How Prompt Language Structures Influence Perceived Camera Movement and Usage Intentions in AI-Generated Video: An S–O–R Framework

With the widespread application of generative artificial intelligence (AIGC) in video content creation, the linguistic structure of prompts has emerged as a key factor influencing both generation outcomes and user experience. Grounded in the Stimulus–Organism–Response (S–O–R) theoretical framework, this study investigates how the linguistic structure of prompts affects users’ willingness to use AI-based video generation tools by shaping their perception of camera movement. The research adopts two types of prompt structures—terminology-driven and context-guided—as independent variables. Using image-to-video generation as experimental stimuli, the study employs a within-subjects experimental design and structural equation modeling (SEM) for empirical analysis. The results indicate that five perceptual dimensions—visual coherence, sense of immersion, realism and naturalness, output professionalism, and trust—significantly and positively predict users’ behavioral intention to use the tools. However, the difference in user ratings across perceptual dimensions between the two prompt types did not reach statistical significance, suggesting that prompt language structures have not yet exerted a measurable impact on subjective user perception. Furthermore, mediation analysis did not reveal significant indirect effects through perceptual variables. This study extends the applicability of the S–O–R model within AIGC contexts and demonstrates that semantically clear and logically structured natural language prompts can be equally effective in guiding AI video generation. The findings offer empirical support for reducing creative barriers and expanding user participation in generative media environments.