A brain signal-based credibility assessment approach

2011

The proposed system is designed to make EEG database of criminals during the investigation by crime branch. System will be detecting the criminals on the bases of thought related to crime to be stored in brain, when the question related to the crime will be asked automatically remind the past event. The whole system will be tested after complete the investigation by crime branch.

Indian Journal of Medical Specialities, 2011

The classical methods of forensic science like finger printing, polygraphic lie detector test, narcoanalysis and DNA finger printing are being questioned for their reliability and accuracy. Newer developments in investigating crime are evolving like the use of event related evoked potentials to retrieve the sequence of events of crime as recorded in the brain of the criminal/accused. The principle of event related evoked potentials is simple. In an oddball paradigm two stimuli of varying characteristics are presented to the individual, one repeated frequently and the other being rare, is given randomly at infrequent intervals. The subject is asked to respond to the infrequent or target stimulus by pressing a button. The response to target stimulus is recorded as P3 wave of the cognitive evoked potentials. P3 wave of the ERPs has wide applications in cognitive forensic science and interrogative polygraphy in particular. Late vertex positivity in ERP reflects guilt detection in a test known as guilty knowledge. Memory and encoding related multifaceted electroencephalographic responses (MERMER) have been used to determine whether the subject had relevant information (committing of crime) in his brain, after presenting him with probing (relevant), target and non-target stimuli. Two high amplitude P3 waves are recorded in response to target and probing stimuli in case of a knowledgeable person and only one P3 in case of an innocent person. The newly emerging, late responses to ERP is a step forward to quantify the thought processing feature of the brain. Brain wave science has emerged as a new discipline with promising application in field of crime investigation.

2015 International Conference on Automation, Cognitive Science, Optics, Micro Electro-Mechanical System, and Information Technology (ICACOMIT), 2015

Lie detection mechanism is used to evaluate the ethical and moral values of people. In general, these moral and ethical values are difficult to determine. Earlier technique like polygraph does not provide accuracy in determining mental behaviour. Lately, it is possible to study psychological response of a person using electroencephalogram (EEG). Based on this, we develop a standalone solution that can detect brain signals using EEG. To this end, we extract the desired features from the brain signals acquired through sixteen electrodes using various extraction techniques. Then, we implement linear discriminant analysis (LDA) classification technique to differentiate the positive and negative samples from the signals obtained from sensors in order to achieve a decision for either guilty subject or innocent subject accordingly. The whole work is designed with the help of MATLAB and Xilinx tool. For efficiency, the whole system is implemented on FPGA. The amount of correct deception detection in guilty and innocent subjects are 85%. It is easy and more convenient approach than other previously used methods.

Computer Methods and Programs in Biomedicine, 2009

c o m p u t e r m e t h o d s a n d p r o g r a m s i n b i o m e d i c i n e 9 4 ( 2 0 0 9 ) 48-57 Event-related potential (ERP) P300 Lie detection

Deception detection studies have become a trending topic over the past 34 years. Lie detection seems to have moral, ethical, methodological, clinical, and legal implications. Although researchers have studied the P300 (P3) Event-Related Potential (ERP) component towards an understanding of the phenomenon extensively, deception remains a well-hidden cognitive process. The basic set-up of most of the deception detection studies contains a mock-crime scenario and the use of a Concealed Information Test (CIT). The same design requires a deceptive response from the participants. Towards this direction, the present study employed a fundamental research question via examining if meaningful stimuli can modulate the P300 component compared to target & irrelevant stimuli in guilty participants. The aim of the present study was: a) to check if a non-deceptive design could produce similar results to previous ERP-based deception detection studies, and b) to raise interest concerning the set-up of the experimental design of previous applications. In line with the literature, and with no deceptive behavior, P3 results revealed similar patterns and statistical significance. The unexpected findings of Early Posterior Negativity (EPN) and P100 (P1) provide a new index to be further explored in future deception detection studies. While the present results should not be taken as a new index for deception, they appoint a new path against the phenomenon. Future research suggestions are also addressed.

