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

This study introduces an electroencephalography (EEG)-based dataset to analyze lie detection. Various analyses or detections can be performed using EEG signals. Lie detection using EEG data has recently become a significant topic. In every aspect of life, people find the need to tell lies to each other. While lies told daily may not have significant societal impacts, lie detection becomes crucial in legal, security, job interviews, or situations that could affect the community. This study aims to obtain EEG signals for lie detection, create a dataset, and analyze this dataset using signal processing techniques and deep learning methods. EEG signals were acquired from 27 individuals using a wearable EEG device called Emotiv Insight with 5 channels (AF3, T7, Pz, T8, AF4). Each person took part in two trials: one where they were honest and another where they were deceitful. During each experiment, participants evaluated beads they saw before the experiment and stole from them in front of a video clip. This study consisted of four stages. In the first stage, the LieWaves dataset was created with the EEG data obtained during these experiments. In the second stage, preprocessing was carried out. In this stage, the automatic and tunable artifact removal (ATAR) algorithm was applied to remove the artifacts from the EEG signals. Later, the overlapping sliding window (OSW) method was used for data augmentation. In the third stage, feature extraction was performed. To achieve this, EEG signals were analyzed by combining discrete wavelet transform (DWT) and fast Fourier transform (FFT) including statistical methods (SM). In the last stage, each obtained feature vector was classified separately using Convolutional Neural Network (CNN), Long Short-Term Memory (LSTM), and CNNLSTM hybrid algorithms. At the study's conclusion, the most accurate result, achieving a 99.88% accuracy score, was produced using the LSTM and DWT techniques. With this study, a new data set was introduced to the literature, and it was aimed to eliminate the deficiencies in this field with this data set. Evaluation results obtained from the data set have shown that this data set can be effective in this field.

Unveiling the efficacy of the feedback concealed information test in collaborative crime detection.

Collaborative crime poses severe social hazards. In collaborative crime scenarios, previous studies have indicated that perpetrators' collaborative encoding can impair the detection efficiency of P300-based complex trial protocols due to the collaborative encoding deficit. The feedback concealed information test (fCIT), a unique variation of the concealed information test, provides participants with feedback on how well they conceal information from memory. The fCIT, which has proven to be highly efficient, detects concealed information using recognition P300 along with feedback-related event-related potentials, and reflects the subject's motivation to conceal. However, no studies have examined the fCIT's effectiveness in identifying collaborative criminals. We propose that the fCIT's efficiency persists in cases of collaborative crime and test this hypothesis using a sample of 48 participants. The participants in the collaborative groups were instructed to have hushed conversations about theft to simulate the collaborative crime process. Subsequently, they completed the fCIT. The findings indicate a significant decline in recognition P300's detection efficiency when participants committed crimes collaboratively. Nevertheless, the detection efficiency of feedback P300 and feedback-related negativity remained high. This study's outcomes illustrate the capacity of the fCIT to detect perpetrators involved in collaborative crime.

Examining levels of processing using verbal & pictorial stimuli with the complex trial protocol in a mock theft scenario.

The Complex Trial Protocol (CTP) is an EEG-based Concealed Information Test (CIT). Depth of processing influences memorability where deeper processing increases recollection. The CTP's performance as a function of shallow versus deep levels of processing has not been explored. Two experiments were conducted, one with verbal stimuli and the other with their pictorial referents. In both experiments, participants were randomly assigned to three groups, Innocent Control (Control) condition, Guilty Immediate Shallow Processing (Shallow) condition, and Guilty Immediate Deep Processing (Deep) condition. Shallow and Deep participants from both experiments underwent the same mock theft scenario and all three groups were later exposed to either a verbal (N = 41) or pictorial (N = 43) stimulus on a computer monitor. In the word study, no differences in CIT effect were found between any of the groups. Areas under the curve (AUCs) did not differ from chance (.624 and .679 for Shallow and Deep groups respectively). In the image study, the CIT effect for the Shallow and Deep groups differed from the Control one. The AUCs (.755 and .943 for the Shallow and Deep groups respectively) differed significantly from each other. Levels of Processing (LOP) did not appear to have any bearing on CTP performance when words were used as probes but did have an effect when images were used. The findings may hint at some of the limitations of the CTP and fail to replicate similar experiments, especially when words are used as probes, from the late Peter Rosenfeld's laboratory.

