by emotional arousal^71,72. Preference evaluation involves not only attention but is also related to emotional experiences^57,73. When presented with positive or favorable stimuli, P200 amplitude increases accordingly. For example, Fudali-Czyż et al.⁷⁴ observed that when subjects viewed beautiful paintings, their P200 amplitudes were larger than when viewing less attractive paintings. Based on research findings, we propose Hypothesis 6:

H6: Preferred heritage landscapes elicit larger P200 amplitudes.

N200 is an ERP component that peaks within the 200–350 ms time window after stimulus onset, closely related to cognitive processes such as automatic stimulus recognition, selective attention, and perception, and is considered an endogenous negative component^75,76. Researchers have found that people exhibit larger N200 amplitudes when confronted with disliked or low-esthetic items. For example, Handy et al.⁷⁷ observed increased N200 amplitude in the frontocentral region when participants viewed disliked logos in a commercial branding study. Similarly, Telpaz et al.⁷⁸ also found that products with low preference indices elicited greater N200 responses compared to highly preferred products. Therefore, combining the above research, we propose Hypothesis 7:

H7: Disliked heritage landscapes elicit larger N200 amplitudes.

LPP is a late, sustained positive component that typically reaches its peak within 300–800 ms after stimulus onset and can persist for at least several hundred milliseconds, associated with sustained attention allocation and emotional evaluation^79,80. Compared to emotionally neutral stimuli, intense emotional stimuli result in more pronounced LPP^81,82. Additionally, according to our previous research findings, there exists a close connection between LPP and preference evaluation⁸³. We found that disliked tiles elicited larger LPP amplitudes. Kurohara et al.⁸⁴ also found that pleasant images elicited greater LPP responses compared to unpleasant images. Therefore, we propose Hypothesis 8:

H8: Preferred heritage landscapes evoke larger LPP in people.

In summary, this study integrates heritage landscape types, place imagination, sense of awe, and visual esthetic variables under the DPT framework to construct an “explicit (questionnaire)-implicit (ERP)” dual-pathway preference model (Fig. 1), aiming to analyze Generation Z’s preference mechanisms for cultural heritage landscapes, focusing on answering: (1) Do different types of heritage landscapes trigger significant preference differences? (2) What mediating and moderating factors exist in the formation of explicit preferences? (3) How do landscapes stimulate rapid, automated implicit neural responses and their phasic characteristics? (4) Are explicit and implicit preference results consistent or complementary, and what kind of interactive relationship do they present?

Fig. 1

Theoretical framework.

The main contributions of this research include: (1) Introducing ERP neural measurement into heritage tourism research, establishing a mixed methodology system that integrates subjective surveys with neural indicators, overcoming the limitations of previous research relying solely on questionnaire surveys, and precisely capturing Generation Z tourists’ explicit evaluations and implicit neural responses to heritage landscapes; (2) Based on the DPT, a dual-pathway mechanism was constructed to explain how heritage landscapes influence preferences, thereby enhancing the robustness of the conclusions and enriching the interpretive framework of heritage tourism experience research;(3) Identifying neural indicators P200, N200, and LPP applicable to heritage preference assessment, confirming the application potential of cognitive neuroscience in heritage research; (4) The empirical analysis reveals a chained mediation effect between place imagination and awe, as well as a moderating role of visual esthetics. These findings provide insights into the psychological mechanisms underlying the interaction between Generation Z and cultural heritage.

In conclusion, the research findings not only promote disciplinary integration between heritage science and cognitive neuroscience but also provide more objective and precise scientific evidence for heritage site evaluation, planning, and utilization, holding significant theoretical value and practical significance.

Methods

This study employs a research strategy centered on a specific population group, rather than a cross-group comparative design. This experimental paradigm, which uses generational cohorts as units of analysis, has been widely applied in the fields of tourism and neuroscience^85,86,87. Grounded in the theoretical model, a mixed-methods design combining questionnaires and EEG experiments was adopted, focusing on Generation Z as the core sample to investigate their preference mechanisms for cultural heritage landscapes. Using Jingdezhen, a popular heritage tourism destination for Generation Z, as a case study, consists of two parts: Study One employs subjective scales to assess explicit preferences for four types of heritage landscapes, and uses SPSS 27 to analyze mediating and moderating effects; Study Two collects ERP data based on the EEG Oddball paradigm, preprocesses it using EEGLAB, and analyzes neural responses of components such as P200, N200, and LPP through repeated measures ANOVA to reveal implicit preference mechanisms.

Research area

This study selects Jingdezhen City in the northeastern part of Jiangxi Province as the research area (Fig. 2). Jingdezhen began porcelain production during the Five Dynasties period, developed rapidly during the Song and Yuan dynasties, and established imperial kilns at Zhushan during the Ming and Qing dynasties, becoming the national center for porcelain production and earning the title “Millennium Porcelain Capital.” There are currently 160 porcelain industrial sites, 108 historic alleyways, and 41 provincial or higher-level key cultural protection units⁸⁸, forming a rich ceramic cultural heritage system. In 2019, the State Council approved the establishment of the “Jingdezhen National Ceramic Culture Inheritance and Innovation Experimental Zone”; In 2025, its “Handicraft Porcelain Industry Remains” was included in China’s World Heritage nomination list, further highlighting its cultural strategic position and inheritance value.

Fig. 2

The study area.

In recent years, Jingdezhen has accelerated the integration of ceramic culture with cultural tourism, and since becoming one of the first cities to offer a 240-h transit visa exemption in 2024, tourism popularity has continued to rise. During the 3-day Qingming Festival holiday in 2025, Zhushan District received 409,000 tourist visits, generating tourism revenue of 388 million yuan⁸⁹. With its excellent reputation, Jingdezhen has become a phenomenal “internet-famous” destination, attracting numerous Generation Z tourists, ceramic artists, and digital nomads who are active in cultural venues such as Sculpture Porcelain Factory, Taoxichuan, and Sanbao Ceramic Village, engaging in creative experiences and interdisciplinary exchanges. The convergence of profound historical heritage and cultural practices of emerging groups provides an ideal field background and observation platform for this study.

