Neuro signal based lie detection

Polygraph has been used for deception detection as an alternative way to provide proof for investigation; however the significant shortcoming occurs based on its reliability. This paper proposes to understanding the relationship between lying and frontal lobe during some specific tasks by mapping their EEG signals. In the present study, Multiplayer neural network are used for bio-signal classification. For each group subject, features from Short-time Fourier transform (STFT) are computed, for each channel. Multi-layer Perception (MLP) is used for classification to differentiate between deception and truth types of EEG classes with the accuracy of around 90%. Feature extractions were strong enough and mental processes linked with the activation of the frontal cortex. Three students´ alpha waves were collected while they play the “Pokemon card” and this card chosen in order to challenge participants during testing duration. The goal of this study is to evaluate different state of truth and deception for the extraction of EEG features that are most suitable for the discrimination between each stage. The EEG recordings were obtained alpha waves from four frontal electrodes and two midline electrodes.