Decoding Algorithm Fatigue: The Role of Algorithmic Literacy, Information Cocoons, and Algorithmic Opacity

Abstract

Algorithmic technologies are dominating our online experiences, from content recommendations to personalized services. However, it also introduces a new challenge: algorithm fatigue. Algorithm fatigue describes the phenomenon where users experience mental and emotional exhaustion in prolonged interaction with algorithms. To decode this phenomenon, we explored personal and technical antecedents of algorithm fatigue and its impact on user behavior. Using data collected from 393 users of algorithm-driven applications, we identified three key drivers: algorithmic literacy, information cocoons, and algorithmic opacity. Particularly, while it is commonly assumed that knowing more about algorithms can facilitate more satisfying interactions, our findings indicate the opposite—greater algorithmic literacy exacerbates fatigue. Moreover, information cocoons significantly contribute to algorithm fatigue, emphasizing the detrimental effect of homogeneous content. Algorithmic opacity can also trigger fatigue, emphasizing the importance of transparency design in algorithmic systems. Additionally, we revealed a strong link between algorithm fatigue and resistance behavior, suggesting that fatigued users are more likely to resist algorithmic recommendations. Overall, this study suggests developers and policymakers design more user-centric algorithms that not only excel in personalization but also preserve user well-being.

Keywords: Algorithm fatigue, algorithmic literacy, information cocoons, algorithmic opacity, algorithm resistance