14 Detecting lies and deceit: Pitfalls and opportunities in nonverbal and verbal lie detection
Abstract: People are generally poor at detecting deceit when observing someone’s behaviour or listening to their speech. In this chapter I will discuss the major factors (pitfalls) that lead to failures in catching liars: the sixteen reasons I will present are clustered into three categories: (i) a lack of motivation to detect lies; (ii) difficulties associated with lie detection; and (iii) common errors made by lie detectors. Discussing pitfalls provides insight into how lie detectors can improve their performance (for example, by recognising common biases and avoiding common judgment errors). The second section of this chapter discusses 11 ways (opportunities) to improve lie detection skills. Within this section, I first provide five recommendations for avoiding common errors in detecting lies. Next, I discuss recent lie detection research that introduces novel interview styles aimed at eliciting and enhancing verbal and nonverbal differences between liars and truth tellers. The recommendations are relevant in various settings, from the individual level (e.g., “Is my partner really working late?”) to the societal level (e.g., “Can we trust this suspect when he claims that he is not the serial rapist the police are searching for?”).
Key Words: deception, lie detection, interviewing to detect deception, pitfalls in lie detection, opportunities in lie detection
1 Introduction
People are generally poor at detecting deceit when observing someone’s behaviour or listening to their speech. A meta-analysis of the nonverbal and verbal lie detection literature that included 206 studies involving a total of 24,483 observers revealed that these observers achieved an average accuracy rate of 54%, just greater than the rate expected by chance. Vrij (2008) examined whether such professionals as law enforcement personnel were any better at lie detection than lay-persons (typically college students). The 79 studies that included laypersons as observers and 28 studies that included professionals revealed no difference between the two groups. The laypersons’ accuracy rate was 54.27%; whereas the professionals’ rate was 55.91%.
These findings show that laypersons and professionals perform very slightly above the level of chance in lie detection when observing someone’s behaviour or listening to their speech. The situation may even be worse than it seems. Levine (2010) argues that the just above chance rates are caused by a few transparent liars, and that the majority of truth tellers and liars are detected at chance level. In this chapter I will discuss the major factors (pitfalls) that lead to failures in catching liars: the sixteen reasons I will present are clustered into three categories (Vrij 2007/2008): (i) a lack of motivation to detect lies; (ii) difficulties associated with lie detection; and (iii) common errors made by lie detectors. Discussing pitfalls provides insight into how lie detectors can improve their performance (for example, by recognising common biases and avoiding common judgment errors). The second section of this chapter discusses 11 ways (opportunities) to improve lie detection skills. Within this section, I first provide five recommendations for avoiding common errors in detecting lies. Next, I discuss recent lie detection research that introduces novel interview styles aimed at eliciting and enhancing verbal and nonverbal differences between liars and truth tellers. The recommendations are relevant in various settings, from the individual level (e.g., “Is my partner really working late?”) to the societal level (e.g., “Can we trust this suspect when he claims that he is not the serial rapist the police are searching for?”). For a more detailed discussion of the pitfalls and opportunities presented in this chapter see Vrij (2008) and Vrij, Granhag and Porter (2010).
The focus of this chapter is lie detection through observation of nonverbal and verbal cues but lies can also be detected in other ways, such as measuring people’s physiological responses (e.g., skin response heart rate, blood pressure) or brain activity (brain waves or neural activities); however such measurements are intrusive as examinees must be attached to or examined by a machine. This makes measuring physiological responses and brain activity not applicable to many real life situations and they will not be discussed in this chapter.
Before commencing discussing the common pitfalls associated with lie detection, two issues merit attention: (1) underlying theoretical premises of verbal and nonverbal cues to deception and its detection; and (2) research methods used in deception research.
2 Underlying theoretical premises of (non)verbal cues to deceit
Different theoretical approaches have been proposed to predict verbal and nonverbal cues to deception, particularly Ekman and Friesen’s (1969) leakage and deception cues approach; Zuckerman, DePaulo and Rosenthal’s (1981) multi-factor model; Ekman’s (1985/2009) emotion approach; Buller and Burgoon’s (1996) Interpersonal Deception Theory; and DePaulo’s self-presentational perspective (De-Paulo 1992; DePaulo et al. 2003). These approaches have in common the notions that compared to truth tellers liars may (i) experience stronger emotions (particularly fear as a result of detection apprehension); (ii) experience higher levels of cognitive load, and (iii) be inclined to use more and different strategies to make a convincing impression on others.
Traditionally, verbal and nonverbal lie detection focused on the difference in emotions experienced by liars and truth tellers. Ekman’s (1985/2009) analysis of facial micro-expressions of emotions is a prime example (see the TV program ‘Lie to Me’), but lie detection techniques promoted in police manuals are also primarily based on the notion that liars are more concerned and nervous than truth tellers (Vrij and Granhag 2007). Nowadays, most deception researchers use a cognitive lie detection approach which postulates that lying is mentally more taxing than truth telling. Another strain of recent lie detection research aims to exploit the fact that liars use more and different strategies to avoid detection than truth tellers. In sum, in verbal and nonverbal lie detection the emphasis has moved in recent years from emotion-based lie detection techniques to cognitive – load lie detection techniques that exploit liars’ and truth tellers’ different psychological states and take their differential strategies into account.
3 Deception research methods
In research where nonverbal and verbal cues to deception have been examined, trained raters watch video-footage or analyse transcripts of such footage of truth tellers and liars. They analyse the frequency of occurrence or duration of various nonverbal and verbal cues displayed by the truth tellers and liars (e.g., body movements, eye contact, smiling, pauses, number of detail, type of detail, contradictions) and compare the truthful and deceptive responses. There are two types of studies. In field studies video-footage of real-life settings is analysed, such as police – suspect interviews (Mann, Vrij and Bull 2002). In laboratory studies video-footage/transcripts of participants instructed by researchers to tell the truth or to lie for the sake of the experiment are analysed. Although field studies seem more realistic, there are problems associated with conducting them, particularly establishing the ground truth: only responses known to be true or false can be analysed. To establish ground truth satisfactorily, independent case facts, such as medical evidence, material evidence or DNA evidence are needed. Such independent case facts are often not available in real-life cases. In laboratory studies, researchers ask participants, mostly university students, to tell the truth or lie, and measure their nonverbal and verbal responses during truth telling and lying. In such studies, participants have told the truth or lied about many different topics, including a film they had just seen, whether they had a certain object in their pocket, their involvement in money that went missing, their feelings about certain people, or some controversial issues. More recently, scenarios have been introduced that better reflect forensic real-life situations. For example, participants were sent to a shop to buy a product (truth tellers) or to steal a wallet (liars) and were interviewed about the alleged shop visit (Hartwig, Granhag, Strömwall and Kronkvist 2006); participants were sent on a mission to receive a package at a certain location and deliver it somewhere else and were interviewed about this mission (liars had to hide the details of this mission) (Vrij, Leal, Mann and Granhag 2011). Other studies have employed similar realistic scenarios (Strömwall, Granhag and Jonsson 2003; Vrij, Granhag, Mann and Leal 2011). The advantage of laboratory studies is that the ground truth can be established; however, laboratory studies have limitations. Study participants do not choose to lie, but are instructed to do so by the experimenter. Thus lying is condoned. In addition, the stakes (negative consequences of being caught or positive consequences of being believed) are not high (Ekman, 1985/2009; Miller and Stiff 1993).
In a typical lie detection study, observers (often undergraduate students, but sometimes professionals such as police officers or police detectives) are given short video fragments of people they do not know who are either telling the truth or lying. The fragments the observers judge are typically derived from the studies that have been discussed in the previous paragraph. The observers are asked to indicate after each fragment whether the person (often called the sender) was telling the truth or lying. In such a study, simply guessing whether the sender spoke the truth or lied would result in correctly classifying 50% of the truths (truth accuracy rate) and 50% of the lies (lie accuracy rate), resulting in a total accuracy rate (truth and lie accuracy rate combined) of 50%.
