Dept. of Artificial Intelligence, University of Groningen, AK Groningen, The Netherlands.
Neuroimage. 2011 Sep 1;58(1):137-47. doi: 10.1016/j.neuroimage.2011.05.084. Epub 2011 Jun 16.
In this paper, a model-based analysis method for fMRI is used with a high-level symbolic process model. Participants performed a triple-task in which intermediate task information needs to be updated frequently. Previous work has shown that the associated resource - the problem state resource - acts as a bottleneck in multitasking. The model-based method was used to locate the neural correlates of 'problem state replacements'. To analyze the fMRI data, we fit the computational process model to the behavioral data and regressed the model's activity against the fMRI data. The brain region responsible for the temporary representation of problem states, the inferior parietal lobule, and the brain region responsible for long-term storage of problem states, the inferior frontal gyrus were thus identified. These results show that model-based fMRI analyses can be performed using high-level symbolic cognitive models, enabling fine-grained exploratory fMRI research.
本文采用基于模型的分析方法，结合高级符号过程模型，对 fMRI 进行分析。参与者执行了一项三重任务，其中需要频繁更新中间任务信息。先前的研究表明，相关资源——问题状态资源——在多任务处理中起着瓶颈作用。基于模型的方法用于定位“问题状态替换”的神经相关物。为了分析 fMRI 数据，我们将计算过程模型拟合到行为数据中，并将模型的活动与 fMRI 数据进行回归。因此，确定了负责问题状态临时表示的脑区——下顶叶，以及负责问题状态长期存储的脑区——下额叶。这些结果表明，可以使用高级符号认知模型对基于模型的 fMRI 分析进行操作，从而实现精细的探索性 fMRI 研究。
