Department of Psychology, University of Miami, Coral Gables, FL, USA.
Department of Psychology, University of Miami, Coral Gables, FL, USA.
Neuroimage. 2018 Aug 1;176:477-488. doi: 10.1016/j.neuroimage.2018.04.015. Epub 2018 Apr 11.
Analysis of task-based fMRI data is conventionally carried out using a hypothesis-driven approach, where blood-oxygen-level dependent (BOLD) time courses are correlated with a hypothesized temporal structure. In some experimental designs, this temporal structure can be difficult to define. In other cases, experimenters may wish to take a more exploratory, data-driven approach to detecting task-driven BOLD activity. In this study, we demonstrate the efficiency and power of an inter-subject synchronization approach for exploratory analysis of task-based fMRI data. Combining the tools of instantaneous phase synchronization and independent component analysis, we characterize whole-brain task-driven responses in terms of group-wise similarity in temporal signal dynamics of brain networks. We applied this framework to fMRI data collected during performance of a simple motor task and a social cognitive task. Analyses using an inter-subject phase synchronization approach revealed a large number of brain networks that dynamically synchronized to various features of the task, often not predicted by the hypothesized temporal structure of the task. We suggest that this methodological framework, along with readily available tools in the fMRI community, provides a powerful exploratory, data-driven approach for analysis of task-driven BOLD activity.
基于任务的功能磁共振成像（fMRI）数据的分析通常采用假设驱动的方法进行，其中血氧水平依赖（BOLD）时间序列与假设的时间结构相关联。在某些实验设计中，这种时间结构可能难以定义。在其他情况下，实验者可能希望采用更具探索性、数据驱动的方法来检测任务驱动的 BOLD 活动。在这项研究中，我们展示了一种用于基于任务的 fMRI 数据探索性分析的受试者间同步方法的效率和能力。我们结合瞬时相位同步和独立成分分析的工具，根据脑网络的时间信号动态的组间相似性来描述全脑任务驱动反应。我们将此框架应用于在执行简单运动任务和社会认知任务期间采集的 fMRI 数据。使用受试者间相位同步方法的分析揭示了大量大脑网络，这些网络动态地与任务的各种特征同步，这些特征通常是任务假设的时间结构所无法预测的。我们认为，这种方法框架以及 fMRI 社区中现成的工具，为分析任务驱动的 BOLD 活动提供了一种强大的、探索性的数据驱动方法。