In lie detection studies, observers are typically not given any background information about the senders and their statements, so the only source of information available to them is senders’ nonverbal and verbal behaviour. (Exceptions are the Strategic Use of Evidence studies discussed later in this chapter, Opportunities 11). Compared to real life this is a somewhat unusual way of detecting lies. When college students recalled an instance in which they had detected that another person had lied to them, the students reported that less than 2% of the lies were detected at the time the lie was told by relying exclusively on the liars’ nonverbal behaviour or speech content. More commonly, lies were discovered via information from third parties (38%), physical evidence (23%), and confessions (14%) (Park, Levine, McCornack, Morrisson and Ferrera 2002).
4 Pitfalls in lie detection
4.1 Lack of motivation to catch liars
4.1.1 The Ostrich-Effect
Lies often remain undetected because people do not attempt to uncover the truth, a phenomenon labelled the Ostrich effect (Vrij 2008). People are sometimes not motivated to learn the truth because it is not in their best interest to do so. For example, why bother trying to discover whether mendacious compliments about one’s body shape, hairstyle, dress sense, or achievements and so on, are truthful? More serious lies also remain undiscovered for this reason. Not knowing what to do when having learned the truth may also be the reason why some people overlook evidence for possible infidelity by their romantic partners, and remain in denial (Feldman and Cauffman 1999).
4.2 Lie detection is a difficult task
4.2.1 Absence of Pinocchio’s growing nose
Pinocchio’s nose grew larger each time he lied, but was unaltered each time he spoke the truth. Consequently, his growing nose was a reliable cue to deceit. In reality such a cue to deceit does not exist. It does not exist in behaviour, speech, physiological responses or brain activity.
4.2.2 Subtle differences
DePaulo et al.’s (2003) meta-analysis of cues to deception included 50 verbal and nonverbal cues that were examined in five or more deception studies. Of these 50 cues, 14 (28%) showed a significant relationship with deception. The average effect size of these 14 significant cues was d = .25. Cohen (1988, 1992) has argued that d-values around .20 represent a small effect and that a small effect is a barely perceptible but real difference. He gave as an example the difference in height between 15- and 16-year-old girls (Cohen, 1988; Rice and Harris 2005). DePaulo et al.’s meta-analysis thus indicates that most verbal and nonverbal cues do not appear to be related to deception at all and that those that are show only a weak relationship with deception. Consequently, cues to deceit are faint and unreliable.
4.2.3 Countermeasures
A further complication for lie detectors is that liars – particularly those communicating high-stakes lies – often deliberately attempt to appear credible to avoid detection by invoking countermeasures. A verbal veracity assessment tool widely used by professional lie catchers in alleged child sexual abuse cases is Statement Validity Assessment (SVA). SVA assessments are accepted as evidence in some North American courts (Ruby and Brigham 1997) and in criminal courts in several West-European countries (Vrij 2008). However, children (and adults) who are knowledgeable about the workings of this tool can tell stories that sound plausible to experts who use this tool (Vrij, Akehurst, Soukara and Bull 2002, Vrij, Kneller and Mann 2000).
If liars realise that observers are paying attention to their behavioural reactions to detect lies, liars may attempt to control behaviour that could betray their lies (Buller and Burgoon 1996). They may wish to avoid exhibiting behaviours that they believe will create a dishonest impression on others and may try instead to show behaviours that they believe to appear credible (Hocking and Leathers 1980). Behaviours such as ‘gaze aversion’ and ‘making grooming gestures’ are among those most naïve observers associate with deceptive behaviour (see below); thus, liars may attempt to avoid displaying such behaviours. Liars appear to be successful in this as gaze aversion and grooming gestures are unrelated to deception (DePaulo et al. 2003).
The successful use of countermeasures is a major challenge for professional lie detectors. If terrorists, spies and criminals know which lie detection tools will be used to catch them, they may learn more about these tools and attempt to thwart their use and render them ineffective.
4.2.4 Embedded lies
Unfortunately for lie detectors, lies are often embedded in truths. Rather than telling blatant lies that are entirely untruthful, liars tend more to change specific vital details in otherwise truthful stories (Leins, Fisher and Ross 2013). Thus, when a man wants to conceal his illicit activities on Tuesday night, he could describe what he really did on another night (say Monday night) but pretending that those events happened on the Tuesday night in question. In a similar vein, when examining false identities given by criminals, Wang, Chen and Atabakhsh (2004) found that criminals typically change only a small portion of their original identity. Of course, the less deceptive liars are in their responses, the more difficult their lies will become to detect.
4.2.5 Inadequate feedback
Lie detectors often do not receive adequate feedback about their judgments and therefore cannot learn from their mistakes. Helpful feedback is provided frequently, reliably, and immediately. Thus, observers should be informed immediately after every interaction with another person whether that person was lying or not. They could then learn how liars truly behave and what they really say and incorporate such knowledge into their lie-catching activities. However, adequate feedback often is unavailable (DePaulo and Kirkendol 1989). People often never discover that they have been lied to, or such knowledge is gained long after the interaction (Park, Levine, McCornack, Morrison and Ferrara 2002). In many wrongful conviction cases, the police and/or judge only discover their credibility assessment errors years or decades after they occur. Customs officers also face feedback problems (DePaulo and Pfeifer 1986). Part of their jobs is to detect smugglers amongst travelers. They receive virtually no feedback at all from the numerous passengers they decide not to search. Some of them may be smugglers, but once the officers let them pass unsearched, they almost never find out that they made a mistake. Officers may not get adequate feedback from the people they search; that is those whose illegal goods remain undetected in such searches.
4.2.6 Violation of conversation rules
The act of lying becomes increasingly difficult when the lie detector asks appropriate questions that follow an initial account by the sender (Vrij 2008). However, further questioning in daily life conversations can violate social norms, and be seen as inappropriate, strange or impolite. Conversation partners will probably not appreciate utterances such as “Could you elaborate on that?”, and “Could you repeat what you just said?”.
Although focusing on a speaker’s body movements could benefit the lie detector because they may reveal signs of deceit (DePaulo et al. 2003), such movement scanning might seem strange and inappropriate in everyday interactions. Conversation rules dictate that a listener should look into a speaker’s eyes, but the eyes themselves generally do not reveal reliable information about deception (DePaulo et al. 2003). Therefore, the conversational rules (i) discouraging probing questions and (ii) encouraging the maintenance of eye gaze can hamper lie detection.
4.2.7 Good liars
Lie detection can be made more difficult because some people are good liars. Surprisingly little research has addressed the features of a good liar, but six features may be especially important. The best liars are those individuals (i) whose natural behaviour disarms suspicion; (ii) who do not find it cognitively difficult to lie; (iii) who do not experience emotions such as fear, guilt or delight when they are lying; (iv) who are good actors and who display a seemingly honest demeanour; (v) whose attractiveness may lead to an inference of virtue and honesty; and/ or (vi) who have good insight into another person’s thought processes, which they can use to fool the other person into believing that they are telling the truth (see Vrij 2008; Vrij, Granhag and Porter 2010 and Vrij, Granhag and Mann 2010 for in-depth discussions of these factors).
4.3 Common errors made by lie detectors
People not only fail to catch liars because they are unmotivated to catch them or because lie detection is difficult, they also make systematic errors in the evaluation process. Eight common errors will be identified in this section.
4.3.1 Examining wrong cues
Overwhelmingly, both laypersons and professional lie catchers expect liars to act nervously, with exhibiting gaze aversion (“liars look away”) and displaying grooming gestures (“liars fidget”) amongst the most popular beliefs (Strömwall, Granhag and Hartwig 2004; The Global Deception Team 2006; Vrij 2008). Despite their overwhelming endorsement internationally, cues such as gaze aversion and grooming gestures are not reliable cues to deception (DePaulo et al. 2003). Nonetheless, police and other legal professionals are encouraged to use such incorrect cues in detecting lies (Johnson 2006a, b). For example, in their influential police manual, Inbau, Reid, Buckley and Jayne (2013) identify several nonverbal cues as being diagnostic of deception, including avoiding eye contact and grooming gestures, but also cues such as frequent posture changes, placing hands over mouth/ eyes and lack of illustrators. Research has revealed that only one of these cues is associated with deception: Lying is associated with a decrease in illustrators (e.g., DePaulo et al. 2003). It is therefore not surprising that when police officers saw video fragments of suspects who told the truth or lied during their police interviews, the more the police officers endorsed the lie cues promoted in the Inbau et al. manual, the worse they became at distinguishing between suspects’ truths and lies (Mann, Vrij and Bull 2004). The cues reported by Inbau et al. are cues to nervousness and police manuals typically advise investigators to pay attention to signs of nervousness when attempting to detect deceit (see Vrij and Granhag 2007, for a review of such manuals).
