Deception Detection Research Articles

Veracity judgement, not accuracy: Reconsidering the role of facial expressions, empathy, and emotion recognition training on deception detection.

People hold strong beliefs about the role of emotional cues in detecting deception. While research on the diagnostic value of such cues has been mixed, their influence on human veracity judgements is yet to be fully explored. Here, we address the relationship between emotional information and veracity judgements. In Study 1, the role of emotion recognition in the process of detecting naturalistic lies was investigated. Decoders’ veracity judgements were compared based on differences in trait empathy and their ability to recognise microexpressions and subtle expressions. Accuracy was found to be unrelated to facial cue recognition and negatively related to empathy. In Study 2, we manipulated decoders’ emotion recognition ability and the type of lies they saw: experiential or affective (emotional and unemotional). Decoders received either emotion recognition training, bogus training, or no training. In all scenarios, training did not affect veracity judgements. Experiential lies were easier to detect than affective lies; however, affective unemotional lies were overall the hardest to judge. The findings illustrate the complex relationship between emotion recognition and veracity judgements, with abilities for facial cue detection being high yet unrelated to deception accuracy.

The verifiability approach to deception detection: A preregistered direct replication of the information protocol condition of Nahari, Vrij, and Fisher (2014b)

SummaryNahari, Vrij, and Fischer [(2014b), Applied Cognitive Psychology, 28, 122–128] found that, when participants were forewarned that their statements would be checked for verifiable details, truth tellers gave much more verifiable details than liars. In this direct replication (n = 72), participants wrote a statement claiming they had carried out their regular campus activities, whereas liars had actually stolen an exam. Statements were coded for verifiable details. Our primary prediction was confirmed: Truth tellers provided significantly more verifiable details than liars. Of note, the replication effect size (d = 0.49) was less than half that of the original (d = 1.14), and – like in the original study – was smaller than the lie‐truth effect size for total details (verifiable and unverifiable details combined; d = 0.80). We hope this will stimulate other independent investigations of VA to tell whether or not coding for verifiability will pass Ockham's razor test.

Acoustic-Prosodic and Lexical Cues to Deception and Trust: Deciphering How People Detect Lies

Humans rarely perform better than chance at lie detection. To better understand human perception of deception, we created a game framework, LieCatcher, to collect ratings of perceived deception using a large corpus of deceptive and truthful interviews. We analyzed the acoustic-prosodic and linguistic characteristics of language trusted and mistrusted by raters and compared these to characteristics of actual truthful and deceptive language to understand how perception aligns with reality. With this data we built classifiers to automatically distinguish trusted from mistrusted speech, achieving an F1 of 66.1%. We next evaluated whether the strategies raters said they used to discriminate between truthful and deceptive responses were in fact useful. Our results show that, although several prosodic and lexical features were consistently perceived as trustworthy, they were not reliable cues. Also, the strategies that judges reported using in deception detection were not helpful for the task. Our work sheds light on the nature of trusted language and provides insight into the challenging problem of human deception detection.

Deception in the eyes of deceiver: A computer vision and machine learning based automated deception detection

There is growing interest in the use of automated psychological profiling systems, specifically applying machine learning to the field of deception detection. Several psychological studies and machine-based models have been reporting the use of eye interaction, gaze and facial movements as important clues to deception detection. However, the identification of very specific and distinctive features is still required. For the first time, we investigate the fine-grained level eyes and facial micro-movements to identify the distinctive features that provide significant clues for the automated deception detection. A real-time deception detection approach was developed utilizing advanced computer vision and machine learning approaches to model the non-verbal deceptive behavior. Artificial neural networks, random forests and support vector machines were selected as base models for the data on the total of 262,000 discrete measurements with 1,26,291 and 128,735 of deceptive and truthful instances, respectively. The data set used in this study is part of an ongoing programme to collect a larger dataset on the effects of gender and ethnicity on deception detection. Some observations are made based on this data which should not be interpreted as scientific conclusions, but pointers for future work. Analysis of the above models revealed that eye movements carry relatively important clues to distinguish truthful and deceptive behaviours. The research outcomes align with the findings from forensic psychologists who also reported the eye movements as distinctive for the truthful and deceptive behavior. The research outcomes and proposed approach are beneficial for human experts and has many applications within interdisciplinary domains.

