Exploring the Effect of Intelligent Recommendation Systems on Users’ Emotional Attachment: The Moderating Role of Personality Trait

Abstract

Users’ emotional attachment is the key to promoting their loyalty. However, few studies have explored how to improve users’ emotional attachment in the context of intelligent recommendation systems. To bridge the research gaps, our research mainly uses attachment theory and uses and gratifications theory as our research basis to explore the effect of intelligent recommendation systems (i.e., accuracy, serendipity, and personalization) on users’ emotional attachment. The mediating effect of self-construction (i.e., self-actualization, self-pleasure, and self-expressiveness) and the moderating effect of personality trait are also explored in our research. To examine our theoretical model, we conduct a survey and collect 305 valid questionnaires. The research results show: (1) Accuracy, serendipity, and personalization have positive and significant effect on users’ self-actualization, self-pleasure, and self-expressiveness respectively; (2) Self-pleasure and self-expressiveness are positively associated with users’ emotional attachment, while self-actualization do not have significant effect on users’ emotional attachment; (3) Extraversion has a positive moderating effect on the relationship between personalization and self-expressiveness. However, our research does not find evidence to support the moderating effect of extraversion on the relationship between accuracy and self-actualization as well as the relationship of serendipity and self-pleasure. This study can help developers of intelligent recommendation systems understand users’ continuous usage behavior from the perspective of emotional attachment.