4.3.2 Over-emphasis on nonverbal cues
In specific situations observers primarily rely on speech content when attempting to detect deceit. First, observers who are knowledgeable about the facts discussed by the sender rely on verbal content and compare their knowledge with the story provided by the sender. Second, occasionally observers may have obtained multiple statements from the same person, or they may have obtained statements from different people. In such cases, observers focus on the verbal cue ‘level of consistency between the statements’ (Granhag and Strömwall 1999). Third, observers rely on verbal cues when they are distinctive, particularly when a statement appears to be against the story-teller’s self-interest (Noller 1985), such as a confession.
When professional lie catchers possess no factual information, have no statements for comparison, and when the speech content is not distinctive, they are inclined to pay greater attention to nonverbal behavior than to verbal behavior (Mann, Vrij and Bull 2004; Porter, Woodworth and Birt 2000). The preference for nonverbal behaviors as indicators of deception may result from training that encourages such an emphasis. For example, police training manuals give greater attention to nonverbal than to speech content cues as cues to deceit (See Vrij and Granhag 2007 for a review of visual cues mentioned in police manuals). This nonverbal dominance is further emphasised with explicit statements. For example, Inbau and colleagues stated in their training manual that “According to various social studies as much as 70 percent of a message communicated between persons occurs at the nonverbal level” (Inbau et al. 2013: 122). This quote is derived from the work of Albert Mehrabian. In his studies participants were sometimes instructed to respond with single words, not a situation that gives verbal content much chance to make an impact on the exchange of information. Also, Mehrabian’s studies dealt with the communication of feelings and attitudes (i.e., like-dislike). As Mehrabian (1981: 79) said: “Incidentally, we should be careful to note that these assertions about the disporportionate contribution of implicit, relative to verbal, cues is limited to feelings (pleasure, arousal, dominance) and like-dislike.” In police – suspect interviews police detectives mostly discuss a suspect’s alleged activities rather than his/her feelings or attitudes.
Overemphasis on nonverbal cues to deception is problematic. Meta-analyses have shown that many speech-related cues are more diagnostic of deception than nonverbal cues (DePaulo et al. 2003; Vrij 2008a). In addition, observers who pay sole attention to nonverbal cues are less accurate in discriminating between truths and lies than those who consider speech content (Bond and DePaulo 2006).
4.3.3 The Othello-Error
Lie detectors make the common mistake of interpreting too readily certain behaviours as diagnostic of deception, particularly signs of nervousness. Specifically, they fail to consider that truth tellers (e.g., an innocent suspect) can be as nervous as liars. Truth tellers can be nervous as a result of being accused of wrongdoing or due to fear of not being believed, because they too could face negative consequences if they are not be believed (Ofshe and Leo 1997). Deception researchers have labelled the misinterpretation of signs of nervousness in truth tellers as signs of deceit as the Othello-error (Ekman, 1985/2009), based on Shakespeare’s character.
4.3.4 Use of heuristics
Instead of carefully scrutinizing someone’s responses in evaluating his/her credibility, observers may rely on general decision rules or heuristics (Fiedler and Walka 1993). Applying heuristics can be an effective way for observers with limited time and attentional resources to deal with complex environments or information processing demands (Albrechtsen, Meissner and Susa 2009), but not all heuristics are functional. The question then is which heuristics to use and to avoid. Deception researchers have focused considerable attention on problematic heuristics but little on potentially effective ones.
Although discussing all the heuristics is beyond the scope of this chapter, consider the facial appearance heuristic (Vrij 2004). This heuristic refers to the tendency to judge people with attractive, symmetrical faces or baby-faced appearances as honest, and people with certain facial characteristics suggesting anger and unkindness as dishonest (Porter, Juodis, England and ten Brinke 2008). This is an inappropriate heuristic to use, as a link between physical characteristics and deception does not exist.
4.3.5 Neglect of interpersonal differences
There are large individual differences in people’s behaviour and speech (DePaulo and Friedman 1998). Some people typically make many movements, others do not; some people are eloquent, others are not, etc. Verbal lie detection tools such as SVA attempt to control for these interpersonal differences by using a Validity Checklist, a list of factors that is thought to have an impact on people’s verbal responses (Vrij 2008).
However, when evaluating behavioural responses, interpersonal differences often are uncontrolled (e.g., Meissner and Kassin 2002), and the result is that people whose natural behaviour looks suspicious (e.g., people who naturally avert their gaze or fidget a lot) risk being falsely accused of lying. The social clumsiness of introverts and the impression of tension, nervousness or fear that is naturally given off by socially anxious individuals may appear suspicious (Riggio, Tucker and Throckmorton 1988). When people of different ethnic backgrounds or cultures interact with each other, behaviours naturally displayed by members of one ethnic group or culture may appear suspicious to members of another ethnic group or culture, thus potentiating errors (Johnson 2006a, b; Vrij and Winkel 1991, 1992).
4.3.6 Neglect of intrapersonal variations
Not only do different people respond differently in the same situation (interpersonal differences), the individual also responds differently in different contexts (intrapersonal differences). Neglecting or underestimating intrapersonal differences is another error that lie catchers make. In police interviews, detectives are advised to examine a suspect’s natural, truthful, behavior during the small-talk part at the beginning of the interview and to compare this behavior with the behavior shown by the suspect during the actual interview. Differences in behavior could then be interpreted as ‘significant’ (Inbau et al. 2013). Although the approach sounds appealing, it is prone to incorrect judgments because an incongruent comparison is made. Engaging in small-talk and discussing the crime are fundamentally different situations. In low-stakes small-talk conversations, the suspect’s responses are unlikely to produce any negative consequences. In contrast, in the high stakes core investigative elements of the interview the suspect’s reactions and responses are critical. Therefore, unsurprisingly, both guilty and innocent suspects tend to show different behaviours during small-talk compared to the actual interview (Vrij 1995).
4.3.7 Existing interview techniques
Many interview strategies advocated by police manuals can impair lie detection. For example, police detectives sometimes confront suspects at the beginning of the interview with the evidence they have gathered earlier on in the investigation process (Hartwig, Granhag, Strömwall and Kronkvist 2006; Leo 1996). This tactic is designed to show suspects that it is fruitless to remain silent and that they are better off confessing. Research has revealed that this interview style hampers lie detection (Hartwig, Granhag, Strömwall and Vrij 2005). Because liars do not know what the observer knows, they do not know what they can say without running the risk of contradicting facts known to the observer. If police officers disclose the facts they know to suspects, they facilitate the ease of lying. Disclosing evidence early on provides liars with the opportunity to adapt their stories and to give an innocent explanation for the evidence.
Another misguided strategy is to accuse someone of lying. This gives deceptive suspects the opportunity to ‘escape’ from the interview situation by saying that they will no longer cooperate with the investigation, claiming that further co-operation is futile because they are not believed anyway. Also, accusing someone of lying may elicit the same responses in liars and truth tellers. That is, both suspects correctly accused of lying and those wrongly accused of lying may become afraid of not being believed. Because of that fear, truth tellers may show the same nervous responses as liars (Ofshe and Leo 1997).
4.3.8 Overconfidence in lie detection skills
Professional lie catchers tend to overestimate their ability to detect deceit; however, although professionals are more confident in their veracity judgments than laypersons, they are no more accurate (Kassin, Meissner and Norwick 2005). This tendency to be confident is not unique to police officers, but common amongst many groups of professionals in carrying out various tasks (Allwood and Granhag 1999).