LieToMe: An Ensemble Approach for Deception Detection from Facial Cues.

Deception detection is a relevant ability in high stakes situations such as police interrogatories or court trials, where the outcome is highly influenced by the interviewed person behavior. With the use of specific devices, e.g. polygraph or magnetic resonance, the subject is aware of being monitored and can change his behavior, thus compromising the interrogation result. For this reason, video analysis-based methods for automatic deception detection are receiving ever increasing interest. In this paper, a deception detection approach based on RGB videos, leveraging both facial features and stacked generalization ensemble, is proposed. First, a face, which is well-known to present several meaningful cues for deception detection, is identified, aligned, and masked to build video signatures. These signatures are constructed starting from five different descriptors, which allow the system to capture both static and dynamic facial characteristics. Then, video signatures are given as input to four base-level algorithms, which are subsequently fused applying the stacked generalization technique, resulting in a more robust meta-level classifier used to predict deception. By exploiting relevant cues via specific features, the proposed system achieves improved performances on a public dataset of famous court trials, with respect to other state-of-the-art methods based on facial features, highlighting the effectiveness of the proposed method.

Deception detection and emotion recognition: Investigating F.A.C.E. software

Objective: Investigation of deception within psychotherapy has recently gained attention. Micro expression training software has been suggested to improve deception detection and enhance emotion recognition. The current study examined the effects of micro expression training software on deception detection and emotion recognition. Method: The current study recruited 23 counseling psychology graduate students and 32 undergraduate students and randomly assigned them to either a training group or a control group. The training and control group received all the same materials and measures pre- and post-test, with the training group differing by receiving the micro expression training. Results: Findings revealed no significant difference in deception detection between the control group and training group. The training did reveal significant improvement for emotion recognition, specifically in contempt, anger, and fear. State and trait anxiety did not predict deception detection nor did it mediate emotion recognition. No significant difference was found between graduate trainees and undergraduate students. Conclusion: The use of the F.A.C.E. software was not effective for increasing deception detection but did serve to increase emotion recognition. Implications for training, practice, and research are discussed.

Seeing the Bigger Picture: Susceptibility to, and Detection of, Deception.

An extended time window was used to examine susceptibility to, and detection of, deception in rugby union. High- and low-skilled rugby players judged the final running direction of an opponent "cutting" left or right, with or without a deceptive sidestep. Each trial was occluded at one of eight time points relative to the footfall after the initial (genuine or fake) reorientation. Based on response accuracy, the results were separated into deception susceptibility and deception detection windows. Signal-detection analysis was used to calculate the discriminability of genuine and deceptive actions (d') and the response bias (c). High-skilled players were less susceptible to deception and better able to detect when they had been deceived, accompanied by a reduced bias toward perceiving all actions as genuine. By establishing the time window in which players become deceived, it will now be possible to identify the kinematic sources that drive deception.

Are criminals better lie detectors? Investigating offenders' abilities in the context of deception detection

SummaryThe present research examined lie detection abilities of a rarely investigated group, namely offenders. Results of the studies conducted thus far indicated a better performance of offenders compared to non‐offenders when discriminating between true and false messages. With two new studies, we aimed at replicating offenders' superior abilities in the context of deception detection. Results of Study 1 (N = 76 males), in contrast, revealed that offenders were significantly worse at accurately classifying true and false messages compared to non‐offenders (students). Results of Study 2 (N = 175 males) revealed that offenders' discrimination performance was not significantly different compared to non‐offenders (clinic staff). An internal meta‐analysis yielded no significant difference between offenders and non‐offenders, questioning the generalizability of previous findings.

Can Emotional Awareness of Liars Influence Deception Detection Effectiveness?

