運用腦波特徵於持續專注力評估與恢復方法之研究

The Study of Sustained Attention and Recovery Methods based on EEG

劉良驥(Liang-Chi Liu)

指導教授 ： 姜琇森

國立臺中科技大學/資訊與流通學院/資訊管理系碩士班/碩士(2014年)

https://doi.org/10.6826/NUTC.2014.00103

摘要

隨著資訊科技的進步與競爭環境快速變化，人們面臨新型態的挑戰，環境變化、提升學習效率與工作成效等都需要高度集中且持續的專注力。而行動網路的快速發展，帶來更多不同的應用與便利，無形中分散學生學習、工作的專注力，往往帶來諸多不良的效果。
腦電圖(Electroencephalogram, EEG)是一項重要的生理參數，記錄大腦活動時的電波變化，透過腦波特徵的量測可即時反應專注力的程度，腦電圖已被運用在專注力的研究，著重在專注力的類別(持續專注、選擇專注)的量測，已被證實是相當快速且客觀的專注力評估工具。然而，腦電波圖(EEG)屬於較微弱的信號，除不易辨識外，也容易受外界干擾或雜訊影響而失真，因此，如何擷取與識別腦部發出的生理訊號是提升專注力評估準確率的重要課題。
奇異值解析是一種能有效將訊號去除外部雜訊，保留原本訊號不失真，傅立葉轉換透過線性積分轉換的方式，讓腦波訊號在時域與頻域相互轉換，並可從頻域中擷取與專注力有關的腦波特徵頻段。最小熵原理法可從腦波特徵資訊中，建立出最有關專注力的模糊隸屬函數，關聯派翠網路圖形化的呈現方式，並以關聯函數運算與Apriori演算法作為基礎計算腦波特徵參數與專注力的連結性與相關程度，能呈現與提供腦波特徵變化與不同專注力程度的關係。本研究欲結合奇異值解析、傅立葉轉換、最小熵原理法與關聯派翠網路等技術，發展兼具專注力評估與提升的模式，透過分析腦電圖擷取的腦波生理參數，精確量測專注力程度。

關鍵字

腦電圖 ； 持續專注力 ； 奇異值解析 ； 最小熵原理 ； 關聯派翠網路

並列摘要

Nowadays, technology improvements and competition environment make people need to enhance both learning and work efficiency. Moreover, how to increase sustained attention periods are people face new challenges.
In recent years, the smartphone not only make our life more convenient but also to distraction many student attention on study. However, if people can’t be attention for long period it will leads to cannot focused, memory loss, lack of understanding and lack of patience. In cognitive psychology, more recent set of research point out the people attention with learning, memory and resolution have a significant positive relationship. In past study, confirms that when people learning, reminders and real time feedback will help people back to attention status and improve learning attention. In addition, correct recovery method and enhance attention will help people to avoid lax of attention, lack of patience and distraction condition. Therefore, an accurate to measure attention level and find out appropriate recovery method will be major topic.
In this study, we developed a model to assessment sustained attention level and find out associated between Electroencephalography and sustained attention. According to attention level, we recommend recovery method to help user enhance attention.

並列關鍵字

Sustained Attention ； Attention Level ； Electroencephalography ； Recovery Method

參考文獻

孟瑛如, & 簡吟文. (2009). 注意力訓練對國小學習障礙學生在課堂學習時注意力影響之探討. 特殊教育與復健學報, 20, 25-52.

湯雅雯. (2005). 腦波量測系統之研製與腦波信號之非線性分析. 成功大學電機工程學系學位論文, 1-70.

Shih D. H., Chiang H. S. & Lin B. (2007) A Generalized Associative Petri Net for Reasoning. IEEE Transactions on Knowledge and Data Engineering, 19(9), 1241-1251.

Christensen, R. (1980). Entropy minimax sourcebook. Lincoln, MA:Entropy Limited.

Allan Cheyne, J., Solman, G. J. F., Carriere, J. S. A., & Smilek, D. (2009). Anatomy of an error: a bidirectional state model of task engagement/disengagement and attention-related errors. Cognition, 111(1), 98-113.