Search

Interviewees sometimes deliberately omit reporting some information. Such omission lies differ from other lies because all the information interviewees present may be entirely truthful. Truth tellers and lie tellers carried out a mission. Truth tellers reported the entire mission truthfully. Lie tellers were also entirely truthful but left out one element of the mission. In truth tellers' statements, only the parts that lie tellers were also asked to recall were analysed. Interviews were carried out via the Cognitive Credibility Assessment, Reality Interview, or standard interview protocol. Dependent variables were the details, complications and verifiable sources interviewees reported. A questionnaire measured three deception strategies: 'Tell it all', 'keep it simple' or 'paying attention to demeanour'. Lie tellers reported fewer details, complications and verifiable sources than truth tellers and reporting these variables was negatively correlated with the 'keep it simple' and 'demeanour' strategies. The type of interview protocol did not affect the results.

Past research explored the relationship between personality, moral disengagement, and deception and found a general trend showing that the lower people score on the big five personality factors, but the higher they score on moral disengagement and Machiavellianism, the higher their lying tendency. However, a limitation of past research is that it has usually adopted a variable-centred approach, whereas a person-centred approach might describe people in more detail and provide further insight into the relationship between personality and morality. In the present study, we collected data from 316 participants and asked them to fill an on-line questionnaire which included measures on personality, moral disengagement, and lying tendency (perceived lying ability, frequency, negativity and contextuality). The latter was measured via the newly developed Structure of Deception (SoD) scale (Makowski et al., Current Psychology, 2021 (See CR30)). We had to aims. First, to validate an Italian version of the SoD, which showed a good factor structure, gender measurement invariance, and good construct and criterion validity. Second, to explore the association between personal characteristics and lying tendency. Personality and morality scores were combined to obtain subpopulations of participants by a mean of cluster analysis. We obtained four clusters, one of which was marked by high Machiavellianism and moral disengagement but low scores on the personality factors, and one of which showed the opposite trend. The results also showed that cluster membership, and hence personal characteristics, was associated with lying tendency. The person-centred approach can be applied in research on lying. Limitations of the study and future suggestions are also discussed., Author(s): Nicola Palena [sup.1] , Letizia Caso [sup.2] , Lucrezia Cavagnis [sup.1] , Andrea Greco [sup.1] , Aldert Vrij [sup.3] Author Affiliations: (1) grid.33236.37, 0000000106929556, Department of Human and Social [...]

Sketching while narrating involves describing an event while sketching on a blank paper (self-generated sketch) or on a printed map. We compared the effects of self-generated sketches and printed maps on information elicitation and lie detection. Participants (N = 211) carried out a mock mission and were instructed to tell the truth or to lie about it in an online interview. In the first phase of the interview, all participants provided a free recall. In the second phase, participants provided another free recall or verbally described the mission while sketching on a blank paper or on a printed map. Truth tellers provided richer accounts than lie tellers. Larger effect sizes emerged for the self-generated sketch condition than for the printed map and free recall conditions. This suggests that self-generated sketches are more effective lie detection tools when information on routes and locations is sought.

In a comparable truth baseline (CTB), a knowingly truthful baseline statement is compared to a statement of interest, and deviations in verbal details possibly indicate deceit. In two experiments, we investigated whether a CTB can improve truth/lie discrimination when verbal details are coded by independent raters (Experiment 1) and when judged by naive observers (Experiment 2). In addition, we investigated whether lie tellers would calibrate their lies to match the detailedness of their baseline. Results showed no evidence of calibration. As expected, truths were more detailed than their corresponding baselines, while lies were less detailed. Significant differences emerged for spatial, visual and action details. Experiment 2 did not show that a CTB improved observers' lie detection accuracy. Taken together, our results showed that a deviation in details from a CTB may serve as a helpful aid in lie detection.

The Devil's Advocate protocol has been developed to assist making veracity assessments when someone discusses their opinion. The present experiment focused on protester actions rather than controversial issues and also included an adapted version of the Verifiability Approach. Participants told the truth or lied about protester actions and the participants' answers to the eliciting opinion and Devil's Advocate questions were compared. The Devil's Advocate approach predicts the difference in answers (residue scores) to be more pronounced in truth tellers than in lie tellers in terms of quantity of the answers (number of words, details and arguments) and quality of the answers (plausibility, immediacy and clarity). The hypothesis was supported but only in terms of quality: Truth tellers' answers sounded more plausible and immediate and somewhat clearer than lie tellers' answers. Truth tellers also reported more digital verifiable sources than lie tellers.

