事件相關電位分析於精神分裂症研究之應用

Abstract

精神分裂症是一重大的慢性神經精神疾患。過去的研究結果，均指出此疾患涉及多層面的認知功能缺損，而這些認知功能缺損或可做為標記與界定精神分裂症易感受性基因的內在表現型。然而實際上在對這些認知功能進行量測時的信度常有問題：因為難以判定精神分裂症病患的表現不佳，是否真的不會，還是由於其負性症狀而導致缺乏動機（知道答案，但不想回答）。因此諸如失匹配負波(MMN)、P50/N100這類注意力前(pre-attentional)、不需個案配合測試(task-independent)的事件相關電位，近年來在訊息處理的研究中，已逐漸顯現出其重要性。 P50 與N100（通常由Cz 紀錄）；為使用paired-click paradigm 所進行的聽覺門控(sensory gating)量測；失匹配負波（其中以Fz 為最顯著）是由聽覺oddball paradigm所引發的負向事件相關電位成分波，所對映的是注意力前的聽覺記憶。 在本論文中，進行的四個事件相關電位研究，係依照三個維數層面(疾病狀態層面；事件相關電位與其他量測層面；事件相關電位分析方式層面) 來進行 1. 第二章係單獨使用傳統 P50 分析方法。過往的P50 研究經常強調S1 的角 色 (“gating in”) ，而本研究則從個案反應中找出S2 (“gating out”)的重要性。 2. 第三章則進行精神分裂症個案與對照組的失匹配負波與心理衡鑑，合併失匹配負波與心理衡鑑的多元模式，可以有較佳的正確診斷分類。 3. 第四章針對早期廣義的風險狀態(E-BARS) 起不同精神病風險程度的受試者，進行事件相關電位分析。我們發現：不論精神病風險程度如何，失匹配負波所反應出來的偏差檢測受損均已存在；相較之下P50和N100所反應的感覺門控(sensory gating)缺陷則隨著精神病風險程度而有差異。 4. 第五章利用時頻分析：發現相較於控制組個案，精神分裂症個案失匹配負波較差的主要原因，係由於其相位一致性(inter trial phase coherence)不佳。 總之，我們建立了一個包含防電磁隔音艙（硬體）與標準作業流程（人員）的事件相關電位研究室，也建立了與事件相關電位相關的研究相關網路(疾病狀態層面；事件相關電位與其他量測層面；事件相關電位分析方式層面)。事件相關電位是連結基因型（諸如基因與基因表現）跟表現型（諸如人類認知與疾病診斷）之間的橋樑，它能夠增加人類在腦與心智科學上的研究視野。

Schizophrenia is a major chronic neuropsychiatric disorder. Previous studies consistently revealed several cognitive deficits, which may serve as candidate endophenotype markers for identifying the genes for schizophrenia susceptibility. However, the reliability of some cognitive measurements had been criticized since the schizophrenia patients with negative symptoms usually did not have strong motivation to cooperate with such attention-dependent tests. Therefore, the role of pre-attentional, task-independent event-related potentials, such as Mismatch Negativity (MMN) and P50/N100, is important in information processing research. P50 and N100, usually collected from Cz electrodes, are auditory sensory gating measures using paired-click paradigm. MMN, usually collected from Fz electrodes, is a negative voltage component elicited in an auditory oddball paradigm and represents auditory sensory memory. In this dissertation, we proposed 4 different views of ERP research according to the three dimensions (Disorder dimension; ERP and other measure dimension; ERP analysis method dimension) : 1. Chapter 2 demonstrated a traditional P50 analysis method alone. Previous studies have placed more emphasis on S1's role in P50 sensory “gating in” while our study highlights the significant role played by the S2 response (“gating out”). 2. In Chapter 3, MMN was combined with several neuropsychological measurements to differentiate schizophrenia patients from healthy subjects. The multivariate approach combining MMN and neurocognition had a better classification utility. 3. Chapter 4 calculated the ERP in high-risk group (suspected pre-psychotic state, also called E-BARS, early/broad at-risk mental states) vs. control subjects. Impaired deviance detection shown by MMNs already exists in people at a pre-psychotic state regardless of clinical severity, while sensory gating deficits shown by P50/N100 varies depending on the risk levels in prodromal period. 4. Chapter 5 utilized time-frequency analysis for MMN and we found subjects with schizophrenia have lower inter trial phase coherence than controls. In conclusion, we built up an ERP laboratory with sound-attenuating and electrically shielded booth, standard operation procedure for ERP.We also set up ERP research network in three different dimensions. ERP is a bridge connecting genotype (gene, gene expression) and phenotype (human cognition and disorder). ERP actually made a lot of impact in brain and mind research.