腦性麻痺學童於腦機介面訓練前後 之認知控制與腦電特徵分析

NCU Institutional Repository > 生醫理工學院 > 認知與神經科學研究所 > 博碩士論文 > Item 987654321/98459

請使用永久網址來引用或連結此文件: https://ir.lib.ncu.edu.tw/handle/987654321/98459

題名:	腦性麻痺學童於腦機介面訓練前後之認知控制與腦電特徵分析
作者:	梁瑄;Liang, Francis Suean
貢獻者:	認知與神經科學研究所
關鍵詞:	腦性麻痺;腦機介面;動作想像;事件相關電位;時頻分析;認知控制;Cerebral Palsy (CP);Brain-Computer Interference (BCI);Motor Imagery (MI);Event Related Potential (ERP);Time-Frequency Analysis;Cognitive Control
日期:	2025-07-21
上傳時間:	2025-10-17 12:48:04 (UTC+8)
出版者:	國立中央大學
摘要:	本研究旨在探討腦性麻痺兒童接受八週動作想像腦機介面（brain–computer interface, BCI）認知訓練後，執行Go/NoGo任務時行為表現與神經指標的變化，以及兩者之間的關聯性。研究共納入18名兒童，於訓練前後分別完成Go/NoGo反應抑制任務，同步記錄反應時間、正確率及64通道腦電圖（electroencephalography, EEG）。透過獨立成分分析（ICA）與基線校正，去除動作與眼動雜訊後，針對N2與P3成分振幅與潛時，以及theta（3–7 Hz）、alpha（8–12 Hz）、beta（13–30 Hz）頻段的事件相關去同步化/同步化（ERD/ERS）進行分析。行為數據以成對樣本t檢定比較訓練前後差異，神經指標則以Pearson相關性與簡單迴歸模型檢視與行為改變的關聯。研究結果顯示，訓練後受試者反應時間有縮短趨勢，平均正確率則未達顯著改變。ERP分析顯示，NoGo條件下Pz通道的N2振幅顯著增強，而P3振幅多呈下降趨勢，可能反映早期衝突偵測活化增強及後期刺激評估的資源再分配。時頻分析則觀察到額葉及頂葉區域theta功率整體上升，尤其Pz通道的theta同步化與反應時間改善呈穩定負相關。進一步的迴歸分析中，theta功率變化對行為改變的解釋力最高，N2振幅則提供輔助性預測。然而，多數模型解釋力偏低，顯示在小樣本下神經行為關聯仍具高度變異性，應謹慎詮釋。本研究整合ERP與時頻指標，初步驗證BCI動作想像訓練對腦性麻痺兒童認知控制具潛在調節作用，並提供具量化潛力的神經指標（如N2振幅與theta同步化）作為早期效果評估工具。未來建議擴大樣本數、延長訓練期程，並結合源定位及腦網絡分析，以進一步驗證訓練對神經可塑性及執行功能的長期影響。;This study aimed to investigate the effects of an eight-week motor imagery brain–computer interface (BCI) training program on behavioral performance and neural indices in children with cerebral palsy during a Go/NoGo cognitive control task. Eighteen participants completed the Go/NoGo task before and after training, while reaction time, accuracy, and 64-channel electroencephalography (EEG) were recorded. EEG data underwent artifact rejection via independent component analysis (ICA), re-referencing, and baseline correction. Subsequent analyses focused on N2 and P3 amplitudes and latencies at frontal, central, and parietal sites, as well as event-related desynchronization/synchronization (ERD/ERS) within theta (3–7 Hz), alpha (8–12 Hz), and beta (13–30 Hz) frequency bands. Behavioral changes were assessed using paired-sample t-tests, and relationships between neural changes and behavioral improvements were examined via Pearson correlations and linear regression models. Following training, participants exhibited a trend toward faster reaction times without significant changes in overall accuracy. ERP results revealed enhanced N2 amplitudes over parietal channels during NoGo trials, alongside generally decreased P3 amplitudes, suggesting strengthened conflict monitoring processes and potentially more efficient stimulus evaluation. Time–frequency analyses indicated increased theta synchronization, particularly in parietal regions, with theta power changes demonstrating the strongest negative association with reaction time improvements. Regression models further supported the explanatory value of theta power as the most robust predictor of behavioral gains, while N2 amplitude provided supplementary predictive information. However, most models exhibited modest explanatory power, highlighting the variability inherent in small-sample studies. Overall, these findings provide preliminary evidence that BCI-based motor imagery training can modulate low-frequency neural oscillations and early conflict monitoring processes, potentially supporting cognitive control improvements in children with cerebral palsy. The results also underscore the promise of EEG-based neural indices—particularly N2 amplitude and theta synchronization—as sensitive markers of training-related neuroplasticity and behavioral change. Future research should involve larger cohorts, longer interventions, and source-level and connectivity analyses to strengthen causal inferences and model stability.
顯示於類別:	[認知與神經科學研究所 ] 博碩士論文

文件中的檔案:

檔案	描述	大小	格式	瀏覽次數
index.html		0Kb	HTML	6	檢視/開啟

在NCUIR中所有的資料項目都受到原著作權保護.

社群 sharing

資料載入中.....