Research has repeatedly shown that accusatory questions posed during an investigative interview are indicative of biased beliefs about suspect guilt. Linguistic research has shown that the verbs used in utterances can be indicative of biased beliefs about another person. In the present study we examined question type and the verbs used in question formulation using non-police participants to explore the influence of guilt presumption on interview questions. In Study1 we used the Linguistic Category Model (LCM; Semin and Fiedler European Review of Social Psychology, 2, 1-30, 1991 (See CR33)) and in Study 2, the Question-Answer Paradigm (QAP; Semin et al. Personality and Social Psychology Bulletin, 21, 834-841, 1995 (See CR34)) to analyse verb abstraction and positive/ negative valence of the formulated interview questions. We also explored whether interviewers' guilt presumptions changed over the course of the interview as well as their motivations for creating the questions they chose to ask (Study 2). We found that participants who presumed guilt were more likely to formulate accusatory questions and use a higher verb abstraction with negative valence (Study 1 and 2). Interviewers asked more questions to gather additional information overall; however, the number of questions was negligible for trying to find support for alternative scenarios or to falsify existing guilt beliefs (Study 2). Interviewers who presumed guilt were also less likely to change their views during the interview and were more likely to report using behavioural cues to solidify their guilt presumptions (Study 2). The overall findings are in line with previous research in both guilt presumptive interviewing and linguistically biased language; however, we expanded on previous research by allowing participants to come to their own conclusions regarding guilt, as well as formulating their own questions for the suspect. Finally, we conclude that there are extensive limitations for using the LCM in applied interview settings and these are discussed., Author(s): Nicole M. Adams-Quackenbush [sup.1] [sup.2] [sup.3] , Aldert Vrij [sup.3] , Robert Horselenberg [sup.2] , Liam P. Satchell [sup.4] , Peter van Koppen [sup.2] [sup.5] Author Affiliations: (1) grid.6374.6, [...]

The current experiment examined the effects of combining two interview techniques on information elicitation and lie detection in multiple interviews. Participants were interviewed online over three weeks. Two-thirds of the participants (n = 114) were presented with the model statement (MS) interview technique in Interview 1 and were asked to sketch and narrate in Interview 2 (MS-Sketch condition) or vice versa (Sketch-MS condition). In Interview 3, all participants provided a free recall. The remaining one-third of participants (n = 61) provided only a free recall in the three interviews. Truth tellers reported more information than lie tellers in all three interviews. In Interview 2, the Sketch-MS condition elicited more new information and cues to deceit than the other conditions. The results imply that veracity effects are consistent across multiple interviews and the MS is effective for eliciting information and cues to deceit when preceded by an initial recall.

Research has shown that complications are more common in truth tellers' accounts than in lie tellers' accounts, but there is currently no experiment that has examined the accuracy of observers' veracity judgments when looking at complications. A total of 87 participants were asked to judge 10 transcripts (five truthful and five false) derived from a set of 59 transcripts generated in a previous experiment by Deeb et al. Approximately half of the participants were trained to detect complications (Trained), and the other half did not receive training (Untrained). Trained participants were more likely to look for complications, but they did not detect them accurately, and thus their veracity judgments did not improve beyond Untrained participants' judgments. We discuss that the training may have been too brief or not sensitive enough to enhance decision-making. [ABSTRACT FROM AUTHOR]

Behaviour detection officers' task is to spot potential criminals in public spaces, but scientific research concerning what to look for is scarce. In two experiments, 52 (Experiment 1A) and 60 (Experiment 2A) participants carried out a mission involving a ferry crossing. Half were asked to smuggle an object; the other half were non-smugglers. In Experiment 2A, two confederates appeared to approach as if looking for someone on the ferry. Smugglers, more than non-smugglers, reported afterwards to have felt nervous, self-conscious, and conspicuous and to attempt behavioural control during the ferry crossing. The secretly videotaped ferry crossings were shown to 104 (Experiment 1B) and 120 (Experiment 2B) observers, tasked to identify the smugglers. Although they reported paying attention mostly to signs of nervousness, lie detection accuracy rate was poor (48% in Experiment 1 and 39.2% in Experiment 2) because their perceptions of nervousness did not match the experiences of nervousness reported by the (non)smugglers.

