Open Access Repository
Predicting individual decision-making responses based on single-trial EEG

Full text not available from this repository.
Abstract
Decision-making plays an essential role in the interpersonal interactions and cognitive processing of individuals. There has been increasing interest in being able to predict an individual’s decision-making response (i.e., acceptance or rejection). We proposed an electroencephalogram (EEG)-based computational intelligence framework to predict individual responses. Specifically, the discriminative spatial network pattern (DSNP), a supervised learning approach, was applied to single-trial EEG data to extract the DSNP feature from the single-trial brain network. A linear discriminate analysis (LDA) trained on the DSNP features was then used to predict the individual response trial-by-trial. To verify the performance of the proposed DSNP, we recruited two independent subject groups, and recorded the EEGs using two types of EEG systems. The performances of the trial-by-trial predictors achieved an accuracy of 0.88 ± 0.09 for the first dataset, and 0.90 ± 0.10 for the second dataset. These trial-by-trial prediction performances suggested that individual responses could be predicted trial-by-trial by using the specific pattern of single-trial EEG networks, and our proposed method has the potential to establish the biologically inspired artificial intelligence decision system.
Item Type: | Article |
---|---|
Authors/Creators: | Si, Y and Li, Fali and Duan, K and Tao, Q and Li, C and Cao, Z and Zhang, Yangsong and Biswal, B and Li, P and Yao, D and Xu, P |
Keywords: | EEG, decision making, electroencephalogram, discriminative spatial network pattern, brain network, single-trial prediction |
Journal or Publication Title: | Neuroimage |
Publisher: | Academic Press Inc Elsevier Science |
ISSN: | 1053-8119 |
DOI / ID Number: | 10.1016/j.neuroimage.2019.116333 |
Copyright Information: | Copyright 2019 Published by Elsevier Inc. |
Related URLs: | |
Item Statistics: | View statistics for this item |
Actions (login required)
![]() |
Item Control Page |