Mental Workload Estimation Using Wireless EEG Signals

Abstract

Previous studies have shown that electroencephalogram (EEG) can be used in estimating mental workload. However, developing fast and reliable models for cross-task, cross-subject and cross-session classifications of workload remains a challenge. In this study, a wireless Emotiv EPOC headset was used to evaluate workload in two different mental tasks: n-back task and mental arithmetic task. 0-back task and 2-back task were employed as low and high workload in the n-back task while 1-digit and 3-digit addition were used as the two different workload levels in the arithmetic task. Using power spectral density as features, a fast signal processing and feature extraction framework was developed to facilitate realtime estimation of workload. Within-session accuracies of 98.5% and 95.5% were achieved in the n-back and arithmetic tasks respectively. Adaptive subspace feature matching (ASFM) was applied for cross-session, cross-task and cross-subject classifications. The feature adaptation provided average crosssession accuracies of 80.5% and 74.4% in the n-back and the arithmetic tasks respectively. An average cross-task accuracy of 68.6% was achieved while cross-subject accuracies were 74.4% and 64.1% in the n-back and arithmetic tasks respectively. The framework generalised well across subjects and tasks, and it provided a promising approach towards developing subject and task-independent models. This study also shows that a consumerlevel wireless EEG headset can be applied in cognitive monitoring for real-time estimation of workload in practice.

Multimodal Study of the Effects of Varying Task Load Utilizing EEG, GSR and Eye-Tracking

“…• Most responsive electrode: FC6. Our results are similar to those reported in a recent study by Adewale and Panoutsos (2019) which reported a power increase with workload in frontal regions. In their study, the increase was highest in FC6, next to AF4 as reported here.…”

“…Further, Adewale and Panoutsos (2019) found that this region was also most responsive in the mid and high beta band, the same bands as in this study (see Table 5). The corroboration with the results reported by Adewale and Panoutsos (2019) suggests that the mid-and high-beta bands responses from FC6 require further investigation in task load experiments. • Most responsive frequency band: Low Beta (12.5-18 Hz), followed by higher frequency bands.…”

Cross-task-oriented EEG signal analysis methods: Our opinion

“…The results showed that common components of cross-task EEG signals could be identified through this method. Furthermore, Adewale et al designed a signal processing and feature extraction framework based on PSD (Adewale and Panoutsos, 2019 ), and found that PSD could be used as an excellent feature for mental workload estimation. Therefore, PSD features show excellent performance in cross-task EEG signal analysis.…”

“…Research on cross-task EEG signals analysis methods has become a fast-growing research hotspot. In recent years, more and more researchers applied the features, which were widely used in EEG signal analysis to cross-task EEG signal analysis studies, including power spectral density (PSD) features (Touryan et al, 2016 ; Adewale and Panoutsos, 2019 ), fusion features (Kakkos et al, 2021 ), etc. The objective aimed to find ways to effectively deal with the differences between tasks.…”