PLOS ONE, 2015

Deception is an impactful social event that has been the focus of an abundance of researches over recent decades. In this paper, an electroencephalography (EEG) study is presented regarding the cognitive processes of an instructed liar/truth-teller during the time window of stimulus (question) delivery period (SDP) prior to their deceptive/truthful responses towards questions related to authentic (WE: with prior experience) and fictional experience (NE: no prior experience). To investigate deception in non-experienced events, the subjects were given stimuli in a mock interview scenario that induced them to fabricate lies. To analyze the data, frequency domain network and connectivity analysis was performed in the source space in order to provide a more systematic level understanding of deception during SDP. This study reveals several groups of neuronal generators underlying both the instructed lying (IL) and the instructed truth-telling (IT) conditions for both tasks during the SDP. Despite the similarities existed in these group components, significant differences were found in the intra-and inter-group connectivity between the IL and IT conditions in either task. Additionally, the response time was found to be positively correlated with the clustering coefficient of the inferior frontal gyrus (44R) in the WE-IL condition and positively correlated with the clustering coefficient of the precuneus (7L) and the angular gyrus (39R) in the WE-IT condition. However, the response time was found to be marginally negatively correlated with the clustering coefficient of the secondary auditory cortex (42L) in the NE-IL condition and negatively correlated with the clustering coefficient of the somatosensory association cortex (5L, R) in the NE-IT condition. Therefore, these results provide complementary and intuitive evidence for the differences between the IL and IT conditions in SDP for two types of deception tasks, thus elucidating the electrophysiological mechanisms underlying SDP of deception from regional, interregional , network, and inter-network scale analyses.

Biomedical Signal Processing and Control, 2019

Deception is a complex cognitive process in which liars always want to conceal some information. Concealed information test (CIT) is a useful paradigm which widely used in deception detection. Recent evidences from brain research suggest deception involves various cognitive activities, and most electroencephalograph (EEG) based concealed information test basically focuses on the signals from few electrodes and analyzed separately. In order to investigate the functional connectivity in different brain regions and the features from spatial domain, we applied graph theoretical concept to evaluate the changes of functional brain networks in guilty group compared with innocent in this article. Five different connectivity methods, including linear and nonlinear interdependence analysis, were explored to the multi-channel EEG signals, to explore which method is best for CIT. The result shows deception responses showed an increased connectivity level. Intraregional and interregional connectivity analysis also showed that deception was associated with increased activity in certain brain areas. Statistical analysis of network parameters showed these features form the two groups were significantly different, and mutual information was the best approach during which for network construction in CIT. Simultaneously, the deception response showed increased small-worldness. The results support the hypothesis that deception mainly involved in the process of working memory, which shows an enhanced connectivity and small-world properties. These findings reveal different dynamic networks in deception and truth telling state, and could be used to identify deception in individual subjects.

The study of electroencephalography (EEG) for deception detection has long been regarded as an alternative to the standard polygraphy, whose main shortcoming is its unacceptably low level of reliability. Most of the EEG deception research has focused on the amplitude and topography of P300. However, the characteristics of the P300 component are tightly connected to the experimental design and hence countermeasures are easily available for P300-based deception detection. The goal of this study is to evaluate different epoching criteria for the extraction of EEG features that are most suitable for the discrimination between truthful and deceptive responses. In order to reach this aim, a modified version of the Guilty Knowledge Test was used where EEG recordings were obtained from four frontal electrodes and two midline electrodes. In none of the electrodes the P300 component differed between deceptive and truthful responses. Differences have instead been revealed through the extraction of response-locked epochs and analysis of area under the curve.

Journal of Telecommunication, Electronic and Computer Engineering, 2017

In this paper, the differences in brain signal activity (EEG-P300 component) which detects whether a person is telling the truth or lying is explored. Brain signal activity is monitored when they are first respond to a given experiment stimulus. In the experiment, twelve subjects whose age are around 19 ± 1 years old were involved. In the signal processing, the recorded brain signals were filtered and extracted using bandpass filter and independent component analysis, respectively. Furthermore, the extracted signals were classified with adaptive neuro-fuzzy inference system method. The results show that a huge spike of the EEG-P300 amplitude on a lying subject correspond to the appeared stimuli is achieved. The findings of these experiments have been promising in testing the validity of using an EEG-P300 as a lie detector.

NeuroImage, 2005

Patterns of brain activity during deception have recently been characterized with fMRI on the multi-subject average group level. The clinical value of fMRI in lie detection will be determined by the ability to detect deception in individual subjects, rather than group averages. High-dimensional non-linear pattern classification methods applied to functional magnetic resonance (fMRI) images were used to discriminate between the spatial patterns of brain activity associated with lie and truth. In 22 participants performing a forced-choice deception task, 99% of the true and false responses were discriminated correctly. Predictive accuracy, assessed by cross-validation in participants not included in training, was 88%. The results demonstrate the potential of non-linear machine learning techniques in lie detection and other possible clinical applications of fMRI in individual subjects, and indicate that accurate clinical tests could be based on measurements of brain function with fMRI.