Questionnaire materials

In the initial stage of the research, we invited seven experts from Jiangxi Provincial Department of Culture and Tourism, Jingdezhen Ceramic University, and the Intangible Cultural Heritage Protection Center to assess the main cultural spatial characteristics of Jingdezhen. Combining expert assessments, field investigations, and existing classification standards, and integrating the diverse heritage characteristics of ceramic settlements, we defined four types of cultural heritage landscapes (Table 1):

• (1) Historical urban landscapes: referencing UNESCO⁹⁰ and Höftberger⁹¹, we selected images of historical districts, ancient city walls, and traditional residences preserved during the process of urban development and evolution;

• (2) Industrial heritage landscapes: drawing on studies by Dimitriou⁹² and Zhang et al.⁹³, we focused on production sites and facilities formed during the industrialization process, emphasizing their cultural memory and regenerative potential in urban renewal;

• (3) Religious sacred landscapes: referencing the research of Niedźwiedź and Baraniecka-Olszewska⁹⁴, we selected scene images of famous temples, churches, religious sites, or sacrificial places, highlighting their solemn and sacred qualities;

• (4) Traditional folk landscapes: with definitions referenced from ÖZER⁹⁵, we selected scene images related to intangible cultural heritage, such as handicraft workshops, traditional folk customs, and artisans creating on-site, showcasing local traditional lifestyles, the living inheritance atmosphere of craft culture, and reflecting the continuity of intangible heritage.

Table 1 Four categories of heritage landscapes selected based on literature and expert screening

The experimental image materials came from two sources: first, high-quality images selected through keyword searches on social media platforms such as Weibo and Xiaohongshu, and second, photographs taken on-site by three researchers in September 2024. Photography was conducted at 10 A.M. and 4 P.M., with location selection covering typical sites (such as Jingdezhen’s Taoxichuan Creative District as a revitalized ceramic industrial heritage, the Wind-Fire Immortal Master Temple reflecting fire deity worship, and the Jingdezhen Ancient Kiln Folk Custom Expo Area, which houses the world’s oldest ceramic production line and serves as an intangible heritage inheritance base). All photography was completed using Canon EOS 5D cameras under clear weather conditions. All images were required to have consistent composition, clear image quality, and conform to experimental standards, with 84 images ultimately selected.

Drawing on methods from previous research¹⁵ that used 7-point Likert scales to assess material attributes, this study invited three local experts to rate the contextual characteristics of the images online, selecting the 10 highest-scoring images from each category as experimental materials (40 images in total). All images were uniformly processed for size, specifications, and pixel dimensions using Adobe Photoshop 2023 to control for non-experimental variables that might interfere with visual perception.

Questionnaire design and data collection

The scales used in the questionnaire were appropriately adapted from existing research combined with the context of this study, all utilizing 7-point Likert scales (1 = strongly disagree, 7 = strongly agree), to enhance scale discrimination. Place imagination referenced Ji et al.²⁹ and Li et al.³¹, measuring landscape-inspired place imagery and associations; The sense of awe was based on items developed by Coghlan et al.⁴⁴ and Yan and Jia⁴⁶, evaluating the awe and admiration emotions experienced by participants; Visual esthetics was based on Kah⁹⁶ and Kirillova et al.⁵⁴, used to measure esthetic evaluation of landscapes; The preference scale was constructed using Ren⁹⁷, used to assess preference intensity and future selection willingness (see Table 2 for details).

Table 2 Composition of variable items

The questionnaire was distributed through the Credamo (www.credamo.com) online platform. Participants were randomly assigned to one of four different cultural heritage type groups using a between-subjects design, with each participant evaluating only one heritage type. To ensure semantic consistency, all English scales underwent a translation-back translation procedure, and a small-scale preliminary survey was conducted before formal testing. Results indicated that all scales had good reliability and structural validity. All subjects were native or native-level Mandarin speakers, and after excluding non-Generation Z individuals (born outside 1995–2012) and invalid responses, a total of 320 valid questionnaires were included, comprising 231 females (72.2%) and 89 males (27.8%). Participants explicitly consented to the use of their survey responses for academic research and publication in scholarly journals, with all data being anonymized to prevent disclosure of personal identifying information. Furthermore, to further validate whether the study specifically targets a population with distinct characteristics, an additional control group consisting of non-Generation Z individuals (aged >30 years, n = 320; 100 males and 220 females) was recruited. They completed a rating scale on heritage preferences, followed by a comparative analysis of group differences.

Experimental participants

According to G*Power 3.1 and relevant research recommendations⁸³, the required sample size for this experiment was 22 people, and 22 Generation Z participants were actually recruited. Preliminary test results showed that subjects had no significant perception differences regarding the scenes used in the experiment. Due to excessive artifacts in the EEG data of one subject, they were excluded, resulting in a final inclusion of 21 samples (including 11 females). All subjects were right-handed with normal vision and no history of neurological or psychiatric disorders.

The experiment was conducted in the Cognition and Decision Laboratory of the School of Business Administration at Huaqiao University, approved by the ethics committee (M2023009), and complied with the relevant norms of the Declaration of Helsinki. Before the experiment, participants were informed that the process was safe and painless to alleviate their nervousness, and they signed informed consent forms. Subjects’ consent was confirmed for the use of their relevant data in academic research, while they were informed of their right to terminate the experiment at any time and withdraw permission for data usage, with all data being kept strictly confidential and anonymized.

