If it is bad, why don’t I quit? Algorithmic recommendation use strategy from folk theories

Abstract

Users persist in utilizing algorithmic recommendations despite perceiving their adverse consequences, such as privacy invasion and filter bubbles. This behavior appears contradictory to the innate human inclination to seek benefits and avert disadvantages. Employing folk theories, this study conducted in-depth interviews with 24 users who acknowledged the detrimental effects of algorithms and maintained a non-positive attitude toward them but continued their usage. The findings revealed that users adhere to the principle of rational choice: they engage in positive risk aversion (confining their usage to entertainment domain and affirming their own abilities) and negative self-avoidance (acknowledging the inevitability of privacy invasion, imagining the additional costs of turning off algorithmic recommendations, and believing that everyone relies on algorithms, rendering the costs relatively inconspicuous). Throughout this process of self-adaptation, users assumed diverse roles (the inactive, active, resister, swayer) to achieve a harmonious state and sustain their reliance on algorithmic recommendations.