Emotional awareness is defined as the ability to cognitively process emotional arousal and the expression thereof. When telling a lie, emotional awareness comes into play: the most important objective is to conceal one’s true emotions and fabricate false ones; simultaneously, however, one must control for the affective state of those who are to believe the falsehood; via such efforts, one can assess the potential for success in the deceit. With this in mind, we hypothesized that emotional awareness may play a vital role in the process of creating a convincing lie. Study participants (Group A, N = 40) were asked to complete the Polish adaptation of the Levels of Emotional Awareness Scale (LEAS) and record videos consisting of plotting some truth or lie, which were then rated as true or false by 400 volunteers (Group B). Both samples (Group A and Group B) were recruited among students attending Polish universities. The results allowed us to confirm correlational relationships between emotional awareness (general, self-awareness, and awareness of others’ emotions) and the effectiveness of the deception. We were also able to confirm previous studies’ results about the truth bias in detecting lies and gender differences in emotional awareness.

Deception detection in repeated interviews: The effects of immediate type of questioning on the delayed accounts

AbstractIn this study, we examined how different types of interviewing (eliciting more complete vs. less complete accounts) used in an interview conducted shortly after an event affected truth tellers' and liars' responses when they were interviewed again after a two‐week delay. Participants (n = 80) were shown a mock intelligence operation video and told either the truth or lied about its contents in two interviews, immediately after watching the video, and after a two‐week delay. In the immediate interview participants were instructed either to report everything they remembered, or asked spatial questions related to the event. In the delayed interview, all participants were asked to report everything. The differences between truth tellers and liars were slightly larger in the report everything than in the spatial questions condition. Results suggest that an immediate “report everything” instruction can aid to effectively discriminate between truthful and deceptive accounts.

认知负荷取向下基于记忆-反应冲突的欺骗检测

认知负荷取向下基于记忆-反应冲突的欺骗检测

A review of truth‐default theory: Implications for information behavior research

AbstractDetermining truth and accuracy of information is a key challenge in today's fast‐paced, global information economy dominated by social media. The field of Information Science, while publishing extensively on information seeking and use, has not done sufficient research into how individuals detect falsehood or deception in information they encounter. This paper describes Levine's Truth‐Default Theory (TDT) and links information behavior (IB) research to three vital aspects of the theory. Furthermore, this work demonstrates how TDT can be merged with T.D. Wilson's General Theory of Information Behavior, applying decades of research on deception detection to foundational IB theory. Implications of marrying these two ideas are discussed as well as suggestions for future research.

The Impact of Key Ideas on Automatic Deception Detection in Text

In recent years, with the rise of the Internet, the automatic deception detection in text is an important task to recognize those of documents that try to make people believe in something false. Current studies in this field assume that the entire document containscues to identify deception; however, as demonstrated in this work, some irrelevant ideas in text could affect the performance of the classification. Therefore, this research proposes an approach for deception detectionin text that identifies, in the first instance, key ideas in a document based on a topic modeling algorith mand a proposed automatic extractive text summarization method, to produce a synthesized document that avoids secondary ideas. The experimental results of this study indicate that the proposed method outperform previous methods with standard collections.

A Novel Hybrid Network Model Based on Attentional Multi-Feature Fusion for Deception Detection

A Novel Hybrid Network Model Based on Attentional Multi-Feature Fusion for Deception Detection

A trifactorial model of detection of deception using thermography

ABSTRACT Most theories of detection of deception relate lying to three factors: (1) cognitive load, (2) subjective arousal, and (3) convincing impression. The evidence suggests that a higher temperature of the forehead is related to cognitive load, a lower temperature of the nose is associated with subjective arousal, and a higher temperature of the cheeks is related to convincing impression. Here, we took into account these three factors and, at the same time, associated the thermal change in specific facial regions of interest (ROIs) with each one of them. More importantly, we studied the combination of the thermal changes in the ROIs to establish the best combination to detect deception. Our results confirm an association between thermal changes in different ROIs and the three factors above. The best combination in the thermal changes of the ROIs for detecting deception (producing 83% accuracy and 13% false alarms in Experiment 1) is the one that was termed ‘at least two of the three ROIs’ where there is a lower temperature of the nose and/or a higher temperature of the cheeks and/or a higher forehead temperature. This finding constitutes an advance for detecting deception in multiple forensic contexts.

