Motivation detection using EEG signal analysis by residual-in-residual convolutional neural network

While we know that motivated students learn better than non-motivated but detecting motivation is challenging. Here present a game-based detection approach from the EEG signals. We take an original of using EEG-based brain computer interface to assess if state manifest in physiological signals as well, and what are suitable conditions order achieve goal? To best our knowledge, level proposed for first time this paper. In resolve central obstacle small datasets containing deep features, propose novel unique ‘residual-in-residual architecture convolutional neural network (RRCNN)’ capable reducing problem over-fitting on vanishing gradient. Having accomplished this, several aspects considered, including channel selection accuracy obtained alpha or beta waves also include detailed validation different methodology, comparison with other works relevant. Our achieves 89% detect while learning, where wave frontal asymmetry channels employed. A more robust (less sensitive learning conditions) 88% achieved channels. The results clearly indicate potential states signals, provided methodologies such paper, • Detection proposed. Novel residual-in-residual designed. Proposed reduces overfitting gradient even dataset.