BACKGROUND Functional dysconnection in schizophrenia is underwritten by a pathophysiology of the glutamate neurotransmission that affects the excitation-inhibition balance in key nodes of the salience network. Physiologically, this manifests as aberrant effective connectivity in intrinsic connections involving inhibitory interneurons. In computational terms, this produces a pathology of evidence accumulation and ensuing inference in the brain. Finally, the pathophysiology and aberrant inference would partially account for the psychopathology of schizophrenia as measured in terms of symptoms and signs. We refer to this formulation as the 3-level hypothesis. METHODS We tested the hypothesis in core nodes of the salience network (the dorsal anterior cingulate cortex [dACC] and the anterior insula) of 20 patients with first-episode psychosis and 20 healthy control subjects. We established 3-way correlations between the magnetic resonance spectroscopy measures of glutamate, effective connectivity of resting-state functional magnetic resonance imaging, and correlations between measures of this connectivity and estimates of precision (inherent in evidence accumulation in the Stroop task) and psychopathology. RESULTS Glutamate concentration in the dACC was associated with higher and lower inhibitory connectivity in the dACC and in the anterior insula, respectively. Crucially, glutamate concentration correlated negatively with the inhibitory influence on the excitatory neuronal population in the dACC of subjects with first-episode psychosis. Furthermore, aberrant computational parameters of the Stroop task performance were associated with aberrant inhibitory connections. Finally, the strength of connections from the dACC to the anterior insula correlated negatively with severity of social withdrawal. CONCLUSIONS These findings support a link between glutamate-mediated cortical disinhibition, effective-connectivity deficits, and computational performance in psychosis.

中文翻译：

---

显着网络中的谷氨酸盐和连接障碍：精神分裂症的神经化学、有效连接和计算证据

背景技术精神分裂症中的功能性连接障碍由影响显着网络关键节点中的兴奋-抑制平衡的谷氨酸神经传递的病理生理学承担。在生理上，这表现为涉及抑制性中间神经元的内在连接中的异常有效连接。在计算方面，这在大脑中产生了证据积累和随之而来的推理的病理学。最后，病理生理学和异常推论将部分解释以症状和体征衡量的精神分裂症的精神病理学。我们将此公式称为 3 级假设。方法 我们在 20 名首发精神病患者和 20 名健康对照受试者的显着性网络核心节点（背侧前扣带回皮层 [dACC] 和前岛叶）中测试了假设。我们建立了谷氨酸的磁共振波谱测量、静息状态功能磁共振成像的有效连接性以及这种连接性的测量与精度估计（Stroop 任务中固有的证据积累）和精神病理学之间的相关性之间的 3 向相关性。结果 dACC 中的谷氨酸浓度分别与 dACC 和前岛叶中较高和较低的抑制连接相关。至关重要的是，谷氨酸浓度与对首发精神病受试者 dACC 中兴奋性神经元群的抑制影响呈负相关。此外，Stroop 任务性能的异常计算参数与异常抑制连接相关。最后，从 dACC 到前脑岛的连接强度与社交退缩的严重程度呈负相关。结论这些发现支持谷氨酸介导的皮质去抑制、有效连接缺陷和精神病的计算性能之间存在联系。从 dACC 到前脑岛的连接强度与社交退缩的严重程度呈负相关。结论这些发现支持谷氨酸介导的皮质去抑制、有效连接缺陷和精神病的计算性能之间存在联系。从 dACC 到前脑岛的连接强度与社交退缩的严重程度呈负相关。结论这些发现支持谷氨酸介导的皮质去抑制、有效连接缺陷和精神病的计算性能之间存在联系。