Individual differences in self-reported lie detection abilities.

Previous literature on lie detection abilities bears an interesting paradox. On the group level, people detect others' lies at guessing level. However, when asked to evaluate their own abilities, people report being able to detect lies (i.e., self-reported lie detection). Understanding this paradox is important because decisions which rely on credibility assessment and deception detection can have serious implications (e.g., trust in others, legal issues). In two online studies, we tested whether individual differences account for variance in self-reported lie detection abilities. We assessed personality traits (Big-Six personality traits, Dark Triad), empathy, emotional intelligence, cultural values, trust level, social desirability, and belief in one's own lie detection abilities. In both studies, mean self-reported lie detection abilities were above chance level. Then, lower out-group trust and higher social desirability levels predicted higher self-reported lie detection abilities. These results suggest that social trust and norms shape our beliefs about our own lie detection abilities.

Building a second-opinion tool for classical polygraph.

Classical polygraph screenings are routinely used by critical businesses such as banking, law enforcement agencies, and federal governments. A major concern of scientific communities is that screenings are prone to errors. However, screening errors are not only due to the method, but also due to human (polygraph examiner) error. Here we show application of machine learning (ML) to detect examiner errors. From an ML perspective, we trained an error detection model in the absence of labeled errors. From a practical perspective, we devised and tested successfully a second-opinion tool to find human errors in examiners' conclusions, thus reducing subjectivity of polygraph screenings. We report novel features that uplift the model's accuracy, and experimental results on whether people lie differently on different topics. We anticipate our results to be a step towards rethinking classical polygraph practices.

The use-the-best heuristic facilitates deception detection.

Decades of research have shown that people are poor at detecting deception. Understandably, people struggle with integrating the many putative cues to deception into an accurate veracity judgement. Heuristics simplify difficult decisions by ignoring most of the information and relying instead only on the most diagnostic cues. Here we conducted nine studies in which people evaluated honest and deceptive handwritten statements, video transcripts, videotaped interviews or live interviews. Participants performed at the chance level when they made intuitive judgements, free to use any possible cue. But when instructed to rely only on the best available cue (detailedness), they were consistently able to discriminate lies from truths. Our findings challenge the notion that people lack the potential to detect deception. The simplicity and accuracy of the use-the-best heuristic provides a promising new avenue for deception research.

Orienting versus inhibition: The theory behind the ocular-based Concealed Information Test.

When trying to conceal one's knowledge, various ocular changes occur. However, which cognitive mechanisms drive these changes? Do orienting or inhibition-two processes previously associated with autonomic changes-play a role? To answer this question, we used a Concealed Information Test (CIT) in which participants were either motivated to conceal (orienting + inhibition) or reveal (orienting only) their knowledge. While pupil size increased in both motivational conditions, the fixation and blink CIT effects were confined to the conceal condition. These results were mirrored in autonomic changes, with skin conductance increasing in both conditions while heart rate decreased solely under motivation to conceal. Thus, different cognitive mechanisms seem to drive ocular responses. Pupil size appears to be linked to the orienting of attention (akin to skin conductance changes), while fixations and blinks rather seem to reflect arousal inhibition (comparable to heart rate changes). This knowledge strengthens CIT theory and illuminates the relationship between ocular and autonomic activity.

A new theoretical perspective on concealed information detection.

