Wang Zhidong, Sun Tie, Xiao Feng
State Key Laboratory of Cognitive Neuroscience and Learning, Mc/Govern Institute for Brain Research, Beijing Normal University, Beijing, China.
Department of Education Science, Innovation Center for Fundamental Education Quality Enhancement of Shanxi Province, Shanxi Normal University, Taiyuan, Shanxi, China.
Brain Topogr. 2025 Jan 22;38(2):24. doi: 10.1007/s10548-024-01099-3.
Relational integration is a key subcomponent of working memory and a strong predictor of fluid intelligence. Both relational integration and fluid intelligence share a common neural foundation, particularly involving the frontoparietal network. This study utilized a randomized controlled experiment to examine the effect of relational integration training on brain networks using electroencephalogram (EEG) and microstate analysis. Participants were randomly assigned to either a relational integration training group (n = 29) or an active control group (n = 28) for one month. The Sandia matrices task assessed fluid intelligence, while rest-EEG was recorded during pre- and post-tests. Microstate analysis revealed that, for microstate D, the training group demonstrated a significant increase in occurrence and contribution following the intervention compared to the control group. Additionally, microstate D occurrence was negatively correlated with reaction times (RTs). Post-training, the training group showed a lower occurrence and contribution of microstate C compared to the control group. Regarding transfer probability, the training group exhibited a decrease between microstates A and B, and an increase between microstates C and D. In contrast, the control group showed increased transfer probability between microstates A, B, and C, and a decrease between microstate D and other microstates (B and A). These findings indicate that relational integration training influences frontoparietal networks associated with fluid intelligence. The current study suggests that relational integration training is an effective intervention for enhancing fluid intelligence.
关系整合是工作记忆的一个关键子成分，也是流体智力的一个有力预测指标。关系整合和流体智力都有共同的神经基础，尤其涉及额顶叶网络。本研究采用随机对照实验，利用脑电图（EEG）和微状态分析来检验关系整合训练对脑网络的影响。参与者被随机分配到关系整合训练组（n = 29）或积极对照组（n = 28），为期一个月。桑迪亚矩阵任务评估流体智力，同时在测试前和测试后记录静息EEG。微状态分析显示，对于微状态D，与对照组相比，训练组在干预后出现次数和贡献率显著增加。此外，微状态D的出现次数与反应时间（RTs）呈负相关。训练后，与对照组相比，训练组微状态C的出现次数和贡献率较低。关于转移概率，训练组在微状态A和B之间呈现下降，在微状态C和D之间呈现上升。相比之下，对照组在微状态A、B和C之间的转移概率增加，在微状态D与其他微状态（B和A）之间的转移概率下降。这些发现表明，关系整合训练会影响与流体智力相关的额顶叶网络。当前研究表明，关系整合训练是增强流体智力的一种有效干预措施。
