Braun Gabriel, Yeshurun Yaara, Shetreet Einat
Sagol School of Neuroscience, Tel Aviv University, Tel Aviv 6139001, Israel.
School of Psychological Sciences, Tel Aviv University, Tel Aviv 6139001, Israel.
Proc Natl Acad Sci U S A. 2025 Jun 10;122(23):e2422396122. doi: 10.1073/pnas.2422396122. Epub 2025 Jun 5.
Despite living in an era where the mere concept of truth is increasingly contested, the cognitive processes underlying the processing of information we believe or disbelieve remain largely unexplored. In this fMRI study, we investigated how belief modulates narrative processing through belief context-the initial information indicating the speaker's credibility-and actual belief-the truth value ultimately assigned by the listener. Across two experiments, participants listened to narratives preceded by contexts explicitly stating whether the speaker was lying or telling the truth. Then, after listening to the narratives, they were asked to rate their actual belief. To investigate the effects of (dis)belief on narrative processing, we analyzed neural synchronization using inter-subject-correlation analysis and inter-subject representational similarity analysis. In both experiments, we successfully differentiated (dis)belief contexts by modeling neural synchronization patterns, despite an actual "belief-bias" at the behavioral level. This indicates a unique neural pattern related to each belief context. Furthermore, our results revealed a dissociation between belief and disbelief in both contextual and actual (dis)belief. Per each narrative, belief and disbelief were associated with increased synchrony within the default mode network, but in distinguishable parcels. These findings highlight the influence of (dis)belief on narrative processing at both behavioral and neural levels. Behaviorally, the observed belief-bias supports the notion of belief as a cognitive default. Neurally, we suggest that belief and disbelief can be understood as fostering qualitatively distinct processing or interpretation of the same narrative, which are then reflected in shared neural responses among individuals who hold similar belief states.
尽管生活在一个真理概念日益受到质疑的时代,但我们相信或不信的信息处理背后的认知过程在很大程度上仍未得到探索。在这项功能磁共振成像(fMRI)研究中,我们研究了信念如何通过信念背景(表明说话者可信度的初始信息)和实际信念(听众最终赋予的真值)来调节叙事处理。在两个实验中,参与者听取了在背景中明确说明说话者是在说谎还是说实话之后的叙事。然后,在听完叙事后,他们被要求对自己的实际信念进行评分。为了研究(不)信念对叙事处理的影响,我们使用受试者间相关性分析和受试者间表征相似性分析来分析神经同步。在两个实验中,尽管在行为层面存在实际的“信念偏差”,我们通过对神经同步模式进行建模成功区分了(不)信念背景。这表明与每个信念背景相关的独特神经模式。此外,我们的结果揭示了在背景信念和实际(不)信念方面,信念与不信之间的分离。对于每一个叙事,信念和不信都与默认模式网络内同步性的增加相关,但在可区分的区域。这些发现突出了(不)信念在行为和神经层面上对叙事处理的影响。在行为上,观察到的信念偏差支持了信念作为认知默认的观点。在神经层面上,我们认为信念和不信可以被理解为促进对同一叙事进行质的不同的处理或解释,然后反映在持有相似信念状态的个体之间共享的神经反应中。