Recently, I came across an article from an official psychological content account I follow, which discussed the crucial role of high-quality conversations in our lives. This immediately piqued my interest, prompting me to conduct further research based on the references provided. I reviewed research papers by Ruggeri et al. (2020), Biehle et al. (2012), and Oomen et al. (2022), which inspired me to design my own psychological experiment.

For this experiment, I used volunteer sampling by posting an online advertisement in Changsha, Hunan, targeting university students aged 18 to 23. I successfully recruited 10 participants who provided informed consent.

To assess happiness, I utilized a simplified Chinese translation of the Psychological Wellbeing Scale (PWS-18) (Ryff & Keyes, 1995). This 18-question questionnaire evaluates a person's perception of their current life state and future outlook. Additionally, after participating in a behavioral test, participants answered an open-ended question: "After reading this chat, how do you feel about your current state of mind? Does this chat provide a sense of meaning? Why?"

For the quality of conversation, participants selected two WeChat conversation screenshots from their past interactions—one representing a "high-quality positive conversation" and the other a "neutral conversation," based on criteria from Hall et al. (2023). They then read one of these chats in a "high-quality positive conversation" group, a "neutral conversation" group, or a "no conversation" group (where they focused on a blank sheet of paper).

During the experiment, participants wore an EEG device while reading their assigned chat. Afterward, they removed the EEG device, completed the PWS-18 again, and answered the behavioral question about their state of mind and the sense of meaning derived from the chat. Participants were then compensated for their time.

My hypothesis was that high-quality conversations would significantly impact happiness levels compared to neutral conversations and no conversation. However, due to the small sample size, the experiment's generalizability and reliability were limited, preventing me from validating this hypothesis conclusively. 

​