Affiliations: [a] Computer Engineering Department, School of Engineering, Shiraz branch, Islamic Azad University, Shiraz, Iran | [b] CSE & IT Department, Electrical and Computer Engineering Faculty, Shiraz University, Shiraz, Iran | [c] Department of Physiology, Shiraz University of Medical Sciences, Shiraz, Iran
Abstract: Accurate differentiation of mental fatigue levels by analyzing electroencephalogram (EEG) features is still a challenge. This deficiency is rooted in the inability of conventional EEG features to reveal significant changes amongst fatigue levels. From a different perspective, it is evident for evaluating the amount of alertness that characterizing the P300 component is widely used by performing a recognition task. The goal of this study is to classify the pre-task and post-task fatigue levels by tracking the spatial activation of their P300 sources compared to differentiating the P300 waveform features. To track these sources, P300 was elicited from the background EEG at all channels using the conventional time-locked synchronous grand averaging over all time frames and subjects. Next, standardized low resolution electromagnetic tomography (sLORETA) and shrinking sLORETA were both applied to the elicited P300 of all channels in order to estimate the activity of P300 sources. Herein, thirty healthy subjects participated and their EEG signals were recorded by thirty channels through the pre-task (30 minutes), task (60–90 minutes) and post-task (30 minutes) states. At each recording phase, an equal number of audio and visual stimuli were applied to the participants who were performing both audio and visual recognition tasks. Empirical results show a significant decrease in the activation of P300 sources in the post-task mental fatigue level compared to the pre-task over tempo-parieto-occipital areas (secondary association area). It is interesting that in most channels, no significant change in the amplitude/latency of P300 is observed between the two fatigue levels.
