Journal Article

Differential Sequence Analysis of EEG Brain Signals for Emotional and Cognitive Assessment

Chaudhuri, Sheli Sinha, Chowdhuri, Swati, Paul, Trisha

2026

Overview

To improve mental health and wellness and create specific solutions, it is essential to comprehend how individuals feel and brain functions. In this study, we present a novel approach for emotion recognition and analysing electroencephalography (EEG) data for cognitive evaluation. EEG data were collected from 30 participants using non-invasive electrodes positioned at AF3, AF4, T7, T8, and Pz, corresponding to the frontal, temporal, and parietal lobes.We have obtained real-time EEG data from participantes during various tasks, including as rest, listening to music, answering questions, and completing mathematical puzzles. Our goal was to investigate the brain correlates of different emotional and cognitive states. The recorded signals were pre-processed using a 4–8 Hz bandpass filter targeting theta waves, followed by Fast Fourier Transform (FFT) and sequence pattern mapping. Statistical significance of variations between brain states was confirmed using ANOVA (p < 0.05). A supervised machine learning classifier (Random Forest) achieved 89.2% prediction accuracy, with precision = 0.87, recall = 0.90, and F1-score = 0.885, demonstrating robust differentiation between emotional and cognitive states. We have developed prediction models for emotion recognition and cognitive assessment using linear regression classification based on EEG features extracted from multiple brain areas. Using statistical analysis and graphical representation techniques, the EEG data was visualised and analysed, revealing a variety of patterns associated with different tasks and stimuli. Our study demonstrates that emotional states and cognitive activity may be accurately identified from EEG signals. More specifically, we observed significant differences in EEG patterns between tasks, suggesting that real-time tracking of human emotions and mental processes can be achieved with EEG-based techniques. Applications in human-computer interaction, mental health monitoring, and tailored interventions to improve well-being are possible with the suggested methodology.