Speaker: Dr. Naijia Liu

Dr. Naijia Liu is an Assistant Professor in the Government Department at Harvard University.

Moderator: Ze Han

Ze Han is a Ph.D. candidate in the Department of Politics at Princeton University.

Lecture Summary:

To observe the experiences and behaviors that recommendation algorithms behind social platforms bring to users, Dr. Naijia Liu and her collaborators built a network video platform. They conducted experiments in two policy areas with severe polarization issues, including gun control policy and minimum wage policy. The platform they built provided a customized interface for experimental subjects, manipulating the ideological direction and intensity of recommended video content to adjust the degree of change in the existing YouTube recommendation algorithm. Participants were randomly assigned to different groups. Through three sets of independent multi-round experiments, the authors estimated the impact of the ideological bias of recommendation algorithms on public policy preferences, emotional polarization, and perceptions of media hostility. The study did not find reliable evidence supporting the idea that algorithmic recommendations lead to attitude polarization.