ABSTRACT Objectives This study introduces the concept of news‐bothers‐me perception (NBMP) and examines how its two dimensions—annoyance due to news overload and annoyance due to perceived news bias—shape news media use and political knowledge in a high‐choice digital media environment. Methods Drawing on two waves of a national online panel survey of South Korean adults (N = 1,175) conducted prior to the 2022 presidential election, we employed autoregressive lagged regression and mediation analyses to examine the longitudinal pathways through which news‐bothers‐me perception (NBMP) shapes traditional news use, social media news use, and political knowledge. Results NBMP due to news overload was negatively associated with social media news use, whereas NBMP stemming from perceived news bias was positively associated with social media news use. Neither dimension significantly affected print newspaper or television news use. Regarding political knowledge, overload‐based NBMP was negatively related to political knowledge across cross‐sectional and lagged models and showed a significant indirect effect through decreased social media news use. In contrast, content‐based NBMP was positively associated with political knowledge but showed no significant indirect effect. Conclusion These findings demonstrate that news‐bothers‐me perception (NBMP) is not a uniform construct but operates differently depending on whether annoyance stems from news overload or perceived bias. Annoyance due to news overload was associated with lower levels of social media news use and political knowledge and indirectly reduced political knowledge through decreased social media news use. In contrast, annoyance related to perceived news bias was positively associated with social media news use and political knowledge but did not exert a significant indirect effect through news media use. These results highlight the importance of perception‐based and affect‐sensitive approaches to understanding news use and political learning in contemporary digital media environments.
Kim et al. (Sun,) studied this question.