Police Use of Nonverbal Behavior as Indicators of Deception

Police Use of Nonverbal Behavior as Indicators of Deception

People rely on various sources of information when they form impressions about others. They could pay attention to various characteristics of the tar-get person, such as gender (Hall & Carter, 1999; Stangor, Lynch, Changming & Glass, 1992), age (Hargie & Tourish, 1999; Hummert, 1999), race (Brown, 1995; Ruby & Brigham, 1996), dialect (Giles & John son, 1986; Giles & Powesland, 1975; Street & Hopper, 1982), dress (Vrij, 1993), clothing (Flank & Gilovich, 1988; Vrij, 1997), and facial appear­ance (Bull & Rumsey, 1988). They also could examine what people actually say (speech content, Krauss & Chiu, 1998; Steller & Kohnken, 1989) or observe their behavior (DePaulo & Friedman, 1998). In this article we pri­marily focus on the impact of nonverbal communication on impression formation. Nonverbal behavior does not only include body language, such as movements people make, smiling gaze behavior, etc., but also vocal characteristics, such as speech rate, speech pauses, uhms and ers, pitch of voice, etc. In addition, we primarily focus on a specific area within impres sion formation, which is the judgment about whether or not someone is ly­ing (we will use the words deception and lying interchangeably). As Horvath, Jayne and Buckley (1994) pointed out, making judgements about the veracity of statements is an important aspect of police work.

Correspondence should be addressed to: Aldert Vrij, University of Portsmouth, Psy­chology Department, King Henry Building, King Henry 1 Street, Portsmouth, Ml 2DY, United Kingdom or via email: aldert.vrij@port.ac.uk Our studies concerning real life police interviews which are discussed in this article (Mann, Vrij, & Bull, 2002, in press) were sponsored by the Economic and Social Research Council (Grant R00429734727).

first address the question of whether people pay more attention to speech content or nonverbal communication when they form im­pressions of others. We do not consider the other characteristics (age, gender, etc.) as researchers typically just focus on speech content and nonverbal communication when addressing this issue. This does not mean that the other characteristics do not play a role in impression formation. On the contrary, in our research, for example, conducted in the Netherlands (Vrij, 1993), we found that Dutch police detectives take into account the way people are dressed when making veracity judgments. They judged people as more suspicious when they were untidily dressed. We will see that the impact of nonverbal communica­tion and speech content on impression formation is situation depend­ent, and that in some circumstances police officers primarily rely upon nonverbal displays. Reasons for this will be discussed.

Given the fact that nonverbal behavior plays a role in determining whether someone is lying or not, we then move on to the question of how suspects and people in general actually behave when they are lying. This section shows that no single cue to deception, such as Pinocchio's grow­ing nose, exists, but that some behaviors are more indicative than others.

In the third section of this article we discuss how police officers and people in general think liars behave. We will see that police officers think that several behaviors are associated with deception, however, many of those beliefs are in fact nondiagnostic. An important reason why police officers hold incorrect beliefs wi11 be discussed, as well as the negative consequences that may arise from these incorrect views. We wi11 show that certain groups of people are in a disadvantageous po­sition, and readily make a suspicious impression on police officers be-cause they naturally show the behavioral patterns police officers typically think liars display.

The fourth section of this article answers the question of how good police officers are at detecting truths and lies. Unsurprisingly, given the fact that they hold incorrect views about how liars behave, this section reveals that errors in veracity judgments are often made. However, there are individual differences and some police officers seem to be better lie detectors than others. We will also discuss whether any relationship ex­ists between accuracy in detecting truths and lies and the cues police of­ficers say they pay attention to while attempting to detect these truths and lies. Given the fact that lie detectors are often inaccurate, we discuss in the fifth, and final, section some thoughts regarding possibilities for future research in improving people's ability to detect deceit.

RELATIVE IMPACT OF SPEECH CONTENT AND NONVERBAL
COMMUNICATION ON IMPRESSION FORMATION

Mehrabian and his colleagues (Mehrabian, 1972; Mehrabian & Ferris, 1967; Mehrabian & Wiener, 1967) were among the first researchers to examine the relative impact of speech content and nonverbal commu-

nication on impression formation. The experiments were ingenious. A target person expressed an opinion in different ways and participants were asked to give their impressions about what that person's true opinion actually was. Sometimes the speech content and nonverbal communication of that target person were consistent (both positive, both neutral, or both negative), and sometimes inconsistent (speech content was positive, nonverbal communication was negative, etc.). On the basis of a series of experiments, Mehrabian concluded that nonver­bal communication had a substantially larger impact on forming im­pressions of others than speech content.