The experiment was completed in a quiet room, and participants could terminate at any stage. After the experiment, each subject received a compensation of 80 Chinese yuan.

Experimental stimuli

This study employs four categories of Jingdezhen cultural heritage images consistent with Study 1 as visual stimuli to assess tourists’ potential preferences for heritage landscapes. The experiment adopts a modified Oddball paradigm⁸⁰, a classic design frequently used to detect attention, cognitive processing, and implicit preferences, with a typical structure of alternating high-frequency (85%) and low-frequency (15%) stimuli presentations that induce specific neural responses. This experiment divides stimuli in a 4:1 ratio: non-target stimuli consist of 40 heritage images selected from Study 1, while target stimuli comprise 10 validated neutral everyday object images from the Chinese Affective Picture System (CAPS). All images undergo standardized processing (size, resolution, and aspect ratio) to ensure consistency in experimental control (see Fig. 3).

Fig. 3

The two types of materials in this part of the experiment are four categories of heritage landscapes as non-target stimuli, and household items from CAPS as target stimuli.

Experimental procedure

The experimental task was programmed and presented using Eprime 3.0. The experimental procedure is shown in Fig. 4. The experiment was conducted in an indoor environment with good and stable light sources. The monitor was approximately 65 cm away from the subject, with a visual field maintained at 30.5° × 20.1° (width × height), a resolution of 1920 × 1080, and a size of 24 inches. Subjects were required to view 50 pictures (40 heritage landscape pictures and 10 daily necessities pictures), and to reduce sequence effects, all stimuli were presented in a pseudo-random form, repeated 5 times, for a total of 250 trials. A simple pre-experiment was conducted before the formal experiment. First, instructions appeared on the screen, and subjects were required not to make a choice when heritage landscapes appeared, but to press the letter Y key when daily necessities appeared. During the experiment, try not to clench your teeth or move your head significantly. After understanding the content, subjects pressed the space bar to confirm their familiarity with the experimental operation and entered the formal trial phase.

Fig. 4

ERP experiment flow.

A cross cursor first appeared on the screen, lasting for 1800 ms, followed by the stimulus appearing for a presentation time of 2000 ms, then entering a gray interval phase lasting 1300–1500 ms, after which the next stimulus appeared, alternating until the end of the experiment. The entire experiment lasted 50 min, including two brief rest periods.

ERP data collection and analysis

Data processing. Procedure is shown in Fig. 5. Using the Neuroscan Synamp2 Amplifier system, EEG data were continuously recorded with a 64-channel Ag/AgCl electrode cap according to the international 10–20 system, with a sampling rate of 500 Hz, reference electrodes at bilateral mastoids (M1, M2), and a ground electrode at FCZ. Eye movement electrodes (VEOG, HEOG) were fixed below the right eye orbit and at the outer canthus of the left eye, respectively, with all electrode impedances controlled below 5 kΩ. The data acquisition software was Curry7.

Fig. 5

Three stages of the ERP experiment.

After the experiment, EEG data was imported into the EEGLAB toolbox on the MATLAB 2023a platform for preprocessing, with steps including: (1) converting data formats and electrode positioning; (2) applying 0.1–30 Hz bandpass filtering to remove artifacts; (3) baseline correction using 200 ms pre-stimulus; (4) extracting data segments from −200 ms to +1000 ms time window; (5) manually removing noise segments and bad channels, and removing eye movement, muscle, and other artifacts through independent component analysis (ICA). As shown in Fig. 6a, a list of independent components for individual subjects was derived by running the ICA analysis option. In the list, ICA component 6(b) can be identified as typical eye-blink components, with the judgment criteria shown in Fig. 6b, including anterior brain activity distribution, high component ranking, and high low-frequency energy. This component indicates frequent eye-blinking behavior by the subject during the experiment, necessitating the removal of related interference components through ICA to achieve smoother waveforms.

Fig. 6: Independent Component Analysis (ICA) components in EEG preprocessing.

a ICA component list for a single subject. b Example of an eye-blink ICA component identified by its typical distribution and ranking.

After preprocessing, the average proportion of valid trials retained under each experimental condition exceeded 90%. Finally, the data was re-referenced to the average of bilateral mastoids to reduce the impact of reference electrode differences on waveforms.

In this study, we focused on three ERP components: P200, N200, and LPP. Based on the experimental content, whole-brain component observations, and previous studies^65,69,98,99, nine whole-brain electrode sites were selected (Fig. 7). We selected 6 electrode sites in the frontal-central area (F1, F2, FZ, C1, C2, and CZ) with a time window of 150–200 ms to analyze the P200 component, to assess differences in early perceptual emotional processing for different heritage stimuli; Similarly, with a time window of 250–350 ms, 6 electrode sites in the frontal-central area (F1, F2, FZ, C1, C2, and CZ) were selected to analyze the N200 component, examining the negative bias or emotional resource allocation triggered by different categories of landscapes as non-target stimuli; Finally, the LPP component of 6 electrodes in the central-parietal brain region (C1, C2, CZ, P1, P2, PZ) with a time window of 500–600 ms was selected to analyze tourists’ preference evaluation for stimuli. Furthermore, to examine whether differences in the physical properties of experimental stimuli (e.g., luminance, color, or spatial frequency) might have influenced ERP components, and to enhance the interpretability and transparency of the ERP results, this study conducted a supplementary analysis of the P100 component. The analysis focused on parieto-occipital electrodes (P1, P2, and PZ) within the 80–120 ms time window. Assessing the P100 component allowed us to evaluate the visual input balance across the four stimulus categories, thereby confirming that subsequent ERP differences are more likely attributable to preference processing rather than primary visual discrepancies.

Fig. 7

Distribution map of nine electrode sites selected for the study.

