Spectral Analysis of EEG Signals during Working Memory Tasks

Abstract

Assessment of the working memory performance is important to assist in the design of an effective pedagogical technique that suits the requirements of each learner. Most researchers used bulky equipment and an invasive technique to study working memory performance. The use of non-invasive electroencephalogram (EEG) to examine the working memory status directly from the brain is very limited. This paper describes the analysis of EEG signals during working memory tasks using power spectral density to identify the suitable frequency range for examining the working memory performance. The recorded EEG signals were filtered to remove the unwanted signals. Power spectral density was computed on four frequency band signals to identify the signal strength at each location on the scalp and to examine the brain active regions. Results from the analysis showed that the low-frequency theta could reveal the working memory status for all activities. The findings in this work can be used to develop an EEG-based Working Memory Assessment system that can assist educators to strategise in helping learners to succeed academically.