The current study is aimed at establishing links between brain network examination and neural plasticity studies measured by optical neuroimaging. Sixteen healthy subjects were recruited from the University of Macau to test the Granger Prediction Estimation (GPE) method to investigate brain network connectivity during figurative language comprehension. The method is aimed at mapping significant causal relationships across language brain networks, captured by functional near-infrared spectroscopy measurements (fNIRS): (i) definition of regions of interest (ROIs) based on significant channels extracted from spatial activation maps; (ii) inspection of significant causal relationships in temporal resolution, exploring the experimental task agreement; and (iii) early identification of stronger causal relationships that guide neuromodulation intervention, targeting impaired connectivity pathways. Our results propose top-down mechanisms responsible for perceptive-attention engagement in the left anterior frontal cortex and bottom-up mechanism in the right hemispheres during the semantic integration of figurative language. Moreover, the interhemispheric directional flow suggests a right hemisphere engagement in decoding unfamiliar literal sentences and fine-grained integration guided by the left hemisphere to reduce ambiguity in meaningless words. Finally, bottom-up mechanisms seem activated by logographic-semantic processing in literal meanings and memory storage centres in meaningless comprehension. To sum up, our main findings reveal that the Granger Prediction Estimation (GPE) integrated strategy proposes an effective link between assessment and intervention, capable of enhancing the efficiency of the treatment in language disorders and reducing the neuromodulation side effects.

中文翻译：

---

利用光学神经影像指导比喻语言理解中的神经调节干预的有效连接性研究


目前的研究旨在建立大脑网络检查和通过光学神经成像测量的神经可塑性研究之间的联系。从澳门大学招募了 16 名健康受试者来测试格兰杰预测估计（GPE）方法，以研究比喻语言理解过程中的大脑网络连接。该方法旨在绘制通过功能近红外光谱测量（fNIRS）捕获的跨语言大脑网络的重要因果关系：（i）基于从空间激活图提取的重要通道定义感兴趣区域（ROI）； (ii) 检查时间分辨率中的重要因果关系，探索实验任务一致性； (iii) 及早识别更强的因果关系，指导神经调节干预，针对受损的连接通路。我们的结果提出了在比喻语言的语义整合过程中负责左前额叶皮层感知注意参与的自上而下机制和右半球的自下而上机制。此外，半球间的定向流表明右半球参与解码不熟悉的字面句子，并在左半球的指导下进行细粒度整合，以减少无意义单词的歧义。最后，自下而上的机制似乎是由字面意义中的语标语义处理和无意义理解中的记忆存储中心激活的。 综上所述，我们的主要研究结果表明，格兰杰预测估计（GPE）综合策略提出了评估和干预之间的有效联系，能够提高语言障碍的治疗效率并减少神经调节副作用。