AI Emotional Social Media Intervention under the Effect of Movie Character for Young Women

Abstract

This study investigates the impact of AI-simulated movie character dialogues on the emotional language of single young women. Employing a two-week crossover design, thirty participants were randomly assigned to groups with alternating intervention sequences. Group A engaged with an AI-simulated movie character while Group B don’t get intervention for the first week, Group B engaged with an AI-simulated movie character while the Group A don’t get intervention. Sentiment lexicon-based analysis was applied to all conversation transcripts to quantify emotional expression. Results demonstrated a statistically significant increase in positive emotional language and greater emotional vocabulary diversity during movie character interactions compared to control conditions. No carryover effects were observed between weeks. These findings suggest AI-simulated movie characters can effectively induce positive shifts in emotional expression, offering a compensatory mechanism for intimacy needs. The study provides a novel framework for quantifying AI's emotional impact and contributes to understanding digital intimacy solutions.