A problem associated with confidence in lie detection is that it is not related to accuracy. A meta-analysis revealed the confidence – accuracy relationship to be virtually non-existent, r = .04, and did not differ significantly from zero (DePaulo, Charlton, Cooper, Lindsay and Muhlenbruck 1997). High confidence in one’s ability to catch liars can be harmful when the confidence is unjustified (Kalbfleisch 1992). High confidence often results in making quick decisions on the basis of limited information (Lord, Ross and Lepper 1979), or tunnel vision (Porter and ten Brinke 2010). In addition, it could make investigators prone to attempt to detect lies via demeanor, which could occur at the expense of searching for physical evidence (Colwell, Miller, Lyons and Miller 2006). High confidence also is likely to reduce motivation to learn more about lie detection, as investigators may already consider themselves to be experts in the area.
5 Opportunities in lie detection
5.1 Avoiding the errors
5.1.1 Avoid examining the wrong cues and attend to more diagnostic cues
As discussed, observers often base their veracity decisions on cues that are not diagnostic of deception; thus observers may become better at discriminating between truths and lies if they are taught to pay attention to more diagnostic cues to deception. Unfortunately, training people to pay attention to specific cues on average resulted in a minimal increment (4.26%) in lie detection accuracy (Vrij 2008), underlining the limitations of this approach. However, some have drawn a more positive picture of training effectiveness (Driskell 2011). He reported a medium positive effect (d = .50) of training and concluded that it is an effective means of enhancing detection accuracy. Although training can increase accuracy, Driskell may be too optimistic. After all, Bond and DePaulo (2006) have shown that untrained judges perform only just above the level of chance; thus a medium increase in that level is not overly impressive. Driskell (2011) correctly observes that the cues that are used in the training should in fact be reliable indicators of deception. And that is the problem: Such cues are, at best, faint.
5.1.2 Avoid relying on nonverbal cues only
Research addressing the individual strategies of lie detectors indicates that people are least successful in distinguishing truths from lies when they pay attention to nonverbal cues only and that they become better if speech content is taken into account (see Vrij, Granhag and Porter 2010, for a review). Another body of research suggests that a ‘holistic’ approach to detecting deception, in which verbal and nonverbal cues are observed, may be preferable (Porter and ten Brinke 2010).
Observers can pay attention to nonverbal behavior and speech simultaneously in three different ways. All three methods will enhance lie detection. First, observers can take into account a mixture of nonverbal and verbal cues, without looking at the relationship between the two sets of cues. Second, observers can examine nonverbal behavior in relation to speech content, an approach common in communication research but often ignored by deception researchers. Research has shown the potential of this approach (Caso, Maricchiolo, Bonaiuto, Vrij and Man 2006; Hillman, Vrij and Mann 2012). When the entire interview was taken into account, truth tellers and liars displayed a similar number of illustrators, but differences did emerge when specific types of gestures were examined when answering specific questions.
Third, observers can examine mismatches between nonverbal behavior and speech content (Ekman 1985/2009). Thus, a person who makes a head shake while agreeing to cooperate may not be as cooperative as he wants to convey. However, mistakes are easily made. For example, some people display clear signs of distress when they talk about a negative event they have experienced, whereas others do not (Burgess 1985). Thus, varying communication styles represent a personality factor (Vrij 2008). However, observers, including police detectives, typically believe that an absence of distress during an interview about an upsetting event is a valid indicator of deceit (Greuel 1992). Porter, and ten Brinke (2010) rightly concluded that nonverbal cues only assist investigators who are informed about the complex relationships between behavior and deceit.
5.1.3 Avoid the Othello Error
To avoid the error of mistakenly interpreting signs of nervousness as cues to deceit, the lie detector should consider questions such as: “Is my questioning likely to evoke emotions in the respondent, regardless of whether he or she is guilty?”, “Is the present situation likely to evoke emotions in the respondent anyway?”, “Is this person the type who is likely to be emotional in this situation anyway?” (Ekman 1985/2009). When the answers to those questions are affirmative, emotional responses cannot be relied upon to judge deceit.
In theory, another cluster of cues could betray deception: Cues associated with having to think hard (cognitive load). Such cues are not solely exhibited by liars either and truth tellers may have to think hard while answering questions. Again, in interpreting cues of cognitive load the lie detector should ask him/herself the same kinds of question as when interpreting signs of emotions, such as “Is my questioning likely to evoke cognitive load in the respondent, regardless of whether he or she is guilty?”
5.1.4 Avoid relying on heuristics and rely on multiple cues in a flexible manner
As discussed, deception research has revealed that no single behavioural or verbal cue is uniquely related to deception. Instead, different people show different cues to deception in a given situation (i.e., interpersonal differences) and the same person shows different cues to deception on different occasions (i.e., intrapersonal differences). Therefore, it is inappropriate to use fixed decision rules based on heuristics such as ‘liars look away’ when attempting to detect deceit. Instead, better accuracy rates are achieved if someone uses flexible decision rules that include multiple cues (Ekman, O’Sullivan and Frank 1999; Mann, Vrij and Bull 2004). The question, of course, remains which decision rule to use and which cues to pay attention to in a given individual in a given situation.
5.1.5 Ascertaining the comparable truth
Lie detectors should take interpersonal and intrapersonal differences into account when making veracity judgments. Therefore, the question for the lie detector to ask is whether the nonverbal behavior and speech patterns displayed by a person differ from this person’s known truthful responses. For this technique to work, it is essential that the known truthful response (e.g., baseline response) is made under similar conditions to the response under investigation, labeled the comparable truth (Vrij 2008). People react differently in formal settings than in informal settings. They also react differently when they are accused of wrongdoing (e.g., during the actual interview) than when they are unchallenged (e.g., during small-talk) (Vrij 2006), and respond differently in high-stakes situations than in low-stakes situations (Porter and ten Brinke 2010; Vrij 1995). In addition, people show different behaviors when they are interviewed by different people (Vrij and Winkel 1991). Behavior is also topic-related. People respond differently when discussing a topic that embarrasses them versus a neutral topic (Kleinke 1986), and when they discuss a topic that is important to them versus a topic with which they have less personal involvement (Davis and Hadiks 1995). Finally, people’s behavior sometimes varies over time in the same interview (Buller and Burgoon 1996 or, if they are interviewed on more than one occasion, changes may occur over repeated interviews (Granhag and Strömwall 2002). Therefore, when lie detectors wish to compare a person’s given nonverbal response with a truthful nonverbal response from the same person, they must make sure that the given and truthful responses are taken from the same interview setting; that the person talks about similar topics in the given and truthful parts; and that these topics are discussed within a short period of time.
Vrij and Mann (2001) provided an example of how the comparable truth technique could be employed. During a videotaped real-life police interview, a man suspected and later convicted of murder was asked to describe his activities during a particular day. The murder suspect described his activities during the morning (went to work), afternoon (visited a market) and evening (visited a neighbour). Detailed analyses of the videotape revealed a sudden change in behaviour as soon as he began to describe his activities during the afternoon and evening. A possible reason for this variation may have been that he was lying, a view supported by the evidence. Police investigators could confirm his story with regard to his morning activities, but revealed that his statement about the afternoon and evening was fabricated. In reality, he met the victim in the afternoon and killed her later that day. In this case, we were able to make a good comparison. The man described a seemingly normal day, and there are no good reasons why different behaviours would emerge while describing different parts of that day.
6 Exploiting the different mental processes of truth tellers and liars
The first five guidelines shared one feature: They all aim to examine and interpret more carefully the nonverbal and verbal cues displayed by liars. These guidelines have one serious limitation: The cues lie detectors are encouraged to examine and interpret are both faint and unreliable. This section discusses a fundamentally different approach to nonverbal and verbal lie detection aimed at provoking stronger and reliable cues to deceit. This is achieved by exploiting the different psychological states of truth tellers and liars. There are two different approaches. The first approach (imposing cognitive load) deals with making the interview setting more difficult for interviewees. This potentially affects liars more than truth tellers, thus accentuating differences between them. The second approach (strategic questioning) deals with specific questions that elicit the most differential responses between truth tellers and liars. Both approaches require interviewees to talk. Contrary to police manuals’ characterisations, suspects are generally willing to talk, even in police interviews in which very few (5%) remain silent (Moston, Stephenson and Williamson 1993).
