Paul Katharina, Angus Douglas J, Bublatzky Florian, Wüllhorst Raoul, Endrass Tanja, Greenwood Lisa-Marie, Hajcak Greg, Jack Bradley N, Korinth Sebastian P, Kroczek Leon O H, Lucero Boris, Mundorf Annakarina, Nolden Sophie, Peterburs Jutta, Pfabigan Daniela M, Schettino Antonio, Severo Mario Carlo, Lee Shing Yee, Turan Gözem, van der Molen Melle J W, Wieser Matthias J, Willscheid Niclas, Mushtaq Faisal, Pavlov Yuri G, Pourtois Gilles
Faculty of Psychology and Human Movement Science, University of Hamburg, Hamburg, Germany.
School of Psychology, Bond University, Gold Coast, Australia.
Cortex. 2025 Mar;184:150-171. doi: 10.1016/j.cortex.2024.12.017. Epub 2025 Jan 9.
Two event-related brain potential (ERP) components, the frontocentral feedback-related negativity (FRN) and the posterior P300, are key in feedback processing. The FRN typically exhibits greater amplitude in response to negative and unexpected outcomes, whereas the P300 is generally more pronounced for positive outcomes. In an influential ERP study, Hajcak et al., (2005) manipulated outcome valence and expectancy in a guessing task. They found the FRN was larger for negative outcomes regardless of expectancy, and the P300 larger for unexpected outcomes regardless of valence. These findings challenged the dominant Reinforcement Learning Theory of the ERN. We aimed to replicate these results within the #EEGManyLabs project (Pavlov et al., 2021) across thirteen labs. Our replication, including robustness tests, a PCA and Bayesian models, found that both FRN and P300 were significantly modulated by outcome valence and expectancy: FRN amplitudes (no-reward - reward) were largest for unexpected outcomes, and P300 amplitudes were largest for reward outcomes. These results were consistent across different methods and analyses. Although our findings only partially replicate the original study, they underscore the complexity of feedback processing and demonstrate how aspects of Reinforcement Learning Theory may apply to the P300 component, reinforcing the need for rigorous ERP research methodologies.
两个与事件相关的脑电成分，即额中央反馈相关负波（FRN）和后部P300，在反馈处理中起着关键作用。FRN通常在对负面和意外结果的反应中表现出更大的振幅，而P300通常在对正面结果的反应中更为明显。在一项有影响力的脑电研究中，哈捷克等人（2005年）在一个猜测任务中操纵了结果效价和预期。他们发现，无论预期如何，FRN对负面结果的反应更大，而P300对意外结果的反应更大，无论效价如何。这些发现挑战了ERN的主导强化学习理论。我们旨在通过#EEGManyLabs项目（帕夫洛夫等人，2021年）在13个实验室中重复这些结果。我们的重复研究，包括稳健性测试、主成分分析和贝叶斯模型，发现FRN和P300都受到结果效价和预期的显著调节：FRN振幅（无奖励 - 奖励）在意外结果中最大，而P300振幅在奖励结果中最大。这些结果在不同的方法和分析中都是一致的。虽然我们的发现只是部分重复了原始研究，但它们强调了反馈处理的复杂性，并展示了强化学习理论的各个方面如何适用于P300成分，强化了对严谨的脑电研究方法的需求。
