EEG Hyperscanning Techniques for Assessment of Mutual Engagement
Sharif, Ahmad (2024)
2024
Tietojenkäsittelyopin maisteriohjelma - Master's Programme in Computer Science
Informaatioteknologian ja viestinnän tiedekunta - Faculty of Information Technology and Communication Sciences
This publication is copyrighted. You may download, display and print it for Your own personal use. Commercial use is prohibited.
Hyväksymispäivämäärä
2024-12-16
Julkaisun pysyvä osoite on
https://urn.fi/URN:NBN:fi:tuni-2024120910887
https://urn.fi/URN:NBN:fi:tuni-2024120910887
Tiivistelmä
Hyperscanning is the study of multiple brain activities simultaneously while participants engage in a common task. In recent times, hyperscanning has gained much attention in the research area to investigate social interaction, mutual engagement tasks, teamwork, and so on.
Electroencephalography, or EEG, is a non-invasive neuroimaging technique to measure the brain’s electrical activity. It measures the brain’s neural activity when electrodes are placed on a scalp. EEG devices are ideal for hyperscanning due to their excellent temporal resolution, portability, wireless capability, and low cost. In this EEG hyperscanning recording setup, the leading equipment included two Bluetooth-enabled computers, EEG devices such as the Muse S and Unicorn Hybrid Black, a local area network (LAN), and corresponding applications like LabRecorder and MuseLSL2/Unicorn Suite. The popular Tetris game was chosen to assess the mutual engagement of two participants while they were relaxing and also played three different versions of the Tetris game. The recordings had four Relaxation phases, and three Tetris games were played. Later, these phases were classified, such as Relaxation vs Game as binary and Game1 vs Game2 vs Game3 as multi-class classification. It was observed that binary classification, such as Relaxation vs. Game, outperformed multi-class classification in performance. Interestingly, for binary classification, the Unicorn cross-coherence features provide consistent results across both random and Leave-One-Subject-Out (LOSO) Cross-Validation in all metrics. The total sample size was 10, which is relatively small. More sample data is required to draw any conclusive statement.
Electroencephalography, or EEG, is a non-invasive neuroimaging technique to measure the brain’s electrical activity. It measures the brain’s neural activity when electrodes are placed on a scalp. EEG devices are ideal for hyperscanning due to their excellent temporal resolution, portability, wireless capability, and low cost. In this EEG hyperscanning recording setup, the leading equipment included two Bluetooth-enabled computers, EEG devices such as the Muse S and Unicorn Hybrid Black, a local area network (LAN), and corresponding applications like LabRecorder and MuseLSL2/Unicorn Suite. The popular Tetris game was chosen to assess the mutual engagement of two participants while they were relaxing and also played three different versions of the Tetris game. The recordings had four Relaxation phases, and three Tetris games were played. Later, these phases were classified, such as Relaxation vs Game as binary and Game1 vs Game2 vs Game3 as multi-class classification. It was observed that binary classification, such as Relaxation vs. Game, outperformed multi-class classification in performance. Interestingly, for binary classification, the Unicorn cross-coherence features provide consistent results across both random and Leave-One-Subject-Out (LOSO) Cross-Validation in all metrics. The total sample size was 10, which is relatively small. More sample data is required to draw any conclusive statement.