6.1 Information-gathering interview style
The police commonly use a mixture of two types of interview styles: information-gathering and accusatory (Moston and Engelberg 1993). When using the information-gathering style, interviewers ask suspects to give detailed statements about their activities through open questions (e.g., “What did you do yesterday between 3 pm and 4 pm?”; “You just mentioned that you went to the gym, who else was there?”). By comparison, when employing the accusatory style, interviewers are more confrontive (e.g., “Your reactions make me think that you are hiding something from me”). Information-gathering interviews encourage suspects to talk, whereas accusatory interviews often yield short denials (e.g., “I am not hiding anything”).
A recent meta-analysis revealed that information-gathering approaches significantly increased the likelihood of true confessions and significantly decreased the likelihood of false confessions compared to accusatory approaches. In addition, accusatory approaches significantly increased both true and false confessions compared to control conditions (Meissner, Redlich, Bhatt and Brandon in press). Moreover, information-gathering approaches elicited significantly more relevant information and significantly more diagnostic cues to deceit than accusatorial methods (Meissner et al. in press). Apparently, information-gathering interviews result in longer answers than accusatory interviews and longer stories afford more opportunities for nonverbal cues and verbal cues to deception to occur (DePaulo et al. 2003). Additionally, being accused of wrongdoing (i.e., accusatory interview style) is likely to affect the behaviour of both truth tellers and liars in a similar way, and the accusation can have a stronger impact on someone’s nonverbal behaviour than the act of lying itself (Ofshe and Leo 1997). Consequently, differences in nonverbal behaviour between truth tellers and liars are overshadowed by the effects of the accusation.
In summary, an information-gathering approach yields better results in terms of obtaining confessions, obtaining relevant information and eliciting cues to deceit. Although the information-gathering interview is a good start in discriminating truth and deceit, the approach alone is not sufficient to elicit diagnostic cues to deception (Vrij and Granhag 2007). More sophisticated strategies incorporated within the information-gathering interview are needed and will be discussed in the remaining part of this chapter.
6.2 Imposing cognitive load
As discussed earlier, deception theory postulates that liars may be more nervous and have to think harder than truth tellers. However, research has shown that liars often do not display cues of nervousness and cognitive load and that cues to deception are typically faint and unreliable. Could interviewers employ techniques that elicit and enhance differences in nervousness or cognitive load? Consistent with the National Research Council (2003), it is unlikely that questions can be asked that will necessarily elicit more concern in liars than in truth tellers. None of the interventions that follow therefore aim to raise concern in interviewees. In contrast, it is possible to enhance differences in cognitive load between truth tellers and liars.
Lying can be more cognitively demanding than truth telling for six reasons (Vrij et al. 2008). First, formulating the lie itself may be cognitively demanding. Second, liars are typically less likely than truth tellers to take their credibility for granted and liars will therefore be more inclined than truth tellers to monitor and control their demeanour in order to appear honest to the lie detector. Such monitoring and controlling is cognitively demanding. Third, because liars do not take credibility for granted, they may monitor the interviewer’s reactions more carefully to assess whether their deceit is being successful. Fourth, liars may be preoccupied by the task of reminding themselves to act and role-play which requires extra cognitive effort. Fifth, liars have to suppress the truth while they are lying and this is also cognitively demanding. Finally, while activation of the truth often happens automatically, activation of the lie is more intentional and deliberate, and thus requires mental effort.
A lie detector could exploit the differential levels of cognitive load that truth tellers and liars experience to discriminate more effectively between them. Liars who require more cognitive resources than truth tellers for the act of story telling will have fewer cognitive resources than truth tellers. This makes liars vulnerable, thus if cognitive demand is further raised, which could be achieved by making additional requests, liars may not be as good as truth tellers in coping with these additional requests.
One way to impose cognitive load is by asking interviewees to tell their stories in reverse order. This increases cognitive load because (a) it runs counter to the natural forward-order coding of sequentially occurring events and (b) it disrupts reconstructing events from a schema (Gilbert and Fisher 2006). Another way to increase cognitive load is by instructing interviewees to maintain eye contact with the interviewer. When people have to concentrate on telling their stories – likely when they are asked to recall what has happened – they are inclined to look away from their conversation partner (typically to a motionless point), because maintaining eye contact is distracting (Doherty-Sneddon and Phelps 2005). In two experiments, half of the liars and truth tellers were requested to recall their stories in reverse order (Vrij et al. 2008) or to maintain eye contact with the interviewer (Vrij, Mann, Leal and Fisher 2010), whereas no instruction was given to the other half of the participants. More cues to deceit emerged in the reverse order and maintaining eye contact conditions than in the control conditions. Moreover, observers who watched these videotaped interviews could distinguish between truths and lies better in the reverse order condition and maintaining eye contact conditions than in the control conditions. In the reverse order experiment, 42% of the lies were correctly classified in the control condition, but in the experimental condition, 60% of the lies were correctly classified, more than typically found in this type of lie detection research.
An alternative method of imposing cognitive load on liars is to ensure that in a given interview setting truth tellers will provide more information. Talkative truth tellers raise the standard for liars, who also need to become more talkative to match truth tellers. Liars may be reluctant to add more information out of fear that it will provide leads to investigators and, consequently, give their lies away. They may also find it too cognitively difficult to add as many details as truth tellers do. Alternatively, if liars do add a sufficient amount of detail, the additional information may be of lesser quality or may sound less plausible.
One way of increasing the amount of detail truth tellers generate is by introducing a second interviewer. Participants in an experiment told the truth or lied about their occupation. In the interview room two interviewers were present, whereby the second interviewer remained silent throughout the interview but displayed different demeanours (Mann et al. in press). In one condition he was supportive throughout (e.g., nodding his head and smiling); in a second condition he was neutral and in a third condition he was suspicious (e.g., frowning). Being supportive during an interview facilitates talking and encourages cooperative witnesses (e.g., truth tellers) to talk (Bull 2010; Fisher 2010). Indeed, truth tellers provided the most detail in the supportive condition and only in that condition did they provide significantly more detail than liars (Mann et al. in press).
6.3 Strategic questioning: Unanticipated questions
A consistent finding in deception literature is that, when possible, deceivers plan their lies (Hartwig, Granhag and Strömwall 2007). Preparation makes sense because planned lies contain fewer cues to deceit than spontaneous lies (DePaulo et al. 2003) and are subsequently harder to detect. However, the effectiveness of a liar’s ‘planning’ strategy is limited, as it can only work when liars anticipate the questions that will be asked. Investigators can exploit this limitation by asking questions that liars do not anticipate or by asking questions in an unanticipated format.
In a test of this technique, liars and truth tellers were interviewed individually about having had lunch together at a restaurant (Vrij, Leal, et al. 2009). While the pairs of truth tellers did have lunch together, the liars were instructed to pretend that they had. All pairs were given the opportunity to prepare for the interview. The interviewer asked typical opening questions which (as established after the interview) the interviewees had anticipated (e.g., “What did you do in the restaurant?”), followed by questions about spatial (e.g., “In relation to the front door and where you sat, where were the closest diners?”) and temporal (e.g., “Who finished their food first, you or your friend?”) information which the interviewees had not anticipated. Further, they were asked to draw the layout of the restaurant (unanticipated). The pairs of liars showed the same amount of overlap in their responses to the anticipated general questions as the pairs of truth tellers, but less overlap than truth tellers in their responses to the unanticipated questions. Based on overlap in answer to the spatial questions, 72% of pairs of truth tellers and liars could be classified (i.e., the answers were less alike for the pairs of liars than they were for the truth tellers). This demonstrates the potential of asking spatial questions for lie detection purposes. An even higher percentage of pairs of truth tellers and liars, 78%, could be correctly classified when assessing drawings (i.e., the drawings were less alike for the pairs of liars than they were for the truth tellers). When liars anticipated the questions, their answers made them undistinguishable from truth tellers. Asking unanticipated questions about central topics led to identifiable betrayals among liars.
