Coping with Homogeneous Information Flow in Recommender Systems: Algorithmic Resistance and Avoidance

Abstract

Despite being able to predict accurate user-item interactions, recommender systems also lead to homogeneous information flow, resulting in negative online experiences. However, users’ coping processes for this stressful event remain basically unclear. Drawing on coping theory, this study examines how users appraise two stressors (information narrowing and information redundancy) in homogenous information flow, especially in different user experiences, and thus adopt algorithmic resistance and avoidance coping strategies. Data (N = 407) were collected based on a survey. The empirical results indicate that information narrowing and information redundancy are significantly and negatively related to challenge appraisal, and exert significant and positive effects on threat appraisal. The moderating effect denotes that user experience strengthens the relationship between information narrowing and information redundancy to threat appraisal. Threat appraisal further triggers algorithmic resistance and avoidance of users. The paper provides several implications for theory and research by revealing the mechanism through which recommender systems users cope with homogeneous information flow. Our findings also provide new insights for designers of recommender systems platforms.