For data analysis, we extracted the average amplitudes for all different conditions and regions, and used SPSS 27.0 to conduct 4 (type: historical urban landscape, industrial heritage landscape, religious sacred landscape, and traditional folk landscape) × 9 electrode sites repeated measures analysis of variance (ANOVA) on subjective data and ERP data. Tests for normality and homogeneity of variance were conducted before analysis, and means and standard deviations were reported in descriptive statistics. Additionally, Greenhouse-Geisser correction was applied in cases where sphericity tests were not passed.

Results

Questionnaire results

Study 1 data analysis was conducted using SPSS 27, unfolding in two steps: first, testing the reliability and validity of the questionnaire (Table 3), and second, exploring the causal relationships between variables.

Table 3 Reliability and validity test results

Reliability test results showed that Cronbach’s α values for each variable ranged between 0.827 and 0.893, with composite reliability (CR) all above the recommended standard of 0.70, indicating good internal consistency of the scales. The Kaiser–Meyer–Olkin value was 0.800, and Bartlett’s test of sphericity was significant, satisfying the prerequisites for exploratory factor analysis. Four factors with eigenvalues greater than 1 were extracted through varimax rotation, with a cumulative explained variance of 79.77%. All item factor loadings were >0.4, with average variance extracted (AVE) ranging between 0.696 and 0.802, supporting the convergent and structural validity of the scales, verifying that the questionnaire measurement quality is reliable and can be used for subsequent analysis.

To investigate the relationships among the four core variables—place imagination, sense of awe, visual esthetics, and preference—this study conducted a Pearson correlation analysis. As shown in Table 4, all variable pairs exhibit significant positive correlations (p < 0.01), with the highest correlation observed between place imagination and preference (r = 0.402). Moderate positive correlations were also found between place imagination and awe (r = 0.353), awe and preference (r = 0.373), and visual esthetics and preference (r = 0.271). These findings indicate significant linear relationships among the variables. Furthermore, variance inflation factor (VIF) analysis revealed VIF values ranging from 1.099 to 1.160 and tolerance values above 0.8 for all predictors, suggesting no severe multicollinearity and providing a solid foundation for subsequent mediation and moderation analyses.

Table 4 Results of Pearson correlation analysis

In the subjective preference section, a two-way ANOVA was conducted with 2 (generation: Generation Z vs. non-Generation Z) × 4 (heritage types: historical, industrial, religious, and folk heritage) factors. Results revealed a significant main effect of generation F(1,632) = 22.535, p < 0.001, η²_p = 0.034, and a significant main effect of heritage type F(3,632) = 9.494, p < 0.001, η²_p = 0.043. However, the interaction effect between generation and heritage type was not significant F(3,632) = 7.908, p = 0.160, η²_p = 0.008. These findings preliminarily indicate the heritage preference characteristics of the Generation Z population. Descriptive statistics (M ± SD) showed that, compared to historical urban landscapes (4.49 ± 1.21) and industrial heritage landscapes (4.63 ± 1.27), religious sacred landscapes (5.21 ± 1.08) and traditional folk landscapes (5.25 ± 1.26) were more favored by Generation Z.

To test the impact pathway of heritage type on tourist preferences and the mediating roles of place imagination and sense of awe, this study used the PROCESS macro (Model 6) proposed by Hayes¹⁰⁰ for serial mediation analysis. Data was processed through the PROCESS plugin (v4.2) in SPSS 27.0, with 5000 Bootstrap resamples and a confidence interval set at 95%; if the CI does not contain 0, the mediation effect is significant.

Regression analysis results (Table 5) indicated that heritage type had a significant positive impact on tourist preferences (total effect = 0.285, p < 0.001, 95% CI [0.167, 0.403]), supporting hypothesis H1. Further analysis showed that heritage type significantly influenced place imagination (β = 0.299, 95% CI [0.189, 0.410]), and place imagination also positively predicted tourist preferences (β = 0.296, 95% CI [0.184, 0.408]), with a significant mediation effect (indirect effect = 0.089, 95% CI [0.043, 0.124]), verifying hypothesis H2.

Table 5 Mediation effect results in study 1

Simultaneously, heritage type significantly predicted sense of awe (β = 0.128, 95% CI [0.013, 0.244]), which had a positive effect on preferences (β = 0.256, 95% CI [0.149, 0.363]), with a significant mediation pathway (indirect effect = 0.033, 95% CI [0.002, 0.064]), supporting hypothesis H3.

Further analysis indicated that place imagination had a significant predictive effect on sense of awe (β = 0.128, 95% CI = [0.013, 0.244]), and the serial mediation pathway of heritage type, place imagination, sense of awe, and tourist preference was significant (indirect effect = 0.025, 95% CI = [0.010, 0.038]), suggesting that heritage type first stimulates tourists’ place imagination, then further enhances their preferences through strengthened sense of awe, verifying hypothesis H4.

Additionally, this study used Model 14 in the PROCESS plugin to test the moderating effect of visual esthetics on the relationship between place imagination and awe. Results showed: in the basic model, place imagination significantly predicted awe(β = 0.365, p < 0.001), R² = 0.125; After adding visual esthetics, both main effects were significant (place imagination: β = 0.320, p < 0.001; visual esthetics: β = 0.177, p < 0.001), with R² increasing to 0.168. After further introducing the interaction term (place imagination × visual esthetics), the interaction effect was significant (β = 0.213, p < 0.001), with the model’s explanatory power further increasing to R² = 0.275, indicating that visual esthetics has a significant positive moderating effect between place imagination and awe, supporting hypothesis H5.

Simple slope analysis showed that when visual esthetics was at a high level ( + 1 SD), the impact of place imagination on awe was strongest (β = 0.681, p < 0.001); it was also significant at a medium level (β = 0.370, p < 0.001); but not significant at a low level (–1 SD) (β = 0.059, p = 0.355). This indicates that the higher the visual esthetics, the stronger the positive impact of place imagination on awe, demonstrating an enhancing moderation effect.

