Towards a neurocomputational account of meta-control based on Bayesian inference in a context-specific hierarchy of time scales (A09)

在特定于上下文的时间尺度层次结构中基于贝叶斯推理的元控制的神经计算解释 (A09)

• 批准号: 318638301
• 金额: --
• 依托单位: Professur für Allgemeine Psychologie
• 依托单位国家: 德国
• 项目类别: Collaborative Research Centres
• 资助国家: 德国
• 项目状态: 未结题
• 来源: https://gepris.dfg.de/gepris/projekt/318638301?language=en
• 关键词: Towards; neurocomputational; account; meta; control

The computational mechanisms by which cognitive control processes are adapted to changing demands in dynamic and uncertain environments (e.g., to focus attention on a task versus to monitor the environment for unexpected changes) are insufficiently understood. Project A09 conceives of meta-control as a Bayesi-an inference problem and plans to develop a ‘hierarchical active inference’ model of meta-control and to test it in behavioral and fMRI studies using novel sequential decision-making tasks. The aim is to advance the theoretical framework of the CRC by developing a computational model of meta-control that provides a mechanistic account of context-sensitive control adjustments at different time-scales.

认知控制过程在动态和不确定环境中适应不断变化的需求的计算机制（例如，将注意力集中在一项任务上，而不是监测环境的意外变化）还没有得到充分的理解。A09项目将元控制视为一个贝叶斯推理问题，并计划开发一个元控制的“分层主动推理”模型，并使用新的顺序决策任务在行为和功能磁共振成像研究中对其进行测试。目的是通过开发一个元控制的计算模型来推进CRC的理论框架，该模型提供了不同时间尺度下上下文敏感控制调整的机制说明。