Abstract: 【Purpose/significance】Reveal the mechanism of the impact of emotional information on user non-transaction behaviors in live e-commerce under different cultural backgrounds, help companies gain sustainable exposure, and promote user engagement mon‑ etary conversions in overseas markets.【Method/process】Drawing on the Emotion as Social Information theory, we constructed a model of non-transaction behavior influences by integrating key emotional response and reasoning cues in live commerce scenarios. Largescale samples from Taobao Live and Shopee Live were collected via multimodal data processing, with a negative binomial Poisson re‑ gression model used for cross-cultural hypothesis testing.【Result/conclusion】The study found that: from the point of view of direct emotional response, there are significant cultural differences in the impact of anchor emotions on user emotional valence and arousal; from the point of view of emotional inference information, responsiveness has an inverted U-shaped influence on user emotional va‑ lence and arousal, implying that there exists an optimal level of stimulation; personalization has an inverted U-shaped influence on user emotional valence and arousal has a positive effect; there is an inverted U-shaped relationship between user emotions and nontransactional behaviors. Popularity and user loyalty have a moderating effect between anchor emotional, responsiveness and user emo‑ tion. Based on the results of the empirical study, corresponding recommendations were made.【Innovation/limitation】This study high‑ lights cross-cultural disparities in emotional information effects, with future research recommended to expand sampling and deepen mechanism exploration.