Vulnerabilities to Online Social Network Identity Deception Detection Research and Recommendations for Mitigation

Identity deception in online social networks is a pervasive problem. Ongoing research is developing methods for identity deception detection. However, the real-world efficacy of these methods is currently unknown because they have been evaluated largely through laboratory experiments. We present a review of representative state-of-the-art results on identity deception detection. Based on this analysis, we identify common methodological weaknesses for these approaches, and we propose recommendations that can increase their effectiveness for when they are applied in real-world environments.

Adults' difficulties in identifying concealment among children interviewed with the putative confession instructions

SummaryThe putative confession (PC) instructions (“[suspect] told me everything that happened and wants you to tell the truth”) increases children's honesty. However, research has shown that children who maintain secrecy despite the PC are more convincing. We examined whether (a) the PC undermines adults' deception detection abilities or (b) children who conceal despite the PC are better deceivers. Adults evaluated truthful and deceptive children interviewed with the PC where the PC portion of the interview was either present or absent. Adults' deception detection was no worse when the PC was present; in fact, it was slightly better. Rather than negatively affecting adults' ability to detect deception, the PC selects an unusually convincing group of concealers.

Robust cooperative car-parking: implications and solutions for selfish inter-vehicular social behaviour

Vehicular cooperation mechanisms are known to provide efficiency and scalability benefits but for the mechanisms to be human-centric, there is a need for them to be robust and resilient to anti-social behaviours such as deception. More specifically, decentralised vehicle-to-vehicle cooperation has been shown to be an effective and convenient approach to coordinate the use of dynamically changing common road resources such as car parking. However, the potential for selfish behaviour of some vehicles in the form of sending false information for self-benefit has a significant effect on the value of cooperation. In this paper, we investigate, via extensive simulations, the deception behaviour of malicious vehicles looking to park by sending false information in decentralized vehicle cooperation. Furthermore, Deception Detection Mechanisms (DDMs) are introduced and are shown to be valuable in ameliorating the effects of malicious vehicles. The work has broader implications for an open world of autonomous and adaptive systems with decentralized control and ownership which need to cooperate to use shared resources; they are susceptible to malicious behaviour, and hence, need to be built to be robust to such behaviour.

LieToMe: Preliminary study on hand gestures for deception detection via Fisher-LSTM

Abstract The ability to discern lies, more broadly known as deception detection, is an invaluable skill that can strongly influence the outcome of relevant situations such as court trials and police interrogatories. Several devices currently exist and are being used (e.g., magnetic resonance and polygraphs) to ease those tasks; although, due to the subject awareness of such tools, their effectiveness can be compromised by the person intentional behavioural changes. Thus, alternative ways to discriminate lies without using physical devices, could become critical assets for the aforementioned situations, especially in ever improving smart cities environments. In this letter, we present an unorthodox deception detection approach, based on hand gestures found in RGB videos of famous trials. The proposed system first extrapolates hands skeletons from the RGB sequences, then computes meaningful features which are summarized into Fisher Vectors (FVs), and finally feeds this representation to a Long-Short Term Memory (LSTM) network, defined Fisher-LSTM, to try and discern if a lie is being told. In the experimental results, we show how the FV representation can help a LSTM network grasp hand gestures characteristics that could otherwise be missed. What is more, the devised Fisher-LSTM, due to its real-time computation, can be employed in smart environments as an alternative lie detector in situations requiring an immediate response, such as the aforementioned law enforcement examples.

Artificial eyespots on cattle reduce predation by large carnivores

Eyespots evolved independently in many taxa as anti-predator signals. There remains debate regarding whether eyespots function as diversion targets, predator mimics, conspicuous startling signals, deceptive detection, or a combination. Although eye patterns and gaze modify human behaviour, anti-predator eyespots do not occur naturally in contemporary mammals. Here we show that eyespots painted on cattle rumps were associated with reduced attacks by ambush carnivores (lions and leopards). Cattle painted with eyespots were significantly more likely to survive than were cross-marked and unmarked cattle, despite all treatment groups being similarly exposed to predation risk. While higher survival of eyespot-painted cattle supports the detection hypothesis, increased survival of cross-marked cattle suggests an effect of novel and conspicuous marks more generally. To our knowledge, this is the first time eyespots have been shown to deter large mammalian predators. Applying artificial marks to high-value livestock may therefore represent a cost-effective tool to reduce livestock predation.