Experimental results

Study 2 analyzed the P100, P200, N200, and LPP brain components through repeated measures analysis of variance (heritage landscape type × electrode sites). Table 6 presents the descriptive statistics of the ERP components. Figure 8 shows the grand-average waveforms.

Fig. 8

Grand-average waveforms of valid components elicited by four types of heritage landscapes in the frontal (FZ), central (CZ), and parietal (PZ) regions.

Table 6 Statistical results of P100, P200, N200, and LPP components under different conditions, including time windows and brain region electrodes (M ± SD)

A P100 component was observed in the parieto-occipital region within the 80–120 ms time window (see Figs. 8, 9). Repeated-measures ANOVA revealed a significant main effect of electrode site, F(2,19) = 5.347, p = 0.014, η²_p = 0.360, while the main effect of landscape type was not significant, F(3,18) = 0.549, p = 0.655, η²_p = 0.084. These findings indicate that the different heritage landscapes did not elicit significant differences in early visual cortical P100 responses, suggesting that the stimuli were balanced in terms of low-level visual features, thereby effectively ruling out physical attributes as a source of influence on subsequent ERP components.

Fig. 9

Topographic maps of the P100 component.

P200 component was observed within the 150–200 ms time window, and the corresponding electrode topographic maps were drawn accordingly (Fig. 10). Repeated measures ANOVA results indicated a significant main effect of electrode, F(5,16) = 13.639, p < 0.001, η²_p= 0.810; the main effect of landscape type was also significant, F(3,18) = 12.443, p < 0.001, η²_p= 0.675. Further pairwise comparisons revealed that the P200 amplitude elicited by religious sacred landscapes (−4.65 ± 1.11 µV) was significantly higher than historical urban landscapes (−6.29 ± 1.04 µV, p = 0.003) and industrial heritage landscapes (−7.36 ± 1.06 µV, p < 0.001), indicating that religious sacred landscapes triggered a stronger early attentional awareness response within this time window, supporting hypothesis H6. However, compared to traditional folk landscapes (−5.48 ± 1.21 µV, p = 0.120), the difference did not reach statistical significance.

Fig. 10

Topographic maps of the P200 component under four landscape conditions.

The N200 component was extracted within the 250–350 ms time window, with electrode topographic maps shown in Fig. 11. Repeated measures ANOVA showed a significant main effect of electrode, F(5,16) = 76.167, p < 0.001, η²_p= 0.792; the main effect of landscape type was also significant, F(3,18) = 16.806, p < 0.001, η²_p= 0.737. Pairwise comparisons further indicated that the less-preferred industrial heritage landscapes in Study 1 (−12.14 ± 0.125 µV) elicited significantly larger N200 amplitudes than religious sacred landscapes (−8.40 ± 0.133 µV, p = 0.003) and traditional folk landscapes (−10.09 ± 0.128 µV, p = 0.003), supporting hypothesis H7. However, compared to living landscapes (−11.35 ± 0.101 µV, p = 0.124), this difference did not reach a significant level.

Fig. 11

Topographic maps of the N200 component under four landscape conditions.

Within the 500–600 ms time window, this study conducted repeated measures ANOVA on the LPP component and drew corresponding electrode topographic maps (Fig. 12). Descriptive statistics for LPP are presented in Table 5. Analysis results showed a significant main effect of electrode, F(5,16) = 17.607, p < 0.001, η²_p= 0.846; the main effect of landscape type was also significant, F(3,18) = 16.278, p < 0.001, η²_p= 0.731. Further pairwise comparison results indicated that traditional folk landscapes (−0.944 ± 0.863 µV) elicited significantly larger LPP amplitudes than historical urban landscapes (−4.239 ± 0.772 µV, p < 0.001) and industrial heritage landscapes (−2.945 ± 0.899 µV, p = 0.009), but the difference was not significant compared to religious sacred landscapes (−0.825 ± 0.704 µV, p = 0.848). The data suggest that more preferred traditional folk landscapes and religious sacred landscapes elicited larger LPP amplitudes, supporting hypothesis H8.

Fig. 12

Topographic maps of the LPP component under four landscape conditions.

Discussion

This study, grounded in dual-process theory, employed both questionnaire surveys and ERP experiments. The findings revealed the underlying mechanisms of heritage experience preferences among Generation Z participants, potentially reflecting certain cognitive tendencies within this cohort. Study 1 demonstrates that religious sacred and traditional folk landscapes score highest in explicit preferences, while historical urban and industrial heritage landscapes score relatively lower. Mediation analysis reveals that “place imagination” and “sense of awe” exert chain-mediating effects on how landscape types influence preferences, while “visual esthetics” demonstrates a positive moderating effect on the relationship between place imagination and preference. In Study 2, ERP results indicate that religious and folk landscapes elicit stronger positive potentials during the P200 and LPP periods, while industrial heritage landscapes trigger more significant negative responses during the N200 period, reflecting higher conflict or lower preference. The convergence between implicit neural data and explicit measurements provides partial support for the explanatory power of dual-process theory in understanding heritage tourism preference pathways.