Comparing the answers to anticipated and unanticipated questions can also be used to detect deceit in individual liars. Observers pay attention to detail and the richer an account is perceived to be in detail, the more likely it is to be believed (Bell and Loftus 1989). Liars are aware that interviewers pay attention to detail when making veracity judgements and are therefore motivated to provide details in order to make an honest impression (Nahari, Vrij and Fisher 2012). Hence, liars may prepare a detailed alibi and report it as soon as the opportunity arises. This opportunity arises when an expected question is asked. In contrast, liars will not have prepared answers for unexpected questions and may therefore struggle to generate detailed answers to them. As a result, liars will be more detailed when answering the expected questions than the unexpected questions. For truth tellers, the difference in detail between expected and unexpected questions should be less pronounced. When asked about an event they have experienced, truth tellers will search their memories for details about that event and there is no reason why those details are less accessible for unexpected than for expected questions, as long as the expected and unexpected questions are both about core aspects of the event. If liars’ and truth tellers’ answers to expected and unexpected questions are compared separately, liars’ prepared answers to expected questions may be more detailed than the answers truth tellers typically generate when searching their memory. In contrast, liars‘ answers to unexpected questions may be less detailed than the answers truth tellers typically generate when searching their memory.
In Warmelink, Vrij, Mann, Jundi and Granhag (in press) truth tellers and liars were interviewed about their alleged forthcoming trip. Anticipated questions about the purpose of the trip (e.g., “What is the main purpose of your trip?”) were followed by unanticipated questions about transport (e.g., “How are you going to travel to your destination?”), planning (“What part of the trip was easiest to plan?”), and the core event (“Keep in mind an image of the most important thing you are going to do on this trip. Please describe this mental image in detail?”). Compared to truth tellers, liars gave significantly more detail to the anticipated questions and significantly less detail to the unanticipated questions.
6.4 Strategic questioning: Temporal questions when a scripted answer is suspected
As discussed previously, a good strategy for liars is to provide a story which is, in fact, true, but which happened at a different time from the time of interest (see the embedded lie, above). The only fabricated part in this story is when the liar was at a particular location. Lie detectors should be aware of this lying strategy. Questions about the layout of the location and general activities occurring in it are not necessarily effective because they enable liars to relate true experiences. Instead, the questions asked should be specifically related to the particular time that the interviewee claims to have been there. For example, the interviewer could ask time-related questions about key events, such as what specific individuals were doing and who else was there.
6.5 Specific questioning: The Devil’s Advocate Approach
The Devil’s Advocate approach is designed to detect lies about opinions (e.g., ‘What is your view about the war in Afghanistan?’). Determining the veracity of such conceptual representations can be important in intelligence settings, for example when deciding whether an informant is as much anti-Taliban as he or she claims or whether someone truly wants to enter a country solely for the purpose of university study.
In using this approach interviewees are first asked an opinion eliciting question that asks them to argue in favour of their personal view (“What are your reasons for supporting the Americans in the war in Afghanistan?”). This is followed by a Devil’s Advocate question that asks participants to argue against their personal view (“Playing Devil’s Advocate, is there anything you can say against the involvement of the Americans in Afghanistan?“).
People normally think more deeply about, and hence are likely to be more able to generate, reasons that support rather than oppose their beliefs and opinions (Ajzen 2001). Therefore, truth tellers are likely to provide more information in their responses to the opinion-eliciting question than to the Devil’s Advocate question. This pattern is unlikely to happen in liars as for them the Devil’s Advocate question is more compatible with their beliefs than is the opinion-eliciting question. An experiment testing the Devil’s Advocate approach revealed that truth tellers’ opinion eliciting answers were longer than their Devil’s Advocate answers, and observers judged truth tellers’ opinion eliciting answers to be more immediate, plausible and emotionally involved than their Devil’s Advocate answers. No clear differences emerged in liars’ answers to the two types of question. On the basis of these differences in speech content, 86% of truth tellers and 79% of liars were correctly classified (Leal, Vrij, Mann and Fisher 2010).
6.6 Specific questioning: The strategic use of evidence
Liars (guilty suspects) and truth tellers (innocent suspects) generally enter interviews with different counter-interrogation strategies (Granhag and Hartwig 2008). Research suggests that liars are inclined to use avoidance strategies (e.g., avoiding mentioning where they were at a certain time) or denial strategies (e.g., denying having been at a certain place at a certain time when asked directly). In contrast, truth tellers are generally more forthcoming and “tell the truth like it happened” (Hartwig, Granhag and Strömwall 2007).
When investigators possess critical and possibly incriminating background information (evidence), they can exploit these differential truth tellers’ and liars’ strategies by introducing the available evidence during the interview in a strategic manner (the Strategic Use of Evidence technique, SUE). When questions about the evidence are asked, guilty suspects tend to use more avoidance strategies, whereas innocent suspects use more forthcoming strategies (Granhag and Hartwig 2008); hence innocent suspects’ accounts will be more consistent with the available evidence than guilty suspects’ accounts.
Granhag, Strömwall, Willén, and Hartwig (in press) introduced the so-called Evidence Framing Matrix which suggests that when one piece of evidence is disclosed, two dimensions are particularly helpful in illuminating the different framing alternatives that exist. The first dimension is the strength of the source of the evidence, which can vary from weak to strong. The second dimension is the degree of precision of the evidence, which can vary from low to high. Granhag et al.’s (in press) research showed that using this matrix to reveal the evidence in a stepwise manner moving from the most indirect form of framing (weak source/low specificity, e.g., ‘We have information telling us that you recently visited the central station’) to the most direct form of framing (strong source/high specificity, e.g., ‘We have CCTV footage showing that you collected a package from a deposit box at the central station, ground floor level, on the 24th of August at 7.30 pm’) elicited more and stronger cues to deception than using the most direct form of framing only.
Police trainees were taught to use some basic elements of the SUE-technique. Results showed that participants who received training clearly outperformed their untrained colleagues: 85% vs. 56% deception detection performance (Hartwig, Granhag, Strömwall and Kronkvist 2006). The SUE-technique has been found to be successful in eliciting cues to deception for lying adults and lying children, for lying single suspects and lying multiple suspects, and for suspects lying about their past actions and lying about their intentions (Vrij and Granhag 2012).
7 Conclusion
This chapter has presented an overview of pitfalls and opportunities in nonverbal and verbal lie detection. These 16 pitfalls suggest that nonverbal and verbal cues to deception are ordinarily faint and unreliable thus making lie detection a difficult task as there is no nonverbal or verbal cue upon which lie detectors can truly rely.
Eleven guidelines to improve lie detection were presented. Five of these guidelines are aimed at avoiding common errors made in nonverbal and verbal lie detection. This has been the focus of research for a considerable period of time. The six remaining guidelines aimed at creating more cues and more perceptible and reliable cues to deception by exploiting truth tellers’ and liars’ distinctive psychological states. This has been the focus of recent research. The success of the traditional methods to improve lie detection is seriously hampered by the fact that cues are typically so faint and unreliable. The recently introduced methods attempt to solve this problem, and, as has been demonstrated, with some success. Lie detectors should become actively engaged in exploiting truth tellers’ and liars’ different mental processes. Researchers should focus their efforts on this line of innovative and promising lie detection research.
References
Ajzen, Icek. 2001. Nature and operation of attitudes. Annual Review of Psychology 52: 27–58
Albrechtsen, Justin S., Chris A. Meissner and Kyle J. Susa. 2009. Can intuition improve deception detection performance? Journal of Experimental Social Psychology 45: 1052–1055.
