ERPs, specifically the P3, have been proposed as an alternative to traditional polygraphy, with one approach (i.e., Brain Fingerprinting) being promoted as infallible to justify its use on a commercial basis. Concerns have been voiced, however, that such techniques would have to undergo peer-reviewed studies to satisfy validity concerns. Rosenfeld et al. (2004) found, for example, that mental countermeasures were effective in reducing detection rates using an amplitude based, peak-to-peak measure. The present study attempted to replicate and extend Rosenfeld et al.'s study, and to test Brain Fingerprinting's vulnerability to participant manipulation by employing a highly realistic virtual reality crime scenario, multiple countermeasures, and Bayesian and bootstrapping analytic approaches to classify individuals as being guilty or innocent. Participants reported a high degree of realism supporting the external validity of this study and suggesting future uses of virtual environments. Hit rates across statistical methods were significantly lower for standard guilty and innocent participants as compared to previous studies; countermeasures reduced the overall hit rates even further. Brain Fingerprinting was as vulnerable to countermeasures as other statistical measures, and produced a significant number of indeterminate outcomes. Nevertheless, innocent participants remained protected from being falsely accused across statistical methods, consistent with findings of prior studies. Reaction times were determined unsuitable in determining guilt or innocence in this study. Results suggested that ERP based deception detection measures might lack the level of validity required for use in an applied setting.

Master of Arts / Department of Communications Studies / Gregory Paul / People are sometimes deceptive, meaning that they “intentionally, knowingly, or purposefully mislead another person” (Levine, 2014, p. 37), despite potential negative relational consequences (McCornack & Levine, 1990; Millar & Tesser, 1988) or harsh societal condemnation (Kleinmuntz & Szucko, 1984). In fact, people encounter deception on a daily basis (Serota, Levine, & Boster 2010). Sometimes this deception is especially destructive (Van Swol, Braun, & Malhotra, 2012). Therefore, many law enforcement agencies and academic disciplines are invested in the study of deception. Much of this research is conducted in order to uncover ways to detect deceptive messages and distinguish them from truthful ones. This deception detection research consistently yields the following three findings. First, that people are notoriously bad at distinguishing truthful messages from deceptive ones (Bond & DePaulo, 2006). Second, that people often over-estimate their ability to detect deception accurately (Burgoon & Levine, 2010). Lastly, people are more likely to judge a message as truthful rather than deceptive regardless of the message’s veracity (Levine, Kim, Park, & Hughs, 2006). However, despite these three consistent findings, deception detection research is primarily bifurcated into two different perspectives researchers take when examining the phenomenon of deception (Burgoon & Levine, 2010) which this report labels the dominant and new perspectives. These two perspectives greatly affect the results and implications of the deception detection research being conducted. Therefore, this report examines and discusses each perspective as well as their divergent and sometimes intersecting research streams. Following this discussion, some of the most notable implications for each perspective are listed. Then some remarkable applications of each perspective are also discussed. Finally, some future research directions are suggested. Such discussions lead to an enhanced understanding of deception detection research.

It is currently unknown how lie detection accuracy is affected when someone is speaking in his or her second language. We examined whether language proficiency had an impact on lie detection. We hypothesized that when judging the veracity of second-language speakers, participants would be better able to discriminate between truth- and lie-tellers and would have bias toward picking ‘lying’ since they may display cues associated with lying when communicating.We collected video footage of native- and second-language English speakers who lied or told the truth about a transgression. Undergraduate students (N = 51) then judged the veracity of these clips and indicated how confident they were in their ratings. Participants were most accurate and confident when judging native-language truth-tellers. In addition, participants were more likely to exhibit a truth-bias when observing native-language speakers, whereas they were more likely to exhibit a lie-bias when viewing second-language speakers. Implications for the justice system will be discussed. / UOIT