The concealed information test (CIT) was designed to detect concealed knowledge. It does so by measuring differential physiological and behavioral responses to concealed, compared to control, items - i.e., the CIT effect. Although the CIT has gained extensive empirical support in laboratory studies, scientific validity requires also a theoretical understanding of the method's underlying mechanisms. In this article, we present a new theoretical perspective. Specifically, we elaborate and clarify several features of Orienting Response (OR) theory, which has been the dominant theory in this domain. Importantly, we suggest for the first time that the voluntary (rather than the involuntary) OR modulates the CIT effect. Second, we argue that motivational-emotional accounts of the CIT effect are consistent with OR theory and cannot be considered as alternative approaches. Finally, we discuss some more recent developments which highlight the idea that different physiological measures reflect different underlying mechanisms - an idea dubbed as response fractionation.

Selective memory detection: Context-dependent stimulus significance in the concealed information test.

Previous studies have posited that the significance of a crime-relevant item in a question produces differential physiological responses in the Concealed Information Test (CIT). However, this term is equivocal and needs to be clarified in order to strengthen the theoretical underpinnings of the CIT. The present study examined the hypothesis that differential responding depends on the examinee's understanding of which item in a question is relevant to a given context. Participants performed a mock theft task, in which they were instructed to steal one item from each of two different locations. An identical CIT question asking about each stolen item was presented under different location contexts while skin conductance response, heart rate, and respiratory activity were recorded. Results indicated that only the relevant item specified by the context of the instruction elicited reliable differential physiological responses. This finding implies that differential responding in the CIT is flexible and context-dependent, and that specifying the subject of a given question is important for detecting crime-relevant memories in practical criminal investigations.

Only true and fabricated baseline statements combined might improve lie, but not truth, detection.

OBJECTIVES: We tested the effect of true and fabricated baseline statements from the same sender on veracity judgments. HYPOTHESES: We predicted that presenting a combination of true and fabricated baseline statements would improve truth and lie detection accuracy, while presenting a true baseline would improve only truth detection, and presenting a fabricated baseline would only improve lie detection compared with presenting no baseline statement. METHOD: In a 4 × 2 within-subjects design, 142 student participants (Mage = 23.47 years; 118 female) read no baseline statement, a true baseline statement, a fabricated baseline statement, and a combination of a true and a fabricated baseline statement from 29 different senders. Participants then rated the veracity of a true or fabricated target statement from the same 29 senders. RESULTS: Logistic mixed-effects models with senders and participants as random effects showed no significant differences in overall veracity judgment accuracy between the no-baseline (51%) and either the true-baseline (44%) or the fabricated-baseline (49%) conditions. Equivalence tests failed to show the predicted equivalence of these accuracy rates. Separate analyses of truth and lie detection rates confirmed the assumed improvement of lie detection in the combination-of-true-and-fabricated-baseline condition (accuracy = 39%-61%). No other truth or lie detection rate changed significantly except that, unexpectedly, a true baseline reduced truth detection accuracy (64%-49%). CONCLUSIONS: Baseline statements largely did not affect judgment accuracy and, in the case of true baselines, even had a negative impact on truth detection. The rather small positive effect of two baseline statements on lie detection suggests an avenue for further research, especially with expert raters. (PsycInfo Database Record (c) 2022 APA, all rights reserved).

Detection of concealed knowledge via the ERP-based technique Brain Fingerprinting: Real-life and real-crime incidents.

