LieWaves: dataset for lie detection based on EEG signals and wavelets

This dataset includes EEG signals for lie detection. EEG signals were collected using a wearable and portable EEG device called Emotiv Insight, which has 5 channels, from 27 different subjects. The subjects participated in two experiments, taking on the roles of deceivers and truth-tellers. In each experiment, a box with 5 different beads was given to the subjects, and they were instructed to take 2 beads from the box and place them in their pockets. In the first experiment, subjects were asked to decide whether to assume the role of a deceiver or a truth-teller. In the second experiment, they were required to take on the opposite role. During the experiments, subjects watched a video composed of images of the beads in the box placed in front of them. The video clip started with a 3-second black screen, followed by 2 seconds of bead images and 1 second of a black screen, repeating in this pattern. After obtaining EEG data, the initial 2 seconds of excessive signal data were removed from the raw data, resulting in a total of 75 seconds of EEG data. In the deceiver role, subjects clicked the button in their left hand labeled "no" if the displayed image matched the bead they took, and the button in their right hand labeled "yes" if it did not, thus deceiving about all the images. For the truth-teller role, the opposite actions were taken, clicking "yes" for the taken bead image and "no" for the not-taken bead image, thus telling the truth for all images. EEG signals were recorded following this procedure. The EEG signals underwent an offset removal process to obtain raw EEG data. Both raw data and preprocessed EEG data were stored in .csv format. The purpose of this dataset is to provide EEG signals for lie detection, offering an alternative and diverse dataset with different channel counts.

When this data set is used, the relevant article must be cited. The relevant article citation is below.

Aslan, M., Baykara, M. & Alakus, T.B. LieWaves: dataset for lie detection based on EEG signals and wavelets. Med Biol Eng Comput (2024). https://doi.org/10.1007/s11517-024-03021-2

本数据集收录了用于谎言检测的脑电图（EEG）信号。所采集的EEG信号由名为Emotiv Insight的便携式脑电图装置产生，该装置具备5个通道，并从27位不同受试者处收集。受试者参与了两个实验，分别扮演欺骗者和说真话者的角色。在每个实验中，受试者被给予一个装有5种不同珠子的盒子，并被指示从中取出2颗珠子放入口袋。在第一个实验中，受试者需决定扮演欺骗者或说真话者的角色。在第二个实验中，他们需承担相反的角色。实验过程中，受试者观看一个由放置在他们面前的盒子中珠子图像组成的视频。视频剪辑以3秒的黑色屏幕开始，随后是2秒的珠子图像和1秒的黑色屏幕，以此循环播放。在获得EEG数据后，从原始数据中移除了前2秒的过量信号数据，从而得到总共75秒的EEG数据。在扮演欺骗者角色时，如果显示的图像与取出的珠子相匹配，受试者需点击左手上的标有“否”的按钮，如果不相匹配，则点击右手上的标有“是”的按钮，以此来对所有图像进行欺骗。在扮演说真话者角色时，采取相反的行动，点击“是”以确认取出的珠子图像，点击“否”以确认未取出的珠子图像，从而对所有图像说真话。根据此程序记录EEG信号。EEG信号经过偏移消除处理以获得原始EEG数据。原始数据和预处理后的EEG数据均以.csv格式存储。本数据集旨在提供用于谎言检测的EEG信号，提供了一种具有不同通道计数的不同类型和多样化的数据集。