An Attempt to Evaluate Mental Workload Using Wavelet Transform of EEG

Abstract

An attempt was made to evaluate mental workload using a wavelet transform of electroencephalographic (EEG) signals. Participants performed a continuous matching task at three levels of task difficulty. EEG signals during the task were recorded continuously from Fz, Cz, and Pz. The reaction time increased as the difficulty of the task increased. The percentage correct decreased as the task became more difficult. In accordance with this, the rating score on the NASA-Task Load Index tended to increase with increased task difficulty. The EEG signals were analyzed using wavelet transform to investigate time-frequency characteristics. The total power at theta, alpha, and beta frequency bands and the time that the maximum power appeared for the three frequency bands were extracted from the scalogram. Increasing cognitive task difficulty seems to delay the time at which the central nervous system works most actively. These measures were found to be sensitive indicators of mental workload and could differentiate three cognitive task loads (low, moderate, and high) with high precision. Actual or potential applications of this research include a method that is relatively quick and accurate, compared with traditional methods, for the evaluation of mental workload.