Improving emotion regulation and well-being through immersive interaction with chatbots among working adults in Taiwan

Abstract

This study presents a novel, dual-method validation of how AI chatbot interaction enhances well-being, integrating Flow Theory and Emotion Regulation (Cognitive Reappraisal). Using data from 122 professionals, we combined PLS-SEM to confirm the theoretical paths and an Artificial Neural Network (ANN) to rank variable importance. The results establish that the flow experience drives well-being, with cognitive reappraisal as the mechanism. Significantly, ANN analysis reveals that cognitive reappraisal’s total contribution is severely underestimated by traditional structural models, highlighting its critical, often-overlooked role. This work transforms our understanding of AI-driven mental health, offering actionable design principles to maximize user focus and promote proactive cognitive change among working adults in Taiwan.