Sham marriages occur frequently, and to detect them, partners are sometimes interviewed together. We examined an innovative method to detect deceit in such interviews. Fifty-three pairs of interviewees, either friends (truth tellers) or pretended to be friends (liars), were interviewed about their friendship. Just before the interview, they received the questions that would be asked in the interview and were invited to prepare the answers. We told them that these preinterview deliberations would be recorded. Based on the transcripts, we analysed cues to truthfulness (cues expected to be expressed more by truth tellers) and cues to deceit (cues expected to be expressed more by liars). Truth tellers and liars differed from each other, particularly regarding expressing cues to truthfulness. Preinterview deliberations that are recorded with awareness of the interviewees can be used for lie detection purposes. We discuss further venues in this new line of research.

Due to time constraints, interviews aimed to detect deception in airport settings should be brief and veracity assessments should be made in real time. In two experiments carried out in the departure hall of an international airport, truth tellers were asked to report truthfully their forthcoming trip, whereas liars were asked to lie about the purpose of their trip. In Experiment 1, we examined five verbal cues that we thought had potential to discriminate truth tellers from liars in short airport interviews: elaboration in disclosing information, checkable details, how many people are aware of the trip, complications, and plausibility. In Experiment 2, we attempted to improve the interview protocol by adding a second interview phase in which we introduced an information protocol and model statement. All five cues differentiated truth tellers from liars in both experiments, but the information protocol and model statement did not enhance these differences.

In none of the deception studies that used drawings to date, was the effect of sketching on both speech content and drawing content examined, making it unclear what the full potential is of the use of drawings as a lie detection tool. A total of 122 truth tellers and liars took part in the study who did or did not sketch while narrating their allegedly experienced event. We formulated hypotheses about the total amount of information and number of complications reported and about various features of the drawings. Participants in the Sketch-present condition provided more information than participants in the Sketch-absent condition, and truth tellers reported more details than liars, but only in the Sketch-present condition. In contrast to previous research, no Veracity differences occurred regarding the content of the drawings, perhaps because sketching was introduced as a tool that facilitated verbal recall and not as a stand-alone tool.

When given the opportunity, liars will embed their lies into otherwise truthful statements. In what way this embedding affects the quality of lies, however, remains largely unknown. This study investigated whether lies that are embedded into truthful stories are richer in detail and contain higher quality details compared to lies that are part of entirely fabricated statements. Participants (N = 111) were asked to provide a statement that was either entirely truthful, entirely fabricated, or had the fabricated element of interest embedded into an otherwise truthful story. Results indicated that lies embedded in a fabricated statement are not qualitatively different from lies embedded in an otherwise truthful statement. Supporting Bayes factors provided moderate to strong evidence for this conclusion. Accordingly, verbal credibility assessment tools based on the verbal content measured in this study may be robust against the embedding of lies.

Research has attempted to explain perceived cues to deception based upon self-report of what participants believe are 'good' cues to deception, or self-report of what cues participants say they base their veracity judgements on. However, it is not clear to what extent participants can accurately self-report what influences their decision-making. Using a within-subjects design, 285 participants completed a questionnaire regarding their beliefs about deception before rating a selection of truthful and deceptive statements on a variety of cues. Expert coders also rated the statements for the same cues. Laypeople and expert coders do not conceptualise between-subject consistency in the same way. A lens model showed that whilst perceptions of cues, such as consistency and amount of detail, influence veracity judgements, these perceptions (and overall veracity judgements) are mostly inaccurate. Fundamentally, there seems to be inconsistencies between how deception research examines consistency and how it is understood and used by laypeople.

In information gathering interviews, follow-up questions are asked to clarify and extend initial witness accounts. Across two experiments, we examined the efficacy of open-ended questions following an account about a multi-perpetrator event. In Experiment 1, 50 mock-witnesses used the timeline technique or a free recall format to provide an initial account. Although follow-up questions elicited new information (18-22% of the total output) across conditions, the response accuracy (60%) was significantly lower than that of the initial account (83%). In Experiment 2 (N = 60), half of the participants received pre-questioning instructions to monitor accuracy when responding to follow-up questions. New information was reported (21-22% of the total output) across conditions, but despite using pre-questioning instructions, response accuracy (75%) was again lower than the spontaneously reported information (87.5%). Follow-up open-ended questions prompt additional reporting; however, practitioners should be cautious to corroborate the accuracy of new reported details.

