Proceedings of the Annual Meeting of the Cognitive Science Society

How to improve the persuasiveness of online requests is crucial to achieve acceptance and foster positive social relationships. The effectiveness of persuasion strategies and the influence of the sequence in which these strategies are applied have been demonstrated in the literature. However, existing research has largely overlooked the importance of linking the sequential persuasion strategies to the specific purpose. In this study, we first employ a few-shot Iterative Collaboration Method (ICM) to identify the purpose of the online requests, referencing human needs as well as the persuasion strategies used. Then, the sequential patterns of persuasion strategies supporting respective purposes are mined. Finally, Large Language Models (LLMs), incorporating the identified effective sequential strategies are used to rewrite the original requests. The results indicate that the sequence of strategies used for different purposes, can significantly increase the level of persuasion. The code and dataset can be found at https://github.com/phillip2f/Seq-Strategies-and-Purpose.