複雜廣度測驗(complex span task)常被用於測量工作記憶容量，此測驗為複合式測驗，其中包含了干擾子測驗以及記憶子測驗，其特點在於干擾子測驗的作答時間長度會隨著不同的受試者而改變，進而達到更準確測量出每個人工作記憶容量的目的，然而，在進行測驗當下的過程與電生理訊號相關性仍不清楚。先前關於複雜廣度測驗的電生理訊號研究並沒有在不同的工作記憶負荷(working memory-load)下發現差異，本研究認為有兩項可能的因素導致此結果，首先為過去的研究固定了每位受試者干擾子測驗呈現的時間，這可能造成該測驗沒辦法準確的測量出每個人的工作記憶容量。第二，過去的研究所使用的分析方法主要用來分析線性的數據，用以分析非線性的腦波訊號可能不是那麼合適。因此本研究主要探討在使用希爾伯特-黃轉換(Hilbert-Huang Transform, HHT)的總體經驗模態分解法(ensemble empirical mode decomposition, EEMD)分析複雜廣度測驗在腦電波訊號上是否與其他工作記憶測驗一樣在工作記憶負荷上呈現差異。我們假設隨著工作記憶負荷的增加，在腦波訊號中會有較小的P300以及在alpha與beta頻率能量強度會下降。此結果指出隨著工作記憶負荷的增加，整體的P300會減小，時間頻率訊號在alpha與beta頻率的強度亦會隨著工作記憶負荷的增加而下降。本研究指出在腦波訊號上複雜廣度測驗亦能表現出工作記憶負荷差異。基於此結論，在複雜廣度測驗下以希爾伯特-黃轉換進行腦電波訊號的分析可以因為更好得訊雜比以及效果量而得到更加的結果。;Complex span task is one among the commonly used cognitive tasks to evaluate individual working memory capacity (WMC). It is a dual task containing a distractor subtask and a memory subtask. The distractor subtask is equipped with varying time-limit across individual subjects which could provide a precise measure of individual WMC. However, the electrophysiological correlates of underlying processes of encoding and retrieval in working memory remain unclear. Previous complex span task study with EEG measure finds no significant difference between working memory-loads which a typical index is observed in other working memory tasks (e.g. n-back task and digital span task). There might be two potential problems that blur out the working memory-load difference. Firstly, the fixed-time distractor subtask may curtail in precisely assessing the individual WMC. Secondly, the method employed to analyze EEG data is favorable for linear systems, which may be inappropriate for a nonlinear system such as the human brain. Therefore, the present study seeks to investigate whether complex span tasks have similar or distinctive EEG patterns as other working memory tasks in terms of working memory-load by utilizing ensemble empirical mode decomposition (EEMD) of Hilbert-Huang Transform (HHT), a method for analyzing the nonlinear system. We expected an increase of working memory-load would lead to a decrement in the P300 component of event-related mode (ERM) and a decrease in the power of alpha and beta band frequency. This study found that low load condition had higher P300 amplitude than high load condition. The trend was widespread in many electrodes but only the C3 electrode reached statistical significance. In time-frequency analysis, a significant difference was observed between high and low load conditions at alpha and beta band in the frontal, central, and parietal channels. The results from our study demonstrate precise differences in EEG data pertaining to varied memory-load differences in the complex span task. Thus, by utilizing complex span tasks with the HHT-based analysis may aid in attaining a better signal to noise ratio and effect size for the results in working memory EEG studies.
