講演情報
[5O2-IS-5a-03]An EEG Measurement Methodologyfor Decision-Making Utilizing Graded Conflict Stimuli in Four-Choice Quizzes
〇Tatsuya Hasegawa1, Motoyuki Sanada2, Yashushi Naruse2, Ikuko Eguchi Yairi1 (1. Graduate School of Sophia University , 2. National Institute of Information and Communications Technology)
regular
キーワード：
Brain-Machine Interface (BMI)、Electroencephalogram(EEG)、Decision-making、Event-Related Potential (ERP)、Frequency analysis
This study aims to elucidate the neural mechanisms underlying semantic decision-making by introducing graded conflict stimuli precisely controlled by generative AI. We recorded electroencephalogram (EEG) data during a novel four-choice quiz task incorporating highly plausible distractors. Event-related potential analysis demonstrated an increased evidence accumulation load reflected by P300 amplitude, alongside post-decision monitoring and response inhibition mechanisms driven by N700 and righthemisphere activity. Furthermore, time-frequency analysis revealed the coordination of multi-layered cognitive algorithms: theta waves for conflict detection, alpha waves for blocking external inputs during internal searches, and delayed gamma waves for information integration and cognitive closure. These findings establish a crucial cornerstone for understanding the neural basis of human critical thinking and pave the way for real-time cognitive conflict estimation in an AI-symbiotic society.