Determining the state of truthfulness and falsehood by analyzing the acquired EEG signals

• Main Authors: Cakmak, Rumeysa, Zeki, Akram M.
• Format: Conference or Workshop Item
• Language: English; English
• Published: Institute of Electrical and Electronics Engineers Inc. 2016
• Subjects: TK5101 Telecommunication. Including telegraphy, radio, radar, television
• Online Access: http://irep.iium.edu.my/58531/1/58531_Determining%20the%20state%20of%20truthfulness_SCOPUS.pdf; http://irep.iium.edu.my/58531/2/58531_Determining%20the%20state%20of%20truthfulness.pdf; http://irep.iium.edu.my/58531/; http://ieeexplore.ieee.org/document/7515826/
• Institution: Universiti Islam Antarabangsa Malaysia

Detection of deception is particular importance for the criminal case and cognitive behaviors of individual. To understanding criminal behavior, extracting the characteristic of brain waves have obviously crucial. According to hypothesis, particularly prefrontal lobes associated with the deception. This paper alleged to understanding the relationship between deception and truth from frontal lobe during some specific tasks by mapping their EEG signals. In the present study, multiplayer neural network are used for bio-signal classification to diversify between patterns of lie and truth types of EEG classes with the accuracy of around 96%. Brain activity have been captured and characterized with EEG by focusing alpha waves. During the test, lie detection identified and especially focus to detect lie in individual subjects, rather than group averages. In this research, the classification methods applied and EEG machine differentiated the specific patterns of brain activity from frontal lobes associated with deception and truth. The responses from the 3 subjects was discriminated correctly with 99%. The ranges of accuracy of test from three subjects was between 88% to 96%, there was an exception in round three with subject three with 46%. While the participants were playing with Pokemon card, alpha waves were collected successfully. © 2016 IEEE.