Humans have long sought to use technology to augment human abilities and intellect. However, technology is traditionally employed only to create speedier solutions or more-rapid comprehension. A more challenging endeavor is to enable humans with technology to gain additional or enhanced comprehension that may not be possible to acquire otherwise. One such application is the use of technology to augment human abilities in detecting deception using nonverbal cues. Detecting deception is often critical, whether an individual is communicating with a close friend, negotiating a business deal, or screening individuals at a security checkpoint.The detection of deception is a challenging endeavor. A variety of studies have shown that humans have a hard time accurately discriminating deception from truth, and only do so slightly better than chance. Several deception detection methods exist; however, most of these are invasive and require a controlled environment.This dissertation presents a technological approach to detecting deception based on kinesic (i.e., movement-based) and vocalic (i.e., sounds associated with the voice) cues that is firmly grounded in deception theory and past empirical studies. This noninvasive approach overcomes some of the weaknesses of other deception detection methods as it can be used in a natural environment without cooperation from the individual of interest.The automatable approach demonstrates potential for increasing humans' ability to correctly identify those who display behaviors indicative of deception. The approach was evaluated using experimental and field data. The results of repeated measures analysis of variance, linear regression and discriminant function analysis suggest that the use of such a system could augment human abilities in detecting deception by as much as 15-25%. While there are a number of technical challenges that need to be addressed before such a system could be deployed in the field, there are numerous environments where it would be potentially useful.

Due to the rapid development of information technologies, especially the network technologies, business activities have never been as integrated as they are now. Business decision making often requires gathering information from different sources. This dissertation focuses on the problem of entity matching, associating corresponding information elements within or across information systems. It is devoted to providing complete and accurate information for business decision making. Three challenges have been identified that may affect entity matching performance: feature selection for entity representative, matching techniques, and searching strategy. This dissertation first provides a theoretical foundation for entity matching by connecting entity matching to the similarity and categorization theories developed in the field of cognitive science. The theories provide guidance for tackling the three challenges identified. First, based on the feature contrast similarity model, we propose a case-study-based methodology that identifies key features that uniquely identify an entity. Second, we propose a record comparison technique and a multi-layer naïve Bayes model that correspond respectively to the deterministic and the probability response selection models defined in the categorization theory. Experiments show that both techniques are effective in linking deceptive criminal identities. However, the probabilistic matching technique is preferable because it uses a semi-supervised learning method, which requires less human intervention during training. Third, based on the prototype access assumption proposed in the categorization theory, we apply an adaptive detection algorithm to entity matching so that efficiency can be greatly improved by the reduced search space. Experiments show that this technique significantly improves matching efficiency without significant accuracy loss. Based on the above findings we developed the Arizona IDMatcher, an identity matching system based on the multi-layer naïve Bayes model and the adaptive detection method. We compare the proposed system against the IBM Identity Resolution tool, a leading commercial product developed using heuristic decision rules. Experiments do not suggest a clear winner, but provide the pros and cons of each system. The Arizona IDMatcher is able to capture more true matches than IBM Identity Resolution (i.e., high recall). On the other hand, the matches identified by IBM Identity Resolution are mostly true matches (i.e., high precision).

As organizations and government agencies work diligently to detect financial irregularities, malfeasance, fraud and criminal activities through intercepted communication, there is an increasing interest in devising an automated model/tool for deception detection. We build on Pennebaker's empirical model which suggests that deception in text leaves a linguistic signature characterised by changes in frequency of four categories of words: first-person pronouns, exclusive words, negative emotion words, and action words. By applying the model to the Enron email dataset and using an unsupervised matrix-decomposition technique, we explore the differential use of these cue-words/categories in deception detection. Instead of focusing on the predictive power of the individual cue-words, we construct a descriptive model which helps us to understand the multivariate profile of deception based on several linguistic dimensions and highlights the qualitative differences between deceptive and truthful communication. This descriptive model can not only help detect unusual and deceptive communication, but also possibly rank messages along a scale of relative deceptiveness (for instance from strategic negotiation and spin to deception and blatant lying). The model is unintrusive, requires minimal human intervention and, by following the defined pre-processing steps it may be applied to new datasets from different domains. / Thesis (Master, Computing) -- Queen's University, 2007-11-28 18:10:30.45

No description available.