The self-administered interview (SAI) is a written eyewitness recall tool that elicits more information from cooperative witnesses than written free recall (WFR) formats. To date, SAI research has examined the accounts of cooperative people providing honest reports. In the current experiment, truthful and fabricating participants (N = 128) either completed a WFR or a SAI after witnessing a crime (initial account). After a 1-week delay, participants were interviewed verbally (subsequent interview). Truth tellers reported significantly more detail than liars in both the initial account and subsequent interview, and participants who completed the SAI reported more detail than those completing the WFR. Truth tellers repeated and omitted more information in the subsequent interview than liars; however, there was no significant difference in the number of reminiscent details reported. Although the SAI is effective in eliciting information as an initial eyewitness reporting tool, no benefits for the detection of deception were demonstrated.

In this article, I present my view on the significant developments and theoretical/empirical tipping points in nonverbal and verbal deception and lie detection from the last 30 years and on prospects for future research in this domain. I discuss three major shifts in deception detection research: (a) From observing target persons' nonverbal behavior to analyzing their speech; (b) from lie detection based on differences between truth tellers and liars' levels of arousal to lie detection based on the different cognitive processes or strategies adopted to appear convincing; and (c) from passively observing target persons to actively interviewing them to elicit or enhance verbal cues to deceit. Finally, I discuss my ideas for future research, focusing on initiatives from my own lab. Hopefully, this will stimulate other researchers to explore innovative ideas in the verbal deception research domain, which already has seen so much progress in the last decade.

We examined the effects of cognitive load on the strategy selection in the forced choice test (FCT) when used to detect hidden crime knowledge. Examinees (N = 120) with and without concealed knowledge from a mock crime were subjected to an FCT either under standard circumstances or cognitive load. Cognitive load was implemented through time pressure. The FCT distinguished examinees with concealed knowledge from those without better than chance in both conditions, but the counterstrategies did not differ between conditions. Further investigation revealed that time pressure did affect examinees' ability to follow their intended counterstrategy to produce randomized test patterns, which constitutes an effective counterstrategy in the FCT. Hence, time pressure lowered the success rate of effective counterstrategies, but not their incident rates. Further disambiguation of various cognitive load manipulations and their effects on strategy selection and execution is needed.

Leal, Vrij, Deeb, and Jupe (2018) found--with British participants--that a model statement elicited (a) more information and (b) a cue to deceit: After exposure to a model statement, liars reported significantly more peripheral information than truth tellers. We sought to replicate these findings with Arabs living in Israel. Truth tellers and liars reported a stand-out event that they had (truth tellers) or pretended to have (liars) experienced in the last 2 years. Half of the participants were given a model statement in the second phase of the interview. Replicating Leal et al. (2018a), (a) truth tellers reported more core details than liars initially and (b) a model statement resulted in more additional core and peripheral details in the second phase of the interview. Unlike in Leal et al. (2018a), a model statement did not have a differential effect on truth tellers in the current experiment.

As interviewees typically say less when an interpreter is present, we examined whether this was caused by interpreters not interpreting everything interviewees say or by interviewees providing less information. We further examined (a) the effect of a model drawing on providing information and (b) the diagnostic value of total details and the proportion of complications as cues to deceit. Hispanic, Russian, and South Korean participants were interviewed by native interviewers or by a British interviewer through an interpreter. Truth tellers discussed a trip they had made; liars fabricated a story. Participants received no instruction (condition 1) or were instructed to sketch while narrating without (condition 2) or with (condition 3) being given examples of detailed sketches. Interviewees said less when an interpreter was present because they provided less information. Truth tellers gave more details and, particularly, obtained a higher proportion of complications than liars. The sketching manipulation had no effect.

