Regression Prediction of Perceptual Integration Capabilities and N270 Component Using Enhanced Electroencephalography Network

This study aims to investigate the brain electrical signals and perceptual integration abilities of pilots in low-visibility flight environments, focusing on the N270 event-related potential (ERP). The N270 occurs approximately 200 to 300 milliseconds after receiving visual stimuli and primarily app...

Full description

Saved in:
Bibliographic Details
Published in2024 IEEE 25th China Conference on System Simulation Technology and its Application (CCSSTA) pp. 182 - 187
Main Authors Chen, Chao, Ma, Yiran, Wang, Changming
Format Conference Proceeding
LanguageEnglish
Published IEEE 21.07.2024
Subjects
Brain modeling
deep learning
Enhanced Generative Encoder Network(EGEnet)
N270
Prediction algorithms
Process control
regression prediction
Representation learning
Resource management
Robustness
Stability analysis
Systems simulation
Training
Visualization
Online AccessGet full text
DOI10.1109/CCSSTA62096.2024.10691804

Cover

More Information
Summary:This study aims to investigate the brain electrical signals and perceptual integration abilities of pilots in low-visibility flight environments, focusing on the N270 event-related potential (ERP). The N270 occurs approximately 200 to 300 milliseconds after receiving visual stimuli and primarily appears in the parieto-occipital and occipital areas of the brain. It reflects the brain's activities in processing visual information, attention allocation, emotional processing, and cognitive control, particularly in response to visual stimuli such as faces and facial expressions. To assess and predict pilots' cognitive abilities under low-visibility conditions, this study designed a new deep learning algorithm called Enhanced Generative Encoder Network(EGEnet). Building on the existing EEGNet model, EGEnet incorporates residual modules, normalization layers, and dropout layers to enhance feature learning capabilities and model stability. Through automated feature extraction and regression prediction, EGEnet effectively evaluates pilots' perceptual integration abilities. Moreover, EGEnet demonstrates high accuracy and robustness in processing and reconstructing brain electrical data. This study provides a novel method for individual pilot assessment, promoting the development of intelligent and personalized training and technical support based on passive auditory ERP paradigms.
DOI:10.1109/CCSSTA62096.2024.10691804