The influence of recommendation algorithm's information flow on targeted advertising audience's coping behavior

Targeted advertising equipped with a recommendation algorithm can achieve accurate matching between users and recommended content, but overly precise recommendations may exacerbate negative audience reactions or behaviors. Improving the transparency of algorithm recommendation is one of the ways to address audience concerns or skepticism, and transparency guarantees the audience's right to know and thus brings more trust, which will reduce the audience's negative behavior. But increased transparency may also make the audience feel pressured or threatened, and requiring more cognitive and behavioral effort, which was called coping behavior. In order to clarify the relationship between the transparency of the algorithm recommendation and the audience's coping behavior, based on the persuasion theory, this study discusses the mechanism of the influence of the characteristics of the algorithm recommendation information flow on the audience's coping behavior of targeted advertising from the perspective of the flow mode and transmission principle of information. Based on the data of 120 online pretests and 297 formal tests, the results show that the perceived trust and perceived threat caused by the information flow characteristics of the algorithm recommendation jointly determine the possible coping behaviors of targeted advertising audiences. Additionally, users' self-efficacy regulates the relationship between mental process and coping behavior. Different from previous studies on audience coping behaviors of targeted ads, which mainly start from the perspective of participants and advertising content, this research tries to start from the perspective of information flow. The research results demystify the relationship between recommendation algorithm information flow and the audience's coping behavior, and enrich the algorithmic persuasion framework. The research results have reference value for the improvement of personalized recommendation effect, and provide a new way to further study the transparency of algorithm recommendation in the field of consumer behavior. Meanwhile, it also provides suggestions for the practices of platforms and advertisers in practice.