Brain Fingerprinting (BFP) is an electroencephalogram-based system used to detect knowledge, or absence of knowledge of a real-life incident (e.g., a crime) in a person's memory. With the help of BFP, a potential crime suspect can be classified as possessing crime-related information (Information-Present), not possessing crime-related information (Information-Absent), or Indeterminate (BFP unable to classify a subject). In the lab setting, we compare the ground-truth of a subject (i.e., real-life involvement in an incident) against their classification based on BFP testing. We report two studies: replication of BFP with university students (Study 1) and replication of BFP with parolees (Study 2). In Study 1, we tested 31 subjects (24 females, seven males, mean age = 21.3) on either their own or another subject's real-life incident. BFP correctly classified nine Information-Present and 18 Information-Absent subjects, but with one false positive and three exclusions. In Study 2, we tested 17 male parolees (mean age = 47.5) on their own or another parolee's crime incident. BFP correctly classified two Information-Present and six Information-Absent subjects. However, there was also one false positive classification and three Indeterminates. Additionally, we identified three subjects who could not complete the BFP testing and two exclusions. We posit that BFP is not yet at a stage to be considered a robust and accurate crime-detection tool as claimed in former articles. Nevertheless, after addressing the limitations, BFP has considerable potential as an information detection tool in forensic investigations, especially for detecting idiosyncratic crime-relevant knowledge in a perpetrator, in addition to helping to confirm the accuracy of a suspect's claim of innocence.

Power and lie detection.

Social power undermines focus on others and increases reliance on stereotype-consistent information. Thus, power may enhance focus on stereotypical cues to deception, thereby decreasing lie detection accuracy. In three studies, we tested whether having power affects lie detection accuracy. Participants (overall N = 502) were asked to identify truthful and lying candidates (N = 12) during mock job interviews. Study 1 was a field experiment involving employees who held managerial and non-managerial positions (N = 88). In the following laboratory experiments, we manipulated power and asked participants to imagine themselves as managers (Study 2, N = 214) or provided them with control over resources and the ability to reward others (Study 3, N = 200). In Studies 2 and 3, we additionally manipulated the method of lie detection (direct vs. indirect). In contrast to the original hypotheses, we found that power led to increased veracity assessment accuracy. Having power over others enhances the accuracy of one's veracity assessment, although this increase is small and limited to lie detection (Study 1) or direct judgments (Studies 2 & 3). Together, power affects the processing of social information and what aspects of this information are taken into account.

Effective Connectivity in Cortical Networks During Deception: A Lie Detection Study Based on EEG.

Thus far, when deception behaviors occur, the connectivity patterns and the communication between different brain areas remain largely unclear. In this study, the most important information flows (MIIFs) between different brain cortices during deception were explored. First, the guilty knowledge test protocol was employed, and 64 electrodes' electroencephalogram (EEG) signals were recorded from 30 subjects (15 guilty and 15 innocent). Cortical current density waveforms were then estimated on the 24 regions of interest (ROIs). Next, partial directed coherence (PDC), an effective connectivity (EC) analysis was applied in the cortical waveforms to obtain the brain EC networks for four bands: delta (1-4 Hz), theta (4-8 Hz), alpha (8-13 Hz) and beta (13-30 Hz). Furthermore, using the graph theoretical analysis, the network parameters with significant differences in the EC network were extracted as features to identify the two groups. The high classification accuracy of the four bands demonstrated that the proposed method was suitable for lie detection. In addition, based on the optimal features in the classification mode, the brain "hub" regions were identified, and the MIIFs were significantly different between the guilty and innocent groups. Moreover, the fronto-parietal network was found to be most prominent among all MIIFs at the four bands. Furthermore, combining the neurophysiology significance of the four frequency bands, the roles of all MIIFs were analyzed, which could help us to uncover the underlying cognitive processes and mechanisms of deception.

A novel and effective item-source complex trial protocol: Discrimination of guilty from both knowledgeable and unknowledgeable innocent subjects.