We examined the efficacy of a Model Statement to detect opinion lies. A total of 93 participants discussed their opinion about the recent strikes on two occasions, 1 week apart. In one interview they told the truth and in the other interview they lied. Each interview consisted of two phases. In Phase 1 they discussed their alleged opinion (truth or lie as appropriate). They then either listened to a Model Statement (a detailed account of someone discussing an opinion about a topic unrelated to strike actions) and expressed their opinion again in Phase 2 (Model Statement present condition) or they discussed their opinion again without listening to a Model Statement (Model Statement absent condition). The verbal cues examined were pro‐opinion arguments, anti‐opinion arguments, plausibility, immediacy, directness, clarity, and predictability. The truthful statements sounded more plausible in Phases 1 and 2 than the deceptive statements, providing further evidence that plausibility is a strong veracity indicator. The truthful statements included more pro‐arguments and sounded more immediate and direct than the deceptive statements, but only in Phase 2. The Model Statement had no effect. Reasons for the Model Statement null‐effect are discussed. [ABSTRACT FROM AUTHOR]

People sometimes lie by omitting information. The information lie tellers then report could be entirely truthful. We examined whether the truthful information that lie tellers report in omission lies contains verbal cues indicating that the person is lying. We made a distinction between (i) essential information (events surrounding the omission) and non‐essential information (the rest); and (ii) made a distinction between informing or not informing participants about the key event they witnessed. Participants followed a target person. Truth tellers reported all activities truthfully; lie tellers omitted the key event. Participants were or were not informed what this key event was. In the analyses we discarded the information truth tellers reported about the key event lie tellers omitted. Truth tellers reported more external and contextual details, more complications and fewer common knowledge details and self‐handicapping strategies than lie tellers, but only when discussing essential information. Being informed had no effect. [ABSTRACT FROM AUTHOR]

The current study examined the validity of the forced choice test (FCT) in a forensic scenario when used to detect concealment of semantic memory (SM-FCT). We also compared the SM-FCT validity to the FCT validity in the more commonly investigated episodic memory scenario (EM-FCT). In simulating a scenario of investigating suspected members of a terror organization, 277 students were asked to deceptively deny being enrolled in a college in which they do actually study. Results indicated that the SM-FCT's validity level was within the range of the EM-FCTs' validity levels. Theoretically, the results support a cognitive-based explanation for the FCT operation mechanism. Practically, they imply that FCT can be used in criminal or intelligence investigations of suspected members of terrorist or criminal organizations or suspected perpetrators of illegal acts or acts of terrorism, in which the incriminating evidence being sought is in the realm of designated semantic memory or knowledge. [ABSTRACT FROM AUTHOR]

Studies on the collection of information (intelligence) show the effectiveness of the Scharff technique and the principle of reciprocity. These two aspects have never been considered at the same time. In the present experiment, we compared the effect of two different forms of reciprocity (intrinsic vs. instrumental) and their combination on the elicitation of information in Scharff interviews. Participants were asked to imagine that an interview took place and were exposed to four reciprocity conditions (no reciprocity, intrinsic reciprocity, instrumental reciprocity, intrinsic and instrumental reciprocity). The results showed that instrumental reciprocity resulted in more information than intrinsic reciprocity. [ABSTRACT FROM AUTHOR]

We examined whether speech-related differences between truth tellers and liars are more profound when answering unexpected questions than when answering expected questions. We also examined whether the presence of an interpreter affected these results. In the experiment, 204 participants from the United States (Hispanic participants only), Russia, and the Republic of Korea were interviewed in their native language by a native-speaking interviewer or by a British interviewer through an interpreter. Truth tellers discussed a trip that they had made during the last 12 months; liars fabricated a story about such a trip. The key dependent variables were the amount of information provided and the proportion of all statements that were complications. The proportion of complications distinguished truth tellers from liars better when answering unexpected than expected questions, but only in interpreter-absent interviews. The number of details provided did not differ between truth tellers and liars or between interpreter-absent and interpreter-present interviews.

This study examines counter-interrogation strategies employed by liars giving false alibis. Participants (N = 144) visited a restaurant to buy a sandwich (truth-tellers) or to use it as a false alibi (liars). Half of the liars were informed they might be asked for a drawing of the alibi setting if interviewed (informed liars). Participants spent either 10 min (high familiarity condition) or 30 s (low familiarity condition) in the restaurant. All participants were asked to provide two visuospatial statements, which were assessed for salient details, nonsalient details, between-statement consistency, and statement-alibi setting consistency. Informed liars provided significantly more salient and nonsalient details than uninformed liars and truth-tellers, particularly in the high familiarity condition. No differences emerged for statement consistency types. The results suggest that liars are more concerned than truth-tellers about making a positive impression on the interviewer, and they fail to accurately reflect on truth-tellers' visuospatial statements.

