This article presents MindShift, a new technique that uses large language models (LLMs) to address problematic smartphone use. The approach considers the user’s physical context, mental state, app usage behaviors, and goals to generate persuasive content that can help change behavior. The study found that MindShift significantly improved intervention acceptance rates and reduced smartphone use frequency. The intervention also led to a significant drop in smartphone addiction scale scores and an increase in self-efficacy. The research suggests that LLMs have potential for context-aware persuasion in other behavior change domains.
Publication date: 28 Sep 2023
Project Page: https://doi.org/10.1145/nnnnnnn.nnnnnnn
Paper: https://arxiv.org/pdf/2309.16639