Innocent subjects who are knowledgeable of crime-related information will often be misclassified as "guilty" in P300-based complex trial protocol (CTP). Therefore, it is necessary to develop a more rigorous CTP that can effectively discriminate the guilty from both the knowledgeable and the unknowledgeable innocents. Sometimes the guilty and the knowledgeable innocents possess the same item memories but different source memories. The present study designed a novel item-source complex trial protocol based on the differences of source memory among the three kinds of individuals. Either the crime-related probe (e.g., the stolen ring) or one of the crime-unrelated stimuli (e.g., watch, earring, bracelet, or bangle) (item memory) was presented in the first part of each trail, and either a stealing-source word (e.g., stole) or other-source word (e.g., fetched) (source memory) was presented in the second part of each trail. The results showed that: (1) the P300 evoked by item memory could effectively discriminated the guilty from the unknowledgeable innocent (AUC = 0.76) but failed to effectively discriminate the guilty from the knowledgeable innocent (AUC = 0.60); (2) the late positive component evoked by source memory could effectively discriminated the guilty from both the knowledgeable innocent (AUC = 0.94) and the unknowledgeable innocent (AUC = 0.84) in one test.

Detecting concealed information using functional near-infrared spectroscopy (fNIRS) combined with skin conductance, heart rate, and behavioral measures.

In this study, brain imaging data from functional near-infrared spectroscopy (fNIRS) associated with skin conductance response (SCR), heart rate (HR), and reaction time (RT) were combined to determine if the combination of these indicators could improve the efficiency of deception detection in concealed information test (CIT). During the CIT, participants were presented with a series of names and cities that served as target, probe, or irrelevant stimuli. In the guilty group, the probe stimuli were the participants' own names and hometown cities, and they were asked to deny this information. Our results revealed that probe items were associated with longer RT, larger SCR, slower HR, and higher oxyhemoglobin (HbO) concentration changes in the inferior prefrontal gyrus (IFG), middle frontal gyrus (MFG), and the superior frontal gyrus (SFG) compared with irrelevant items for participants in the guilty group but not in the innocent group. Furthermore, our results suggested that the combination of RT, SCR, HR, and fNIRS indicators could improve the deception detection efficiency to a very high area under the ROC curve (0.94) compared with any of the single indicators (0.74-0.89). The improved deception detection efficiency might be attributed to the reduction of random error and the diversiform underlying the psychophysiological mechanisms reflected by each indicator. These findings demonstrate a feasible way to improve the deception detection efficiency by using combined multiple indicators.

The effect of mental countermeasures on a novel brain-based feedback concealed information test.

The feedback concealed information test (fCIT) is a novel form of the CIT, providing participants with feedback regarding their memory concealment performance. The fCIT utilizes event-related potentials (recognition-P300 and feedback-related event-related potentials) and has been shown to provide high efficiency in detecting information concealment. However, it is unclear how well the fCIT performs in the presence of mental countermeasures. To address this question, participants were trained to use countermeasures during fCIT. Results showed that the recognition-P300 efficiency decreased when participants used countermeasures. However, the efficiencies of feedback-related negativity and feedback-P300 were unchanged, with feedback-P300 still showing a high detection efficiency (AUC = 0.86) during countermeasures. These findings demonstrate the potential of fCIT for subverting countermeasures.

Retroactive memory interference reduces false positive outcomes of informed innocents in the P300-based concealed information test.

In the Concealed Information Test (CIT), differential responses between crime-relevant and crime-irrelevant items are indicative of concealed knowledge of a crime, and are used to classify an individual as either "guilty" or "innocent". However, when crime-relevant items are leaked before the test, an innocent examinee can exhibit enhanced responses to the crime-relevant items, thus causing such examinee to be wrongly classified as guilty. In an attempt to solve this problem, we examined the role of retroactive memory interference (RI) in differentiating informed innocents from guilty participants, using a P300-based CIT. Participants acquired crime-related knowledge either by committing a mock crime (guilty group) or reading a paper that described a mock crime (informed innocent group). Subsequently, the participants within each condition were randomly assigned to either an RI group, where they were exposed to new crime-related details before the CIT, or a control group. We found an interaction between guilty and RI groups: in the guilty group, there was a significant difference in P300 amplitude between the probe and irrelevant items, regardless of RI manipulation, whereas in the informed innocent group, a difference in P300 amplitude between the probe and irrelevant items was significant only in the control group, but not in the RI group. This led to an improved detection rate of the informed innocents (31% for the control group vs. 77% for the RI group). These results suggest that RI manipulation could be used to reduce the false positive outcomes of informed innocents without affecting the detection rate of guilty participants.

