An Online Learning and Problem Solving (OLPS) EEG Database for Mental Workload Assessment and Its Initial Benchmark Classification Performance

Mental Workload (MWL) assessment is very necessary in the scenario of online learning. In the case of online learning, many students were unable to understand the topics. Online learning also increases the MWL of students as compared to offline learning. This increase in MWL may cause fear, anxiety, sadness, etc., therefore the need for MWL assessment methods and tools is very important. In this study, an experiment has been designed to mimic online learning and understating scenarios. During the experiment MWL is generated as the cause of online learning through video lectures and solving related problems, simultaneously electroencephalogram (EEG) signals are recorded. EEG signals are used to estimate the MWL. A simple classification model with the help of a one-dimensional convolutional neural network (1D-CNN) is also trained. The trained model is able to classify MWL into low and high classes. The main aim of this study is to provide a new database named as “Online Learning and Problem Solving (OLPS) EEG Database” to explore MWL analysis and assessment methods and models.