Gholam Tamimi Maryam, Daliri Mohammad Reza
Neuroscience and Neuroengineering Research Laboratory, Biomedical Engineering Department, School of Electrical Engineering, Iran University of Science and Technology (IUST), Iran, 16846-13114, Tehran.
Brain Imaging Behav. 2025 Sep 24. doi: 10.1007/s11682-025-01059-w.
Emotion is present in all aspects of human life and serves as a crucial foundation for communication and interaction. Emotional processing (EP) is a complex phenomenon involving dynamic interactions among various brain regions. Despite significant progress in EP research, important challenges remain-particularly in understanding the temporal dynamics of emotion. In this study, we investigated alterations in dynamic functional connectivity (dFC) patterns during an emotional processing task, using fMRI data from 100 healthy participants in the Human Connectome Project (HCP). The brain was parcellated into 90 regions of interest (ROIs) and grouped into six networks and ten well-known brain regions using the AAL atlas. We applied dFC analysis based on sliding window correlation (SWC) and k-means clustering to identify discrete connectivity states. To define the optimum number of states, we employed non-supervised validity criteria silhouette measure. Additionally, we estimated mean dwell times and transition probability matrices between states in both face and shape conditions using a hidden Markov model (HMM). Within these states, we observed state-dependent alterations in within and between regional connectivity between the face and shape conditions. Our findings revealed three distinct dFC states and among them, dFC state with the most significant differences in probability of transitions included brain regions involved in, frontoparietal, limbic and visual networks. Across all three states, several key bilateral regions exhibited significant changes in dFC, involved in limbic (amygdala, hippocampus, parahippocampal and rectus), default mode (anterior cingulate gyrus, median cingulate gyrus, posterior cingulate gyrus and angular), frontoparietal (inferior parietal gyrus, superior parietal gyrus, and middle frontal gyrus), visual (inferior occipital gyrus, fusiform, cuneus, precuneus, lingual and calcarine), temporal-parietal (paracentral lobule, precentral, postcentral, superior temporal gyrus, temporal pole superior and insula), and subcortical (caudate, putamen, pallidum and thalamus) networks. Also, we identified three dFC states between ten brain regions -frontal-central-parietal, frontal-temporal-occipital, and global state.
情感存在于人类生活的方方面面，是沟通与互动的重要基础。情感加工（EP）是一种复杂的现象，涉及多个脑区之间的动态相互作用。尽管EP研究取得了显著进展，但仍存在重要挑战，尤其是在理解情感的时间动态方面。在本研究中，我们使用人类连接组计划（HCP）中100名健康参与者的功能磁共振成像（fMRI）数据，研究了情感加工任务期间动态功能连接（dFC）模式的变化。大脑被划分为90个感兴趣区域（ROI），并使用AAL图谱分为六个网络和十个著名的脑区。我们应用基于滑动窗口相关性（SWC）和k均值聚类的dFC分析来识别离散的连接状态。为了定义最佳状态数，我们采用了无监督有效性标准轮廓测量。此外，我们使用隐马尔可夫模型（HMM）估计了面部和形状条件下状态之间的平均停留时间和转移概率矩阵。在这些状态中，我们观察到面部和形状条件下区域内和区域间连接的状态依赖性变化。我们的研究结果揭示了三种不同的dFC状态，其中转移概率差异最显著的dFC状态包括涉及额顶叶、边缘和视觉网络的脑区。在所有三种状态下，几个关键的双侧区域在dFC上表现出显著变化，涉及边缘（杏仁核、海马体、海马旁回和直肌）、默认模式（前扣带回、中扣带回、后扣带回和角回）、额顶叶（顶下小叶、顶上小叶和额中回）、视觉（枕下回、梭状回、楔叶、楔前叶、舌回和距状回）、颞顶叶（中央旁小叶、中央前回、中央后回、颞上回、颞极上回和岛叶）和皮质下（尾状核、壳核、苍白球和丘脑）网络。此外，我们在十个脑区之间识别出三种dFC状态——额中央顶叶、额颞枕叶和全局状态。