The effects of practicing a novel countermeasure on both the semantic and episodic memory-based complex trial protocols.

This study applied a countermeasure-resistant version of the Concealed Information Test - the Complex Trial Protocol (CTP) - in an information recognition scenario. We replicated and extended the effects of a novel countermeasure developed by Lukács et al., (2016) on both Semantic and Episodic CTPs. We measured participants' response time and P300 event-related potential to rare, crime-relevant probe stimuli, or frequent, non-crime-relevant irrelevant stimuli in two ways: 1) probe vs the average of all irrelevants (PvIall), and 2) probe vs the maximum irrelevant (PvImax). We hypothesized that countermeasure use would only impair information recognition (as indexed by P300) when participants had practiced the countermeasure beforehand. We further hypothesized that recognition of less salient, Episodic information (i.e., jewelry items from a mock crime) would be impaired by countermeasure use more than the recognition of highly salient, Semantic information (i.e., birthdates). Individual diagnostics based on the area under the receiver operating characteristic curve (Semantic CTP: practice n = 22, non-practice n = 23; Episodic CTP: practice n = 19, non-practice n = 18) revealed that the Semantic CTP was affected by the novel countermeasure, but both PvIall and PvImax analyses remained diagnostically useful. The Episodic CTP's performance, however, was reduced to chance, regardless of practice or analysis type. These results are important for both the field of deception detection and the CTP literature. Research on improvements to the Episodic CTP is required.

Progress of Keystroke Dynamics in Deception Research.

Traditional polygraph techniques mostly rely on the changes of an individual's physiological indicators, such as electrodermal activity, heart rate, breath, eye movement and function of neural signals and other indicators. They are easily affected by individual physical conditions, counter-tests, external environment and other aspects, and it is difficult to conduct large-scale screening tests based on the traditional polygraph techniques. The application of keystroke dynamics to polygraph can overcome the shortcomings of the traditional polygraph techniques to a large extend, increase the reliability of polygraph results and promote the validity of legal evidence of polygraph results in forensic practice. This paper introduces keystroke dynamics and its application in deception research. Compared with the traditional polygraph techniques, keystroke dynamics can be used with a relatively wider application range, not only for deception research but also for identity identification, network screening and other large-scale tests. At the same time, the development direction of keystroke dynamics in the field of polygraph is prospected.

Countering information leakage in the Concealed Information Test: The effects of item detailedness.

Concealed Information Tests (CIT) are administered to verify whether suspects recognize certain features from a crime. Whenever it is presumed that innocent suspects were contaminated with critical information (e.g., the perpetrator had a knife), the examiner may ask more detailed questions (e.g., specific types of knives) to prevent false positives. However, this may increase the number of false negatives if the true perpetrator fails to discern specific details from its plausible irrelevant controls, or because detailed crime-scene information may be forgotten. We examined whether presenting items at the exemplar level protects against contamination, and whether it compromises the sensitivity in a physiological CIT. Participants (N = 142) planned a mock-robbery, with critical items encoded either at the category or at the exemplar level. The CIT was administered immediately or after a 1-week-delay, with questions phrased at the categorical or exemplar level. There were no effects of time delay. Results revealed that when item detailedness was congruent at encoding and testing, the SCR, HR, and RLL showed larger differential responses, as compared with incongruent conditions. Participants contaminated with crime knowledge at the categorical level did not show a CIT-effect for crime details at the exemplar level, suggesting detailed questions may counter the leakage problem. Asking questions at the exemplar level did not reduce the CIT detection efficiency as compared to asking questions at the categorical level. The importance of congruency between encoding and testing provides examiners with a challenge, as it is difficult to estimate how details are naturally encoded.