Findings at the explicit subjective pathway level (Study 1) revealed that “place imagination” plays a significant mediating role in heritage tourism preferences. The preference of Generation Z for a particular heritage site may depend on their ability to mentally construct and immerse themselves in its cultural narrative. This also implies a generational tendency: they are accustomed to forming preconceptions about destinations through digital platforms or social media before their trip, often embarking on their journey with vivid expectations¹⁰¹. Relph¹⁰² indicates that the meanings and images people attribute to a place shape their experience of that location, while Lewicka and other scholars further emphasize that strong emotional-cognitive bonds formed based on memory and imagination not only help tourists establish emotional connections with places but also evoke positive emotions and deepen emotional attachment^31,35,103. In terms of this study, Generation Z shows higher preferences for religious and folk heritage sites; as Rise and Schwan¹⁰⁴ describe, heritage sites confer place value through symbolic architecture, ritual activities, or historical attributions, and by providing sufficient “imaginative space,” young tourists more readily invest emotionally and develop identification. Interestingly, this finding differs from the related viewpoint that Generation Z, as digital nomads, generally prefers technology and creativity⁸; however, this study confirms that Generation Z seeks destinations with cultural depth and authenticity to satisfy internal identity needs¹⁰⁵, which aligns with the recent trend of Chinese youth enthusiastically embracing indigenous culture (such as Hanfu and traditional festivals). In contrast, the cultural elements of industrial heritage often present a “de-imaginative” tendency; as Szubert et al.¹⁰⁶ point out, the technological development of tourist destinations may cause experiences to remain more at rational and functional levels, making it difficult to evoke deeper emotional resonance. This study also observes that preferences for urban historical landscapes are relatively low, indicating to some extent that, compared to religious and folk heritage, without distinctive cultural characteristics, it is difficult to construct “place imagination” that generates emotional involvement for Generation Z, thus making it challenging to obtain higher preference evaluations.

“Awe” typically refers to the intense emotional impact experienced by individuals when confronting vast, sublime, or sacred entities^42,48. This study finds that the sense of awe similarly plays a significant mediating role in the formation of heritage tourism preferences. In the case of Jingdezhen, temples frequently enshrine Taoist wind and fire deity statues symbolizing fire god worship and utilize yellow decorations that historically symbolized imperial power¹⁰⁷, complemented by intangible ceramic heritage exhibitions and folk activities such as kiln god sacrifices, creating a collective ritualized sacred experience where young tourists often develop reverence and amazement toward history, craftsmanship, and authoritative beliefs, thereby evoking strong feelings of awe¹⁰⁸. This aligns with related research, as Yan and Jia⁴⁶ discovered that tourists readily experience feelings of awe when confronted with religious art or ancient architecture; Wang and Lyu¹⁰⁹ noted the powerful driving force of awe in tourist experiences, finding that awe-inducing tourism activities (such as viewing magnificent ruins or natural scenery) can enhance tourists’ satisfaction with travel and value perceptions, while Zhang et al.¹¹⁰ also found that digital technologies can be utilized to convey place memories, stimulating pre-travel awe and thereby motivating visitors’ intentions to engage in heritage tourism. In fact, awe generates short-term excitement and curiosity, the latter being one of the key motivational factors driving Generation Z’s travel behavior¹¹¹. This curiosity not only deepens their emotional connection to heritage sites but also contributes to the creation of memorable travel experiences¹¹², thereby enhancing their willingness to visit and their preference for such destinations^41,46.

This study also confirms that place imagination and sense of awe exhibit a chain mediation effect in the formation of cultural experience tourism preferences. Results indicate that place imagination, as a cognitive antecedent, significantly influences the generation of awe, which together shape tourists’ preferences for heritage experiences. Previous research has indicated that the sense of awe often stems from the perception of “vastness” phenomena and the resulting need for cognitive accommodation⁴²; this study further discovers that this cognitive process largely depends on tourists’ “place imagination” of heritage sites for guidance. Di Masso et al.¹¹³ emphasize that place imagination plays a key role in constructing cognitive connections between people and environments, while Chen and Rahman¹¹⁴ also point out that prior knowledge can enhance the imagination or understanding of emotional experiences in cultural heritage. These perspectives are confirmed in the case of Jingdezhen ceramic heritage: when tourists possess richer imagination or understanding of ceramic production techniques, they are more likely to develop a sense of awe toward “exquisite craftsmanship,” thereby enhancing their preference for authentic cultural experiences.

Visual esthetics demonstrate a significant positive moderating effect on young tourists’ selection of heritage landscapes in this study. This may be related to the characteristics of Generation Z, who grew up in the era of digital media and place significant value on destination imagery and visual content¹¹⁵. Specifically, religious landscapes in Jingdezhen (such as temples) possess rich visual information due to exquisite statues and architectural decorations, while folk scenes embody the harmonious beauty of communication between people and local culture, making it easier for Generation Z tourists’ place imagination to transform into positive preference tendencies. This aligns with the “visual stimulation-esthetic response-attitude” pathway proposed by Kaplan⁵³: pleasing or stunning visual effects can strengthen people’s imagination and emotional investment in places. Kirillova and Lehto⁴⁹ also point out that the esthetic quality of tourism environments is a key factor in forming positive destination attitudes. Notably, the moderating effect of esthetics also suggests boundary conditions for the influence of place imagination: when the pleasantness of visual esthetics is higher, the impact of place imagination on preference is most significant. From an interactionist experience model perspective, the influence of one factor (cognitive image) on the outcome (preference) may depend on the level of another factor (sensory environment)¹¹⁶. This also echoes the “perceptual fluency” theory: when environments are more visually acceptable, people’s processing of cognitive impressions tends to be smoother and more positive, and Awe as a positive emotion, has a profound impact on tourist behavior—including pro-environmental actions—by promoting cognitive engagement^42,44, thus creating a stronger promoting effect on preferences⁵⁷.

Study 2 employed EEG and the Oddball experimental paradigm to compare four categories of heritage landscapes (historical urban landscapes, industrial heritage landscapes, religious sacred landscapes, and traditional folk landscapes), aiming to observe the main components of P100, P200, N200, and LPP to discuss the implicit neural pathways of rapid automation in the dual processing of Generation Z’s tourism preferences.

