A Study on Using Blink Parameters from EEG Data for Lie Detection

摘要

The use of EEG signals taken from human brain showcases promising information for detecting lies and false intent in a person with increased accuracy. The physical changes detected by and EEG device are concomitant to human reaction and cannot be controlled by a person. Blinking is one such component that changes when a person lies. Blink rate cannot be simply calculated by physical observation. There are certain other methods to detect blink rate, making use of ocular tracking devices however, this method of using EEG signals to extract ocular characteristics allows this methodology to be combined with other techniques to detect lies using EEG signals. This means that a single EEG acquisition device can be used as a comprehensive lie detector. A person's blink rate decreases drastically while they are lying and then increases rapidly moments after. This can be used to check if a response to a question is true or false. For the purpose of this study we have collected data from 10 subjects, taking 10 readings from each. We have studied the number of blinks and thus the blinks per minute of a subject during and after they have told a lie. The results were then used to check if the accuracy of our method. The outcomes of the study showed a decrease in blink rate during a lie with 95.12% accuracy. This proves that blink characteristics extracted from EEG datasets can be used as an effective means to detect lies in a subject. This method provides an alternative to the existing polygraph tests used in lie detection which can conceivably be made admissible in court in the future.