Biosensor assisted measurement of cognitive participation in English reading: A psychological perspective

Cognitive involvement in English reading is important for understanding and engagement. Traditional techniques of measuring cognitive participation frequently rely on individual evaluations that do not capture real-time physiological reactions. Recent developments in artificial intelligence (AI) and biosensor technology provide intriguing options to close this gap by offering a goal, real-time data. This work seeks to improve the evaluation of cognitive participation in reading in English by combining biosensor data analysis with modern AI algorithms. Participants completed English reading activities, and their electroencephalogram (EEG) and Galvanic Skin Response (GSR) signals were recorded. A median filter was used as a pre-processing to reduce noise. Discrete wavelet transform (DWT) was utilized to extract features to extract specific patterns from the biosensor signals. The new Dynamic White Shark Infused Residual Neural Network (DWS-IResNet) approach was used to model and forecast the level of cognitive participation. The proposed method is implemented using the Python platform. The algorithm used was trained and evaluated based on performance indicators such as accuracy. Using the features of simple, technical, analytical, and emotional, the proposed DWS-IResNet approach is compared with metrics between males and females. In simple features, the accuracy was 90% higher for females; in emotional features, the precision was 90% better for females; and in males and females, the percentage of emotion features was greater at 90% of recall and 90% of F1-score. The suggested technique demonstrates the efficacy of the AI-enhanced biosensor method for assessing cognitive engagement. This study demonstrates the possibility of AI-powered biosensor readings for real-time, accurate evaluation of cognitive involvement during reading.