It was first observed that the P100 component did not show significant differences. The P100 typically occurs approximately 100 ms after stimulus onset and primarily reflects perceptual and attentional responses of the primary visual cortex to visual stimuli^69,70. This stage is largely influenced by the physical characteristics of stimuli. This result is not inconsistent with the findings of Study 1; rather, it suggests a dual-processing characteristic among Generation Z participants. At the explicit evaluation stage, there is a marked focus on visual esthetics, whereas no significant differences are observed during the early, automatic perceptual phase—these only emerge through higher-order cognitive processing. Additionally, this result may reflect the fact that Generation Z has grown up in a highly media-saturated and image-rich environment, which may influence their attention and memory¹¹⁷. This possibly leads to reduced sensitivity to low-level visual differences and necessitates greater attentional resources to detect such distinctions.

P200 is closely associated with higher-order visual stimulus processing and rapid attention allocation, and is considered the primary stage of the evaluation process^63,64. Di Russo et al.¹¹⁸ indicated that P200 appears within 200–300 ms, primarily demonstrating early visual recognition functions for stimuli. Simultaneously, studies have found that P200 is also linked to early automated emotional responses¹¹⁹, particularly pronounced in the frontal lobe region, viewed as an important marker of cognitive classification processes. Previous literature has also shown that when viewing preferred or positive stimuli, the amplitude of P200 tends to be larger^15,74. The results of this study are consistent with this: people’s preferred religious and folk landscapes elicit larger P200 components. This may be because temples or Taoist temples can form a strong initial visual impact on tourists through colorful murals, solemn hall layouts, and ritualized activities at specific time periods, thereby allocating more attentional resources, indicating that Generation Z’s affinity for these heritage types is not only a rational choice but also rooted in their automatic cognitive processing. In contrast, although historical urban or industrial heritage landscapes possess historical and cultural value, their visual symbols are relatively weak in terms of “novelty” or “emotional directionality,” making it difficult to trigger strong early attention in an extremely short time, and thus unable to induce higher P200 amplitudes.

N200 typically appears within approximately 200-300 ms after stimulation, belonging to an endogenous negative ERP component, closely related to selective attention and the processing of complex, incongruent stimuli. In visual working memory tasks, the N200 that appears after people retrieve target stimuli is also considered to be related to the physical characteristics of the stimuli^120,121. Existing research indicates that when individuals face negative or disliked items, N200 amplitude may be higher^77,78, which is consistent with the results of this study: when subjects viewed industrial heritage landscapes with lower ratings, N200 amplitudes were more significant. For Generation Z, the visual imagery of factories may differ from their expectations, triggering cognitive conflicts; compared to intuitive demonstrative folk landscapes and sacred landscapes with strong characteristics, industrial heritage landscapes require more resources to identify and understand, also implying higher cognitive loads. The identification challenge deepened the subjects’ negative emotions, thereby further amplifying the N200 amplitude.

LPP is typically considered a key indicator of emotional processing, particularly sensitive to stimuli with emotional arousal and personal significance^83,84. Extensive literature confirms that when individuals invest more attention and emotional assessment in positive or highly preferred stimuli, LPP amplitude increases accordingly^122,123,124. Furthermore, some scholars further indicate that LPP is not only related to immediate emotional responses but also reflects the process of individuals constructing deeper meanings and encoding memories for stimuli¹²⁵. The results of this study echo this: subjects showed significantly increased mean LPP amplitude when viewing their preferred heritage landscapes (such as traditional folk activities), indicating they invested emotional and cognitive resources of higher intensity during this period. In the heritage tourism context, high activation of LPP signifies tourists’ “deep engagement” with landscapes, encompassing both cognitive information integration and strong emotional resonance. This may also be associated with characteristics specific to Generation Z. Compared to other demographic groups, they tend to favor personalized, interactive, and immersive experiences¹²⁶. Combined with this study’s traditional folk cases, such as pottery throwing and kiln god worship ceremonies, these living crafts or ritual performances, being relatively rare in daily life and highly dynamic and symbolic, can effectively attract and maintain Generation Z tourists’ attention in a short time; after higher-order cognitive processing, they induce higher emotional arousal with awe and immersion, potentially contributing to sustained neural activation, triggering LPP components, with results echoing subjective surveys.

It should be noted that this study, using a Generation Z sample, provides only preliminary insights into this group’s subjective and objective preference tendencies. The upbringing and media exposure of Generation Z may have amplified certain effects—such as their heightened sensitivity to visual novelty and stronger desire for immersive experiences. However, such trends may not be exclusive to Generation Z. Moreover, the consistency between the questionnaire data and ERP results in this study does not undermine the applicability of the DPT; rather, it highlights the temporal and pathway-specific characteristics of the cognitive processing involved, instead of focusing solely on outcome differences. This suggests that, during preference formation, the fast, automatic emotional processing of System 1 and the slow, controlled rational evaluation of System 2 may operate in parallel or in a complementary manner. This aligns with classic views of DPT, such as those proposed by Evans and Stanovich¹⁹, who noted that Systems 1 and 2 may function competitively or complementarily. Lawrie et al.²⁷ further identified three primary models of dual-processing: sequential, parallel, and hybrid. Although this study does not directly test distinctions among these models, the findings offer preliminary support for the applicability of DPT within heritage experience contexts.

The key findings of the study are as follows:

• (1) Generation Z tourists exhibit a dual-process mechanism in forming preferences toward cultural heritage. At the explicit level, this is significantly associated with cognitive-affective factors such as place imagination, awe, and visual esthetics. At the implicit level, their preferences may be shaped by rapid and automatic neural processing, which could interact with the aforementioned factors in shaping young tourists’ attitudes toward heritage landscapes.

• (2) The chain mediation pathway of “heritage type → place imagination → sense of awe → preference” indicates that heritage’s profound cultural value and emotional symbols can effectively enhance young tourists’ sense of meaning and immersion, thereby forming higher explicit preferences.

