Search Abstracts | Symposia | Slide Sessions | Poster Sessions
Encoding as True, Revising Later: Dual-Phase Processing for Narrative Labeled as Fake News
Poster Session D, Thursday, October 1, 4:30 - 6:30 pm, Wangari Maathai
Liu lanfang1; 1Bay Area School of Applied Psychological Sciences, Beijing Normal University at Zhuhai, 2Faculty of Psychology, Beijing Normal University
Introduction ：Narratives constitute a powerful means of communication that shape memory, beliefs, and decision-making, and are therefore frequently exploited to convey misinformation. Although pre-exposure notifications are often introduced to mitigate these effects, they appear to have limited effectiveness. This discrepancy highlights the need for a deeper investigation into how the brain processes and retains narrative information in the face of explicit veracity notifications. Methods ： Experiment 1 combined fMRI and behavioral measures to investigate how veracity notifications may influence the online comprehension and offline memory processes of narrative. Sixty participants listened to a narrative labeled as factual or fake news, followed by covert recall. Experiment 2 used the same material and experimental paradigm but systematically varying the timing of the veracity notification: informing participants that the narrative was factual (Condition 1) or fake (Condition 2) before listening; informing of its fake status before listening and again just before retrieval (Condition 3); and informing twice, before and immediately after listening (Condition 4). Hidden Markov Modeling in combination with Support Vector Machine was applied to explore whether brain network dynamics differed between the fake- and fact-labeled conditions. Results ： Experiment 1 showed that whether the narrative was presented as factual or fake news did not significantly affect participants' subjective engagement or memory fidelity. Consistent with this, the global brain state patterns during narrative comprehension were indistinguishable between the two conditions. However, during subsequent covert memory retrieval, brain state dynamics diverged such that they reliably decoded participants’ condition labels (fake vs. fact). Moreover, brain state features during retrieval predicted memory fidelity, with the functional relevance of specific states differing by condition. Experiment 2 replicated these behavioral findings in a new cohort. It further demonstrated that providing a fake-news notification immediately prior to covert retrieval (in addition to pre-exposure) enhanced memory fidelity. Conclusion： this study reveals a dual-phase model of narrative processing: the brain initially encodes the fake narrative as if it were factual, but revise its representation in light of veracity information during offline retrieval. The dual-phase processing model may reflect a principle of the human brain in response to misinformation, whereby early comprehension operates under a truth bias, fostering immersive engagement, whereas later memory processes involve reconstructive operations that initiate re-evaluation and belief updating. These findings identify the post-exposure phase as a critical window for intervention against misinformation.
Topic Areas: Meaning: Discourse and Pragmatics,