Allwood, Carl M. and Par Anders Granhag. 1999. Feelings of confidence and the realism of confidence judgments in everyday life. In: Peter Juslin and Henry Montgomery (eds.), Judgment and Decision Making: Neo-Brunswikian and Process-Tracing Approaches, 123–146. Mahwah NJ: Lawrence Erlbaum.
Bell, Brad E. and Elizabeth F. Loftus. 1989. Trivial persuasion in the courtroom: The power of (a few) minor details. Journal of Personality and Social Psychology 56: 669–679.
Bond, Charles F. and Bella M. DePaulo. 2006. Accuracy of deception judgements. Personality and Social Psychology Review 10, 214–234.
Bull, Ray. 2010. The investigative interviewing of children and other vulnerable witnesses: Psychological research and working/professional practice. Legal and Criminological Psychology 15: 5–24.
Buller, David B. and Judee K. Burgoon. 1996. Interpersonal deception theory. Communication Theory 6: 203–242.
Burgess, Ann. W. 1985. Rape and Sexual Assault: A Research Book. London: Garland.
Caso, Letizia, Fridanna Maricchiolo, Marino Bonaiuto, Aldert Vrij and Samantha Mann. 2006a. The impact of deception and suspicion on different hand movements. Journal of Nonverbal Behavior 30: 1–19.
Cohen, Jacob. 1988. Statistical Power Analysis for the Behavioral Sciences (2nd edition.). Hillsdale, NJ: Erlbaum.
Colwell, Lori H., Holly A. Miller, Phillip M. Lyons and Rowland S. Miller. 2006. The training of law enforcement officers in detecting deception: A survey of current practices and suggestions for improving accuracy. Police Quarterly 9: 275–290.
Davis, Martha and Dean Hadiks. 1995. Demeanor and credibility. Semiotica 106: 5–54.
DePaulo, Bella M. 1992. Nonverbal behavior and self-presentation. Psychological Bulletin 111: 203–243.
DePaulo, Bella M. and Howard S. Friedman. 1998. Nonverbal communication. In: Daniel T. Gilbert, Susan T. Fiske, & Garder Lindzey (eds.), The Handbook of Social Psychology, 3–40. Boston, MA: McGraw-Hill.
DePaulo, Bella M. and Susan E. Kirkendol. 1989. The motivational impairment effect in the communication of deception. In: John C. Yuille (ed.), Credibility Assessment, 51–70. Dordrecht, the Netherlands: Kluwer.
DePaulo, Bella M., James L. Lindsay, Brian E. Malone, Laura Muhlenbruck, Kelly Charlton and Harris Cooper. 2003. Cues to deception. Psychological Bulletin 129: 74–118.
DePaulo, Bella M. and Roger L. Pfeifer. 1986. On-the-job experience and skill at detecting deception. Journal of Applied Social Psychology 16: 249–267.
Doherty-Sneddon, Gwyneth and Fiona G. Phelps. 2005. Gaze aversion: A response to cognitive or social difficulty? Memory and Cognition 33: 727–733.
Driskell, James E. 2011. Effectiveness of deception detection training: A meta-analysis. Psychology, Crime, & Law, published online.
Ekman, Paul. 1985. Telling lies: Clues to deceit in the marketplace, politics and marriage. New York: W. W. Norton. (Reprinted in 1992, 2001 and 2009).
Ekman, Paul and Wallace V. Friesen. 1969. Nonverbal leakage and clues to deception. Psychiatry 32: 88–106.
Ekman, Paul, Maureen O’Sullivan and Mark G. Frank. 1999. A few can catch a liar. Psychological Science 10: 263–266.
Feldman, Shirley S., Elizabeth Cauffman. 1999. Sexual betrayal among late adolescents: Perspectives of the perpetrator and the aggrieved. Journal of Youth and Adolescence 28: 235–258.
Fisher, Ronald P. 2010. Interviewing cooperative witnesses. Legal and Criminological Psychology 15: 25–38
Gilbert, Julian A. E. and Ronald P. Fisher. 2006. The effects of varied retrieval cues on reminiscence in eyewitness memory. Applied Cognitive Psychology 20: 723–739.
Global Deception Team. 2006. A world of lies. Journal of Cross-Cultural Psychology 37: 60–74.
Granhag, Par Anders and Maria Hartwig. 2008. A new theoretical perspective on deception detection: On the psychology of instrumental mind-reading. Psychology, Crime & Law 14: 189–200.
Granhag, Par Anders and Leif A. Strömwall. 1999. Repeated interrogations: Stretching the deception detection paradigm. Expert Evidence: The International Journal of Behavioural Sciences in Legal Contexts 7: 163–174.
Granhag, Par Anders and Leif A. Strömwall. 2002. Repeated interrogations: Verbal and nonverbal cues to deception. Applied Cognitive Psychology 16: 243–257.
Granhag, Par Anders, Leif Strömwall, Rebecca Willén and Maria Hartwig. In press. Eliciting cues to deception by tactical disclosure of evidence: The first test of the Evidence Framing Matrix. Legal and Criminological Psychology.
Greuel, Louise. 1992. Police officers’ beliefs about cues associated with deception in rape cases. In: Friedrich. Lösel, Doris Bender and Thomas Bliesener (eds.), Psychology and Law: International Perspective, 234–239). Berlin, Germany: Walter de Gruyter.
Hartwig, Maria, Par Anders Granhag and Leif Strömwall, Leif. 2007. Guilty and innocent suspects’ strategies during interrogations. Psychology, Crime, & Law 13: 213–227.
Hartwig, Maria, Par Anders Granhag, Leif Strömwall and Ola Kronkvist. 2006. Strategic use of evidence during police interrogations: When training to detect deception works. Law and Human Behavior 30: 603–619.
Hartwig, Maria, Par Anders Granhag,Leif Strömwall and Aldert Vrij. 2005. Detecting deception via strategic disclosure of evidence. Law and Human Behavior 29: 469–484.
Hillman, Jackie, Aldert Vrij and Samantha Mann. 2012. Um … they were wearing …: The effect of deception on specific hand gestures. Legal and Criminological Psychology 17: 336–345.
Hocking, John. E. and Dale G. Leathers. 1980. Nonverbal indicators of deception: A new theoretical perspective. Communication Monographs 47: 119–131.
Inbau, Fred E., John E. Reid, Joseph P. Buckley and Brian C. Jayne. 2013. Criminal interrogation and confessions 5th edition. Burlington, MA: Jones & Bartlett Learning.
Johnson, Robert R. 2006a. Confounding influences on police detection of suspiciousness. Journal of Criminal Justice 34: 435–442.
Johnson, Robert R. 2006b. Race and police reliance on suspicious nonverbal cues. Policing: An International Journal of Police Strategies & Management 30: 277–290.
Kassin, Saul M., Chris A. Meissner and Rebecca J. Norwick. 2005. “I’d know a false confession if I saw one”: A comparative study of college students and police investigators. Law and Human Behavior 29: 211–227.
Kleinke, Chris L. 1986. Gaze and eye contact: A research review. Psychological Bulletin 100: 78–100.
Leal, Sharon, Aldert Vrij, Samantha Mann and Ronald Fisher, Ronald. 2010. Detecting true and false opinions: The Devil’s Advocate approach as a lie detection aid. Acta Psychologica 134: 323–329.
Leins, Drew A., Ronald P. Fisher and Stephen J. Ross. 2013. Exploring liars’ strategies for creating deceptive reports. Legal and Criminological Psychology 18: 141–151.
Leo, Richard A. 1996. Inside the interrogation room. Journal of Criminal Law and Criminology 86: 266–303.
Levine, Timothy R. 2010. A few transparent liars: Explaining 54% accuracy in deception detection experiments. In: Charles Salmon (ed.), Communication Yearbook, Volume 34, 41–61. Newbury Park, CA: Sage.
Lord, Charles G., Lee Ross and Mark R. Lepper, Mark R. 1979. Biased assimilation and attitude polarization: The effects of prior theories on subsequently considered evidence. Journal of Personality and Social Psychology 37: 2098–2109.