• (3) Visual esthetics positively moderates the mediating chain between place imagination and awe, suggesting that esthetically enriched environments may further enhance Generation Z’s affinity for heritage landscapes.

• (4) Implicit preferences exhibit a progressively deepening process. Beginning with the P100 stage, the brain does not show significant differentiation in response to heritage landscapes, indicating that primary visual processing has not yet diverged. From the P200 stage onward, participants begin to exhibit initial attraction and emotional processing toward stimuli with distinct esthetic or cultural symbolism. During the subsequent N200 phase, cognitive conflict arises when landscape features are difficult to identify or do not match psychological expectations. In the later LPP stage, landscapes that elicit stronger preferences induce enhanced emotional processing and cognitive engagement, reflecting deep emotional connections and the construction of mental imagery in response to the stimuli.

The theoretical contributions of this study are primarily manifested in four aspects: Firstly, by adopting DPT as the analytical foundation and constructing a preference model that integrates implicit affective processes (System 1) and explicit cognitive evaluations (System 2) in parallel, this study offers a new perspective for understanding the mechanisms behind preference formation among Generation Z in complex heritage contexts, addressing the traditional neglect of internal perception in prior research. Secondly, the introduction of ERP technology facilitated the acquisition of heritage preference-related neural indicators P200, N200, and LPP, enabling neurological measurement of tourists’ attention and emotional processing, enhancing the objectivity and explanatory power of the research, promoting interdisciplinary integration between neuroscience and cultural heritage research, and providing methodological support for heritage landscape preference and planning. Third, it validates the chain mediation pathway of “heritage landscape → place imagination → sense of awe → preference,” clarifying how cultural heritage transforms into behavioral preferences through cognition and emotion, enriching the logical chain of tourism experience theory. Finally, it reveals Generation Z participants exhibit implicit preferences for landscapes characterized by ritualistic elements, symbolism, and living heritage. This challenges the view held by some scholars that Generation Z lacks interest in historical culture. The findings suggest that their preferences may stem from a deep emotional resonance rooted in authenticity and participatory experiences, providing empirical evidence for understanding their cognitive orientation toward cultural heritage.

This research has significant practical value. First, in terms of heritage tourism management and marketing, it finds that Generation Z has strong interest in attractions rich in cultural connotations, provided that the presentation method is sufficiently interesting or immersive. It is recommended to shape heritage spaces into cultural fields with a sense of ritual and sacredness through “symbolic reinforcement” and “narrativization” strategies, combined with visual beautification and interactive experiences, stimulating their curiosity and willingness to participate in historical places, thereby enhancing tourism attractiveness. Second, in experience design, digital interpretation and atmosphere creation should be used to guide tourists into the emotional channel of “place imagination-sense of awe.” This can be achieved by incorporating impactful historical elements or immersive craft participation (such as ceramics-making ceremonies), complemented by means such as lighting, music, and elevated viewing positions, meeting Generation Z’s expectations for multi-sensory and deep meaningful experiences. Finally, this study validates the feasibility of neural assessment tools in tourism preference research, suggesting that scenic areas use technologies such as EEG and eye-tracking to pre-test promotional videos or exhibition content, assess emotional responses through EEG components (such as ERP), compensate for the inadequacies of questionnaire methods, optimize cultural presentation more precisely, and stimulate emotional resonance and preferences among young groups.

Although this study partially reveals the dual-pathway mechanism of Generation Z’s heritage landscape preferences, there remain several limitations.

First, the sample primarily focuses on the characteristics of Generation Z. While this cohort is an increasingly important segment of the tourism market, the absence of intergenerational comparison means that we cannot conclude these findings are unique to this group. Our results only represent the characteristics observed within the Generation Z cohort in this study and offer new neuroscientific evidence for related research. Future studies may consider including other generational groups, such as Generations Y and X, to explore potential similarities and differences in the neural processing pathways involved in heritage tourism, thereby enhancing the generalizability and theoretical depth of the findings.

Second, the sample is primarily based on the Generation Z group from China, which may not adequately represent preference characteristics across different cultural backgrounds. Future research should conduct cross-cultural comparisons to explore the moderating effect of cultural factors on the dual-pathway mechanism.

Third, EEG measurements under laboratory conditions still differ from real heritage tourism contexts; subsequent studies could attempt to capture tourists’ neural responses in field environments or with VR/AR technology to more authentically reflect actual environmental conditions.

Fourth, this study primarily focuses on neural responses to visual stimuli, whereas actual tourism experiences often involve multi-sensory stimulation and dynamic interactions. Future studies could employ multimodal neuroscience methods to explore the influence of auditory, olfactory, and other stimuli on heritage preferences.

Fifth, due to the complexity of research variables, only four types of heritage landscapes were selected, possibly failing to comprehensively cover the diversity of heritage landscapes; future research could incorporate more types of cultural heritage, expand research universality, apply this framework to practice, and provide theoretical guidance for global heritage protection and innovative utilization.

Finally, although this study draws on DPT to interpret both implicit and explicit processing pathways in preference formation, we do not advocate for an overextension of the theory. The relevant analyses and interpretations primarily serve the construction of a research paradigm and a multifaceted understanding of the data. Future studies should further examine the applicability and limitations of this theory in the domain of heritage perception by employing more complex task paradigms or refined psychological indicators.

In conclusion, this study conducted a preliminary exploration of the dual-pathway preference formation mechanism regarding cultural heritage among Generation Z by integrating questionnaire surveys with ERP neuroimaging techniques. This work introduces a neuroscientific perspective and methodological example to the field of heritage tourism research. Moreover, it offers valuable insights for developing heritage site conservation and cultural creative marketing strategies that better respond to the younger generation’s demand for immersive cultural experiences.