Lie Detection and Truth Identification form EEG signals by using Frequency and Time Features

Abstract

Electroencephalogram signals show how the brain's electricity is working. An EEG signal is a measurement of voltage changes caused by the flow of ions through the brain's neurons. They have been studied in medical research to help find out what's wrong with people with Alzheimer's and epilepsy. They have also been used in brain-computer interface (BCI) applications.This paper discusses how well the electroencephalogram (EEG) can tell the difference between the truth and a lie based on brain waves (EEG). In the past, the feature sets were introduced as part of an EEG-based identification system. In this paper, they are tested as parts of a detection system. The statistical moments serve as the foundation for the examined feature set. For classification tasks, publicly available EEG datasets like the Dryad dataset, obtained from 15 participants, are fed into a feedforward neural network classifier. The 12 channels were trained separately, where each channel was divided into a different number of blocks, and the results indicated that some channels were bad. Some were very encouraging, reaching 100% in block number 16. Comparison with other recently published efforts shows promising identification results, with a single channel and a few features achieving the best results, up to 100% accuracy.