"High risk non-initiating insider" identification based on EEG analysis for enhancing nuclear security. (March 2018)

• Record Type: Journal Article
• Authors: Suh, Young-A; Yim, Man-Sung
• Is Part Of: Annals of nuclear energy. Volume 113(2018)
• Page Start: 308
• Page End: 318
• Publication Date: 2018-03
• Subjects: Insider threat identification -- EEG signals -- Nuclear security; Nuclear energy -- Periodicals; Nuclear engineering -- Periodicals; 621.4805
• DOI: 10.1016/j.anucene.2017.11.030
• Languages: English
• ISSNs: 0306-4549

Abstract

Highlights: The work studied an approach to identify a non-initiating insider using EEG signals. Relative power and band-to-band power ratios of EEG signals were useful indicators. Thinking and decision making toward becoming an insider evoke different bio-responses. Abstract: The objective of this research is to describe the possibility of identifying a non-initiating insider (insider who has not yet committed an action) using an electroencephalography (EEG) signals analysis for application in nuclear facilities. To support the study, a quantitative EEG analysis (qEEG) was performed by using a wearable device on 11 healthy subjects. Possible indicators from the qEEG records were examined to identify a non-initiating insider. Such indicators included the relative power of different frequency bands of brain waves (i.e., alpha, beta, theta, gamma and the sum of alpha and theta ) and the band-to-band power ratio of beta to alpha and gamma to alpha waves. Results showed that the thinking process of the subjects toward a good (e.g., following the moral compass) versus a bad decision (e.g., cooperation with outsiders) evoked significantly different responses in the relative powers of different frequency band brain waves as well as in the β/α and γ/α band-to-band power ratios. Similar differences were also found for all but the theta band EEG indicators when the subjects committed to a decision related to becoming an insider. The variations in the β/α and γ/α band-to-band power ratios were also found to identify the differences in the level of difficulties involved in the tasks related to becoming an insider. The results indicated that it may be possible to identify a high risk insider based on EEG monitoring using a wearable device.