The current study was to test whether reality monitoring and language use could distinguish identity liars from truth tellers when answering outcome questions and unexpected process questions. Truth tellers (n = 30) and liars (n = 30) discussed their identity in a recruitment interview. No differences emerged between truth tellers and liars in the details they provided. In terms of language use, liars used more positive language than truth tellers, whereas truth tellers used more cognitive process words than liars. However, neither were more pronounced when asking process questions. Overall, process questions elicited more cognitive process and cause words than outcome questions. Therefore, process questions may be able to contribute to the cognitive load approach. The findings suggest that reality monitoring may not be diagnostic when applied to identity deception. We discuss the language use differences in relation to impression management theory.

We experimentally investigated how different mnemonic techniques employed in an interview conducted immediately after an event affected truth tellers' and liars' responses when they were interviewed again after a 2-week delay. We also compared how verbal accounts changed over time within truth tellers and liars, and how consistent both groups were. Participants (n = 143) were shown a mock intelligence operation video and instructed either to tell the truth or lie about its contents in two interviews, one of which was immediately after watching the video and the other after a 2-week delay. In the immediate interview, they were asked to provide a free recall and then asked to provide further information via one of three mnemonics: context reinstatement, sketch, or event-line. In the delayed interview, they were asked to provide only a free recall. Truth tellers reported more visual, spatial, temporal, and action details than did liars both immediately and after a delay. Truth tellers experienced more of a decline in reporting details after a delay than did liars, and this decline was affected by the mnemonic used. Truth tellers thus showed, more than liars, patterns of reporting indicative of genuine memory decay. Liars produced patterns of a "stability bias" instead. Truth tellers and liars were equally consistent between their immediate and delayed statements.

The verifiability approach (VA) is a lie-detection tool that examines reported checkable details. Across two studies, we attempt to exploit liar's preferred strategy of repeating information by examining the effect of questioning adult interviewees before the VA. In Study 1, truth tellers (n = 34) and liars (n = 33) were randomly assigned to either an initial open or closed questioning condition. After initial questioning, participants were interviewed using the VA. In Study 2, truth tellers (n = 48) and liars (n = 48) were interviewed twice, with half of each veracity group randomly assigned to either the Information Protocol (an instruction describing the importance of reporting verifiable details) or control condition. Only truth tellers revised their initial statement to include verifiable detail. This pattern was most pronounced when initial questioning was open (Study 1) and when the information protocol was employed (Study 2). Thus, liar's preferred strategy of maintaining consistency between statements appears exploitable using the VA.

In forced-choice tests (FCTs), examinees are typically presented with questions with two equally plausible answer alternatives, of which only one is correct. The rationale underlying this test is that guilty examinees tend to avoid relevant crime information, producing a nonrandom response pattern. The validity of FCTs is reduced when examinees are informed about this underlying rationale, with coached guilty examinees refraining from avoiding the correct information but trying to provide a random mix of correct and incorrect answers. To detect such intentional randomization, a "runs" test--looking at the distribution of the number of alternations between correct and incorrect answers--has been suggested but with limited success. We designed a runs test based on distinguishing between patterns that look random and patterns that are random. Specifically, we alternated the horizontal presentation (i.e., presentation left or right on the screen) of the correct answer alternative between each trial. As a consequence, guilty examinees were faced with having to choose to randomize either between correct and incorrect answers--leading to chance performance--or between answers presented on the left or right, producing a pattern that "looks" random. As innocent examinees are unaware of the correct answers, they can only randomize between horizontal positions. Results showed that "the number of correct items" selected distinguished guilty from innocent examinees only when they were not informed about the underlying rationale. In contrast, "alternations between correct and incorrect answers" did distinguish informed guilty from innocent examinees. Incremental validity of the alternation criterion and theoretical implications are discussed.

Recently, verbal credibility assessment has been extended to the detection of deceptive intentions, the use of a model statement, and predictive modeling. The current investigation combines these 3 elements to detect deceptive intentions on a large scale. Participants read a model statement and wrote a truthful or deceptive statement about their planned weekend activities (Experiment 1). With the use of linguistic features for machine learning, more than 80% of the participants were classified correctly. Exploratory analyses suggested that liars included more person and location references than truth-tellers. Experiment 2 examined whether these findings replicated on independent-sample data. The classification accuracies remained well above chance level but dropped to 63%. Experiment 2 corroborated the finding that liars' statements are richer in location and person references than truth-tellers' statements. Together, these findings suggest that liars may over-prepare their statements. Predictive modeling shows promise as an automated veracity assessment approach but needs validation on independent data.