Project Details
Description
To address the urgent need to reduce the burden of depression, this study aims to establish empirical evidence regarding the effectiveness of utilizing AI chatbots to shape positive expressive writing for improving mental health. In our preliminary work, an AI chatbot, EasyHealth, was developed (a website and built-in WeChat app) and powered by DeepSeek.This study aims to continue to assess the effectiveness of AI-powered expressive writing while also exploring the mechanisms of behavioral change based on usage patterns. The study is structured into two distinct parts: prompt engineering and effectiveness testing. We will undertake two experiments: the first involves fine-tuning DeepSeek to produce positive expressive writing (N=100), while the control group will focus on neutral expressive writing (NEW). The second experiment aims to assess the validity of our findings across a broader audience (N=500). Through the optimization of existing AI chatbot protocols with validated prompts, this research seeks to offer valuable insights into the application of technology driven interventions for the mitigation of depression and the enhancement of mental wellbeing among Chinese populations.
| Project number | 7020144 |
|---|---|
| Grant type | REG-Small Scale |
| Status | Active |
| Effective start/end date | 1/06/25 → … |
Fingerprint
Explore the research topics touched on by this project. These labels are generated based on the underlying awards/grants. Together they form a unique fingerprint.