Abstract
Short video browsing is a dominant medium today, significantly influencing public opinion and social dynamics. We investigate the effects of video platform Content Polarization (CP) on user perceptions and behaviors by bringing together measurements such as responses to self-report questionnaires, behavior signals, and noninvasive Electroencephalogram (EEG) signals. Our study demonstrated that just one week of exposure to polarized content can result in distinct patterns of behavior and brain activity during the perception of videos. Further, we revealed that exposure to content with diverging polarization directions, i.e. positive or negative, has extremely different effects on participants' emotional states, sentiment judgments, and brain responses. Specifically, participants' sentiment scores showed a significant decrease ([Formula: see text]=[Formula: see text], p=[Formula: see text], r=0.484) after exposure to negative content, whereas positive content led to an average sentiment score increase ([Formula: see text]=0.171, p=0.046, r=0.308). Finally, we show that analyzing participants' brain responses is even more effective at detecting exposure to polarized content than self-report judgments or behavioral signals, achieving superior predictive accuracy ([Formula: see text], [Formula: see text], [Formula: see text]).