This study investigates the complex interplay between language arousal and visual features in social media content across diverse destination contexts. By examining how these multimodal elements jointly influence user engagement, it provides strategic insights for optimizing destination marketing communications.
Grounded in social information processing theory and dual coding theory, this study employs multi-modal machine learning methods to analyze large-scale social media data. It systematically examines how language arousal interacts with visual attributes (hue, saturation and brightness) to affect engagement across urban and natural destination contexts.
First, language arousal significantly enhances user engagement, with a stronger impact in urban destinations than in natural destinations. Second, visual elements moderate the impact of language arousal in context-specific ways: hue amplifies arousal effects in natural destinations, high saturation reduces processing efficiency and appropriate brightness enhances persuasiveness through environmental congruence. Third, the moderating effects of visual features on language arousal demonstrate greater significance in natural destinations than in urban contexts, revealing a complex interaction.
This research advances tourism scholarship by pioneering the integration of linguistics and visual content analysis through machine learning methodologies. It addresses a critical theoretical gap by examining how the complex interplay between language arousal and visual elements shapes user engagement specifically in destination marketing, which differs fundamentally from conventional product marketing due to the experiential and high-involvement of tourism destinations.
