How do agricultural heritage landscape elements influence tourist perception? Evidence from online text and images of the Longji Rice Terraces
Year: 2026
Type: preprint
Abstract: Although tourism development and utilisation help raise public awareness of agricultural cultural heritage, they also run the risk of simplifying its multidimensional value into easily consumable scenic symbols in tourism communication. To analyse this contradiction, this study aims to explore how specific landscape elements of agricultural cultural heritage shape visitors’ perceptions and, in turn, influence their overall emotional responses. Taking the Longji Rice Terraces in Guangxi as a case study, this research analyses user-generated online reviews and their accompanying images. By integrating thematic clustering, aspect-level sentiment analysis, and visual language models, the study compares the distribution of attention toward different landscape elements in text and images, as well as their cross-modal co-occurrence relationships, and further examines the association between these features and visitors’ overall evaluations. The results indicate significant differences in text-image representation: images primarily capture the seasonal colours and spatial textures of the terraces, while text tends to describe the overall scenery. Visitors’ attention is primarily focused on visual features, while agricultural production practices, local knowledge, and sociocultural values are relatively marginalised in text-and-image communication. The study reveals the selective expression characteristics of agricultural cultural heritage in tourism communication, which helps identify the heritage values that are highlighted and those that are downplayed, and provides empirical evidence for optimising the interpretation of agricultural cultural heritage and its sustainable management.
Source: Research Square
Authors: Maojun Xiong, Xiyue Wang
Institution: Beijing Forestry University
Language: English
Subfield: Cultural Studies
Field: Social Sciences
Domain: Social Sciences
Open Access status: green
OpenAlex ID: W7235843857