Mann, Samantha, Aldert Vrij and Ray Bull. 2002. Suspects, lies and videotape: An analysis of authentic high-stakes liars. Law and Human Behavior 26: 365–376
Mann, Samantha, Aldert Vrij and Ray Bull. 2004. Detecting true lies: Police officers’ ability to detect deceit. Journal of Applied Psychology 89: 137–149.
Mann, Samantha, Aldert Vrij, Dominic Shaw, Sharon Leal, Sarah Ewens, Jackie Hillman, Par Anders Granhag and Ronald P. Fisher. In press. Two heads are better than one? How to effectively use two interviewers to elicit cues to deception. Legal and Criminological Psychology.
Mehrabian, Albert. 1981. Silent Messages: Implicit Communication of Emotions and Attitudes (Second Edition). Belmont, CA: Wadsworth Publishing Co.
Meissner, Christian A. and Saul M. Kassin. 2002. “He’s guilty!”: Investigator bias in judgments of truth and deception. Law and Human Behavior 26: 469–480.
Meissner, Christian A., Alison D. Redlich, Sujeeta Bhatt and Susan Brandon. In press. Interview and interrogation methods and their effects on true and false confessions. Campbell Systematic Reviews.
Miller, Gerald R. and James B. Stiff. 1993. Deceptive Communication. Newbury Park, CA: Sage.
Moston, Stephen J., Geoffrey M. Stephenson and Thomas M. Williamson. 1993. The incidence, antecedents and consequences of the use of the right to silence during police questioning. Criminal Behaviour and Mental Health 3: 30–47.
Nahari, Galit, Aldert Vrij and Ronald P. Fisher. 2012. Does the truth come out in the writing? SCAN as a lie detection tool. Law & Human Behavior 36: 68–76.
National Research Council. 2003. The polygraph and lie detection. Committee to Review the Scientific Evidence on the Polygraph. Washington, DC: The National Academic Press.
Noller, Patricia. 1985. Video primacy – a further look. Journal of Nonverbal Behavior 9: 28–47.
Ofshe, Richard J. and Richard A. Leo, Richard A. 1997. The decision to confess falsely: Rational choice and irrational action. Denver University Law Review 74: 979–1112.
Park, Hee Sun, Timothy R. Levine, Steven A. McCornack, Kelly Morrisson and Merissa Ferrara. 2002. How people really detect lies. Communication Monographs 69: 144–157.
Porter, Stephen, Laura England, Marcus Juodis, Leanne ten Brinke, and Kevin Wilson. 2008. Is the face the window to the soul?: Investigation of the accuracy of intuitive judgments of the trustworthiness of human faces. Canadian Journal of Behavioural Science 40: 171–177.
Porter, Stephen and Leanne ten Brinke. 2008. Reading between the lies: Identifying concealed and falsified emotions in universal facial expressions. Psychological Science 19: 508–514.
Porter, Stephen, Michael Woodworth and Angela R. Birt. 2000. Truth, lies, and videotape: An investigation of the ability of federal parole officers to detect deception. Law and Human Behavior 24: 643–658.
Rice, Marnie and Grant T. Harris. 2005. Comparing Effect Sizes in Follow-Up Studies: ROC Area, Cohen’s d, and r. Law & Human Behavior 29: 615–620.
Riggio, Ronald E., Joan Tucker and Barbara Throckmorton. 1988. Social skills and deception ability. Personality and Social Psychology Bulletin 13: 568–577.
Ruby, Charles L. and John C. Brigham. 1997. The usefulness of the criteria-based content analysis technique in distinguishing between truthful and fabricated allegations. Psychology, Public Policy, and Law 3: 705–737.
Strömwall. Leif A., Par Anders Granhag and Maria Hartwig. 2004. Practitioners’ beliefs about deception. In: Par Anders Granhag and Leif A. Strömwall (eds.), Deception Detection in Forensic Contexts, 229–250. Cambridge, England: Cambridge University Press.
Strömwall, Leif A., Par Anders Granhag and Anna Carin Jonsson. 2003. Deception among pairs: ‘Let’s say we had lunch together and hope they will swallow it’. Psychology, Crime, & Law 9: 109–124.
Vrij, Aldert. 1995. Behavioral correlates of deception in a simulated police interview. Journal of Psychology: Interdisciplinary and Applied 129: 15–29.
Vrij, Aldert. 2004. Invited article: Why professionals fail to catch liars and how they can improve. Legal and Criminological Psychology 9: 159–181.
Vrij, Aldert. 2006. Challenging interviewees during interviews: The potential effects on lie detection. Psychology, Crime, & Law 12: 193–206.
Vrij, Aldert. 2007. Deception: A social lubricant and a selfish act. In: Klaus Fiedler (ed.), Frontiers of Social Psychology: Social Communication 309–342. New York: Psychology Press.
Vrij, Aldert. 2008. Detecting Lies and Deceit: Pitfalls and Opportunities, Second Edition. Chichester: John Wiley and Sons.
Vrij, Aldert, Lucy Akehurst, Stavroula Soukara and Ray Bull. 2002. Will the truth come out? The effect of deception, age, status, coaching, and social skills on CBCA scores. Law and Human Behaviour 26: 261–283.
Vrij, Aldert and Par Anders Granhag. 2007. Interviewing to detect deception. In: Sven A. Christianson (ed.), Offenders’ memories of violent crime 279–304. Chichester, England: John Wiley & Sons, Ltd.
Vrij, Aldert and Par Anders Granhag. 2012. Eliciting cues to deception and truth: What matters are the questions asked. Journal of Applied Research in Memory and Cognition 1: 110–117.
Vrij, Aldert, Par Anders Granhag and Samantha Mann. 2010. Invited article: Good liars. The Journal of Psychiatry and Law 38: 77–98.
Vrij, Aldert, Par Anders Granhag, Samantha Mann and Sharon Leal. 2011. Lying about flying: The first experiment to detect false intent. Psychology, Crime, and Law 17: 611–620.
Vrij, Aldert, Wendy Kneller and Samantha Mann. 2000. The effect of informing liars about criteria-based content analysis on their ability to deceive CBCA-raters. Legal and Criminological Psychology 5: 57–70.
Vrij, Aldert, Sharon Leal, Samantha Mann and Oar Anders Granhag. 2011. A comparison between lying about intentions and past activities: Verbal cues and detection accuracy. Applied Cognitive Psychology 25: 212–218.
Vrij, Aldert, Sharon Leal, Samantha Mann, Lara Warmelink, Par Anders Granhag, and Ronald P. Fisher. 2010. Drawings as an innovative and successful lie detection tool. Applied Cognitive Psychology 4: 587–594.
Vrij, Aldert, Samantha Mann, Ronald Fisher, Sharon Leal, Rebecca Milne and Ray Bull. 2008. Increasing cognitive load to facilitate lie detection: The benefit of recalling an event in reverse order. Law and Human Behavior 32: 253–265.
Vrij, Aldert, Samantha Mann, Sharon Leal and Ronald P. Fisher. 2010. “Look into my eyes”: Can an instruction to maintain eye contact facilitate lie detection? Psychology, Crime, & Law 16: 327–348.
Vrij, Aldert, Frans Willem Winkel. 1991. Cultural patterns in Dutch and Surinam nonverbal behavior: An analysis of simulated police/citizen encounters. Journal of Nonverbal Behavior 15: 169–184.
Vrij, Aldert & Frans Willem Winkel. 1992. Cross-cultural police-citizen interactions: The influence of race, beliefs and nonverbal communication on impression formation. Journal of Applied Social Psychology 22: 1546–1559.
Warmelink, Lara, Aldert Vrij, Samantha Mann, Shyma Jundi and Par Anders Granhag. In press. Have you been there before? The effect of experience and question expectedness on lying about intentions. Acta Psychologica.
Zuckerman, Miron, Bella M. DePaulo, and Robert Rosenthal. 1981. Verbal and nonverbal communication of deception. In: Leonard Berkowitz (ed.), Advances in Experimental Social Psychology, volume 14, 1–57. New York: Academic Press.