Real-time mental workload detector for estimating human performance under workload

Abstract

Brain is often studied for behavioral reasons highlighting various impacts of tasks that intersperse awareness, attention, problem solving, decision-making, etc. It has been a topic of fascination for psychologists. In this work, participant has been asked to perform some primary visual task that is based on cognition process and simultaneously asked to follow some secondary auditory instructions with some predefined relaxation spans in between. The neural response to these stimuli is recorded in the form of EEG signals with the help of RMS EEG-32 Super Spec machine. The EEG frequency bands are being studied for analysis of human behavior. A classification model is designed with the help of support vector machine to test the cognitive processes of human mind, while there are changes in resource allotment in terms of primary and secondary workload. Such a model can further be adopted to depict the type of cognition a participant undergoes. Also, the decrease in efficiency due to involvement in secondary tasks can be alarmed. This alarm may help in avoiding many disasters at places where attention toward a particular primary task is very crucial.