I utilized a design science approach to create an automated kiosk that uses embodied intelligent agents to interview individuals and detect changes in arousal, behavior, and cognitive effort by using psychophysiological information systems. This dissertation achieves three primary purposes.First, I describe the creation of this new Information Technology artifact, discuss design choices, and show the completed prototype.Second, related to this new system, I propose a unique class of intelligent agents, which are described as Special Purpose Embodied Conversational Intelligence with Environmental Sensors (SPECIES). I outline a system model that frames the conceptual components of SPECIES agents, provide design principles for developing SPECIES agents, and discuss some of the research implications of the various components in the model.Third, based on the SPECIES paradigm, I present five studies that evaluate different parts of the model. These studies form the foundational research for the development of the automated kiosk. In the first study, participants interacted with an automated interviewing agent via a chat-based modality (108 participants). The study clearly demonstrates the strong, positive correlation of both response time and the number of times a message is edited to deceitful responses. The software developed became the heart of the kiosk. The second study evaluated changing human decision-making by including influence tactics in decision aids (41 participants). This paper-based decision experiment showed that framing decision aids as appeals to individuals' values possibly change individuals' decisions and was the basis for study 4. The third study examined human-computer interaction and how SPECIES agents can change perceptions of information systems by varying appearance and demeanor (88 participants). Instantiations that had the agents embodied as males were perceived as more powerful, while female embodied agents were perceived as more likeable. Similarly, smiling agents were perceived as more likable than neutral demeanor agents. The fourth study assessed how incorporating impression management techniques into embodied conversational agents can influence human perceptions of the system (88 participants). The impression management techniques proved to be very successful in changing user perceptions. Specifically, agents that performed self-promotion were perceived as more powerful, trustworthy and expert. Agents that performed ingratiation were perceived as more attractive. In the fifth study, I used an embodied agent to interview people who had either constructed a fake bomb and packed it into a bag or had only packed clothes into a bag (60 participants). The agent used eye-tracking technology to capture pupil dilation and gaze behavior. When combined with vocal measurements, the kiosk technology was able to achieve over 93% accuracy in one trial.

False confessions and unproductive criminal investigations have resulted in misidentification of verbal and nonverbal deceptive cues. Further, the association of deceptive behavioral responses has not been confirmed based upon quantifiable graphological discrepancies. Guided by the 4-factor model for deceptive behavior, the purpose of this study was to explore the relationship between psycholinguistic cues and graphological spacing discrepancies. Handwriting samples were gathered from a stratified group of college students and law enforcement officers in rural Illinois and Tennessee (n = 113). The research was designed to determine whether graphological spacing discrepancies were evident in left margin indentions, word spacing, and sentence spacing. Two-way analyses of variance by ranks were conducted, combining these spacing discrepancies in a way to maximize the differences between the groups of truthful and deceptive statements. Through multiple regression analyses, the contributing variances were explained, as seen from participants' multiple psychological inventory scores and total spacing variances. Two-way analyses of variance were also conducted with the intent of discovering whether an interaction effect occurred, between deception-induced cognitive load and spontaneous or memory-related influences on graphological traits. Results were confirmed for statistically significant differences between truthful and deceptive sentences, containing spacing variances. Implications for positive social change include fewer false confessions during police investigations and interrogation reports with empirically based findings.

While deception is a common strategy in interpersonal communication, most research on interpersonal deception treats the sex as irrelevant in the ability to detect deceptive messages. This study examines the truth and deception detection ability of both male and female receivers when responding to both true and deceptive messages from both male and female speakers. Results suggest that sex may be an important variable in understanding the interpersonal detection probabilities of truth and of lies. An interaction of variables including speakers’ sex, receivers’ sex, and whether the message is truthful or deceptive is found to relate to detection ability. Both women and men were found to be significantly less accurate than chance in judging the veracity of statements made by men, especially when those statements are lies. On the other hand, both women and men were significantly more accurate than chance in judging the veracity of statements made by women, especially when those statements are truthful. This may suggest that men are better deceivers than women, while women seem more transparent in exhibiting feelings about their messages whether being truthful or deceptive. In recalling real life deceptions discovered previously, women reported that they discovered significantly more lies from female sources than from men they knew. This finding may reflect the previous finding that discovering lies told by women is more likely than is discovering lies told by men.