The growing concern over misinformation stems from its potential to significantly disrupt democratic governance. This study examines the case of Taiwan and employs a survey experiment featuring realworld misinformation examples to evaluate the effectiveness of factchecking corrections in reducing public misperceptions. Grounded in the dual-process theory from social psychology and informed by existing international research, the study proposes and tests four theoretically driven hypotheses—all of which are supported by the empirical findings. The results demonstrate that fact-checking can effectively correct false beliefs, particularly in non-political domains such as health and lifestyle. However, its corrective power diminishes considerably when the misinformation is political in nature—especially when it targets the ruling party. The findings further indicate that motivated reasoning plays a key role in shaping responses to political misinformation. Fact-checks are largely ineffective when they challenge misinformation targeting an individual’s political outgroup, but become significantly more persuasive when the misinformation attacks the respondent’s own political affiliation. These results underscore the highly contextual and political nature of factchecking efficacy. We argue that misinformation should not be treated merely as an individual-level problem of misinformation acceptance or attitude formation. Instead, its persistence and impact must be understood within broader political, social, and historical contexts that drive its production and spread.