Learning Signals Within an Iterated Social Exchange

在迭代的社交交换中学习信号

• 批准号: 7158475
• 负责人: Brooks Casas
• 金额: $ 4.6万
• 依托单位: BAYLOR COLLEGE OF MEDICINE
• 依托单位国家: 美国
• 财政年份: 2006
• 资助国家: 美国
• 起止时间: 2006-07-01 至 2007-06-30
• 项目状态: 已结题
• 来源: https://reporter.nih.gov/project-details/7158475
• 关键词: behavior prediction; behavioral /social science research tag; bioimaging /biomedical imaging; clinical research; computational neuroscience; functional magnetic resonance imaging; human subject; interpersonal relations; learning; neural information processing; neuroimaging; neuropsychology; postdoctoral investigator; psychological reinforcement; reinforcer; social behavior; social competition; social cooperation; social model; social psychology; socioeconomics

DESCRIPTION (provided by applicant): Trust and reciprocity are critical social signals enabling the development of cooperation among conspecifics. This proposal builds on preliminary work that has found that behaviors signaling cooperation, such as generosity in an exchange, elicit activity in the same striatal structures that are responsive to primary rewards. The current work will explore the computational and neural mechanisms through which social learning occurs within iterated exchanges of trust. First, a prediction error model of reinforcement learning will be adapted for this social dilemma and used to guide the analysis of functional magnetic resonance imaging data acquired during an iterated exchange. The specific aim will be to identify whether activity in the dorsal striatum reflects 'predictions' or 'prediction error' signals within the context of a 10-round exchange of trust -- a critical distinction for developing a computational account of the neural mechanisms involved in normative and aberrant social learning. Second, this work will be devoted to the development of novel analytic methods, including cross-brain independent component analysis, for use in identifying neural representations of shared social models.

描述（由申请人提供）：信任和互惠是重要的社会信号，能够在同种个体之间发展合作。这一建议建立在初步研究的基础上，该研究发现，表明合作的行为，如交换中的慷慨，在纹状体结构中引起的活动与对主要奖励作出反应的相同。目前的工作将探索在反复的信任交换中社会学习发生的计算和神经机制。首先，强化学习的预测误差模型将适用于这种社会困境，并用于指导在迭代交换期间获得的功能磁共振成像数据的分析。具体目标将是确定在10轮信任交换的背景下，背纹状体的活动是否反映了“预测”或“预测错误”信号——这是开发涉及规范和异常社会学习的神经机制的计算账户的关键区别。其次，这项工作将致力于开发新的分析方法，包括跨脑独立成分分析，用于识别共享社会模型的神经表征。