Performance analysis of memory recollection and problem-solving cognitive processes using EEG signals

摘要

The human brain has the ability to perform high-level multitasking with great efficiency. Cognition is one such domain of brain multitasking which includes attention, memorisation, recollection, decision-making, and problem-solving. In this study, we aim to study the aspects of brain involving memorisation, problem-solving and recollection. The brain signals are recorded in the form of electroencephalography (EEG) for the three mental tasks. Next, the work is approached as a simple pattern recognition problem. Power spectral density features corresponding to the three mental tasks are extracted from the EEG signals. These features are then fed as inputs to the AdaBoost (ensemble) classifier to distinguish among the different mental states. The accuracy of the three mental tasks, thus obtained, is more than 80%. Further, results show that temporal region, frontal region and parietal region are most active for memorisation, problem-solving and recollection, respectively.