OBJECTIVE: Acoustic lie detection, prized for its covert nature and capability for remote processing, has spurred growing interest in acoustic features that can reliably aid in lie detection. In this study, the aim was to construct an acoustic polygraph based on a variety of phonetic and acoustic features rather than on electrodermal, cardiovascular, and respiratory values.
METHODS: Sixty-two participants from the University of Science and Technology of China, aged 18-30 years old, were involved in the mock crime experiment and were randomly assigned to the innocent and guilty groups. We collected 31 deceptive and truthful audios to analyze the performance of voice onset time (VOT) in lie detection.
RESULTS: Our findings revealed that VOT performed well in lie detection. Both the average sensitivity and specificity of the area under the curve are 0.888, and its lower and upper confidence limit are up to 0.803 and 0.973 respectively at the 95% confidence level. Although the other acoustic features had a lower reference value, they also provided a general trend in the judgment of lie detection.
CONCLUSIONS: Our results suggested that some acoustic features can be effectively used as aids to lie detection. Through a similar approach, we will explore more acoustic and phonetic features that contribute to detecting lies in the future.

In recent decades, many different governmental and nongovernmental organizations have used lie detection for various purposes, including ensuring the honesty of criminal confessions. As a result, this diagnosis is evaluated with a polygraph machine. However, the polygraph instrument has limitations and needs to be more reliable. This study introduces a new model for detecting lies using electroencephalogram (EEG) signals. An EEG database of 20 study participants was created to accomplish this goal. This study also used a six-layer graph convolutional network and type 2 fuzzy (TF-2) sets for feature selection/extraction and automatic classification. The classification results show that the proposed deep model effectively distinguishes between truths and lies. As a result, even in a noisy environment (SNR = 0 dB), the classification accuracy remains above 90%. The proposed strategy outperforms current research and algorithms. Its superior performance makes it suitable for a wide range of practical applications.

Skin conductance (SC) is one of the indices commonly used in the autonomic Concealed Information Test (CIT), but SC amplitude is sometimes difficult to quantify. This study investigated the applicability of SC area to the CIT as an unambiguous measure of SC. Secondary analyses of an existing dataset indicated that SC area could be used to classify examinees according to their knowledge status, although the equivalence of its performance with the SC amplitude was inconclusive. Classification performance was best when the SC signal was converted to the difference from question onset and summed over 10 s after question onset. SC area produced relatively consistent evaluations of differential responses based on the amplitude for inter-item comparisons. In addition, the classification performance of SC area exceeded the chance level even for participants who showed few measurable amplitudes (low-responsive participants). A possible implication is that a tonic increase in SC occurred in response to the relevant question even in low-responsive participants, who are traditionally excluded from analysis. The use of SC area might contribute to more impartial data evaluation and broader application of the CIT. These results indicate that SC area can be used as an alternative measure of SC in the CIT.

In this study, we propose an approach to detect deception during investigative interviews by integrating response latency and error analysis with the unexpected question technique. Sixty participants were assigned to an honest (n = 30) or deceptive group (n = 30). The deceptive group was instructed to memorize the false biographical details of a fictitious identity. Throughout the interviews, participants were presented with a randomized sequence of control, expected, and unexpected open-ended questions about identity. Responses were audio recorded for detailed examination. Our findings indicate that deceptive participants showed markedly longer latencies and higher error rates when answering expected (requiring deception) and unexpected questions (for which premeditated deception was not possible). Longer response latencies were also observed in participants attempting deception when answering control questions (which necessitated truthful answers). Moreover, a within-subject analysis highlighted that responding to unexpected questions significantly impaired individuals\' performance compared to answering control and expected questions. Leveraging machine-learning algorithms, our approach attained a classification accuracy of 98% in distinguishing deceptive and honest participants. Additionally, a classification analysis on single response levels was conducted. Our findings underscore the effectiveness of merging response latency metrics and error rates with unexpected questioning as a robust method for identity deception detection in investigative interviews. We also discuss significant implications for enhancing interview strategies.

Fifty years ago, in a trenchant analysis that challenged applied lie detection theory and science, David Lykken (1974) brought polygraphic interrogation methods to the attention of academia with the hope that these techniques would come under the purview of psychology and psychophysiology. In this perspective, I examine how this application of psychophysiology has evolved over the last half century and how its status has changed for 1) the comparison (control) question test (CQT), used in forensic applications; 2) polygraph screening tests, used to evaluate examinee integrity; and 3) the concealed information technique (CIT), used to assess recognition memory of crime details. The criticisms of the CQT and screening tests advanced by Lykken have been amplified and focused by the academic community over the last half century. However, this has had little effect on how these methods are practiced and has not curtailed their use. Although most private sector employee screening tests are now prohibited, personnel screening of government employees has increased, and screening tests of sex offenders are now commonplace. Even though the CIT has captured the interest of psychophysiologists as a scientifically defensible technique, its field use is negligible. A primary purpose of polygraphic interrogations continues to be the extraction of admissions and confessions. The lack of change in the polygraph testing status quo stems in large part from unwavering government support for the use of these methods. As a result, polygraph theory and research support continues to rest on shaky ground while practice continues unfettered by valid criticism.

Rapid advancements in artificial intelligence (AI) have driven interest in its potential application for lie detection. Unfortunately, the current approaches have primarily focused on technical aspects at the expense of a solid methodological and theoretical foundation. We discuss the implications thereof and offer recommendations for the development and regulation of AI-based deception detection.

Verbal content analyses to differentiate truthful and fabricated statements, such as the Criteria-Based Content Analysis (CBCA), are used in lie detection research as well as in practice to assess the credibility of statements in criminal court proceedings. Meta-analyses demonstrate validity of verbal content analyses above chance, but the traditional research paradigms usually lack either ecological or internal validity. The authors discuss the usage of immersive virtual reality scenarios to solve this dilemma, as both types of validity can be increased by this approach. In this integrative review of existing literature on the current use of virtual scenarios in forensic and victimology research, the authors extract strengths and limitations for possible VR studies in the context of verbal content analysis. Furthermore, novel ethical challenges involved are summarized and implications for future studies proposed. Overall, we argue in favor of using virtual reality scenarios to validate methods for verbal content analysis, but also urge to consider ethical limitations regarding unwanted short- and long-term aftereffects.

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.

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.

Social Defense Theory (SDT) states that anxious attachment reflects an adaptive sentinel strategy, whereby anxious people should be better able to detect lies than secure people. Existing research on this issue, however, has not been able to evaluate whether heightened lie detection among anxious individuals is due to an actual ability or a bias to assume that others are lying (one that pays off when others are, in fact, lying). We addressed this issue in a study in which 254 adults had to determine whether people in videos were lying or telling the truth about their experiences. Contrary to the predictions of SDT, highly anxious people did not have a heightened ability to separate lies from truths, but were biased to assume that others were lying regardless of the authenticity of their statements.