However, the relative impact of speech content and nonverbal com­munication depends on the situation. For example, the knowledge of the lie detector will influence how that person will interpret a situation. In one real life case (The Independent, Friday 20 July 2001, p. 3) Jeffrey Archer, a British politician, asked three journalists to leave his hotel room during a political party conference while he took a call from the Prime Minister. Another politician who saw the three journalists pacing up and down the corridor asked them what they were doing. He imme­diately realized that Archer had lied to the journalists and could not be speaking to the Prime Minister on the phone, because he knew that the Prime Minister was sitting on the conference platform at that very mo­ment. In another real life case (Mann, Vrij, & Bull, 2002), police officers asked a man who was suspected of murder about his whereabouts on a certain afternoon. The man explained in detail that he visited a market in a village near his home. The police detectives knew that this was a lie because, unknown to the suspect, the market had been cancelled on that particular day. In both examples by paying attention to the speech content the lie detectors compared what was known with what the target person said, and discovered that the person was lying. Also, when police officers hear different statements from the same person about a topic, or different statements from different people about a topic, they primar­ily tend to focus on speech content, checking for consistency between the different statements (Granhag & Stromwall, 1999, 2000, 2001a, b; Stromwall, Granhag, & Jonsson, 2003). Hence, police officers believe that there is a relationship between consistency and veracity (Akehurst, Kohnken, Bull, & Vrij, 1996; Granhag & Stromwall, 1999; Greuel, 1992; Stromwall & Granhag, 2003). More specifically, they believe that consistent statements are likely to be truthful and inconsistent state­ments are likely to be deceptive. However, the limited research in this area has shown that there is no such link between consistency and ve­racity, or at least, if there is a link, it is more likely to be in the opposite direction. Granhag, Stromwall and Reiman (2002) found that lying pairs (i.e., two people lying in collusion) were more consistent than truth telling pairs, and that single liars and single truth tellers were equally consistent over time.'

There is evidence that in situations where there is no information to check and only one statement is made, police officers, like the partici-

pants in Mehrabian's experiment, primarily pay attention to nonverbal communication when forming an impression (Greuel, 1992; Rozelle & Baxter, 1975; Vrij, Foppes, Volger, & Winkel, 1992; Walkley, 1985; Waltman, 1983). Meissner and Kassm (2002) pointed out that in Florida, Tom Sawyer, believed to be innocent but accused of sexual as­sault and murder, became a prime suspect because he appeared em­barrassed and his face flushed during an initial interview (see Ofshe, 1989, for a detailed description of the Tom Sawyer case). According to Kaufmann, Drevland, Wessel, Overskeid, and Magnussen (2003), judi­cial decisions are sometimes based on nonverbal communication, even when available evidence points in the other direction. They de-scribe a Norwegian court trial in which (p. 2) 'although the circum­stantial evidence of guilt was strong, the defendant (a financial adviser) was acquitted partly because his nonverbal behavior was confident without evasive eye movements of any sort.'

There are several reasons why in these circumstances people rely so much on nonverbal behavior. First, it might be more revealing than speech (DePaulo & Kirkendol, 1989; Vrij, 2000a). There are automatic links between strongly felt emotions and certain behaviors (Ekman, 1985), whereas there are no such links between emotions and speech content. Anger, for example, results in several cues, including narrow­ing of the lips. This might well give a lie away if an angry person denies being angry. Also, we are more practiced in using words than in using behavior (because we exchange information predominantly via words), and this practice makes perfect. The fact that words are more important than nonverbal behavior in the exchange of information makes people also more aware of what they are saying than of how they are behaving. Suspects during a police interview will probably know and remember most of what they have said. However, it will be more difficult for them to indicate exactly how they behaved, that is, which hand movements they made, what their voice tone was like, etc. Being aware of one's own be­havior is essential when effectively controlling behavior. This requires not only being aware of one's behavior when trying to control it but also knowing the behavior one normally exhibits. Finally, although people can refrain from speech, they can't be silent nonverbally. Suppose a sus pect realizes during his police interview that the police know more that he thought they did about his involvement in the crime. This probably implies that he will have to give a different story from that which he had planned to tell. Verbally he can afford a little rest to think of an appropri­ate response in this awkward situation. Nonverbally, however, there is no possibility of taking a rest. He will display behavior throughout the entire interview, even when he remains silent, and the police officers can observe and interpret this behavior.

Second, sometimes there is little speech content to rely upon be-cause the person just says a few words or just a couple of sentences. In such situations an observer has almost no other choice than to exam­ine someone's behavior.

Third, people may not know which verbal cues to pay attention to even when the target person speaks substantially. In our opinion, speech content can reveal a deception if only observers knew what to pay attention to. Our own research has consistently shown that more accurate truth/lie decisions can be made when both speech content and nonverbal behavior are taken into account instead of just speech content or just nonverbal communication (Vrij, Edward, Roberts, & Bull, 2000; Vrij, Akehurst, Soukara, & Bull, 2003; Vrij & Mann, in press). Also, Porter and his colleagues pointed out that verbal cues have clear potential in deception detection (Porter & Yuille, 1995, 1996; Porter, Yuille, & Birt, 2001; Porter, Yuille, & Lehman, 1999). See Vrij (2003) for a recent review of verbal indicators of deceit.²

The Behavior of Liars

DePaulo, Lindsay, Malone, Muhlenbruck, Charlton, and Cooper (2003) and Vrij (2000a) recently reviewed over 100 scientific studies regarding nonverbal indicators of deceit. Perhaps the main finding was that no single behavioral response was uniquely related to deception. In other words, a straightforward giveaway cue, similar to Pinocchio's growing nose, does not exist. However, the reviews also demonstrated that some behaviors are more likely to occur during deception than others, perhaps depending on three processes that a liar may experi ence: emotion, content complexity, and attempted behavioral control (DePaulo, Stone, & Lassiter, 1985; Vrij, 2000a; Zuckerman, DePaulo, & Rosenthal, 1981).^3.4 The mere fact that a person is lying will not re sult in any particular behavior, but liars might be nervous (emotional), might have to think hard to come up with a plausible and convincing answer (content complexity), or might try to control their behavior in order to give a credible impression (attempted behavioral control). The six behaviours, which, according to the literature reviews, are to some extent associated with deception, can all be explained with these processes. The reviews revealed that liars tend to speak with a higher pitched voice which might be the result of arousal experienced by liars (Ekman, Friesen, & Scherer, 1976). However, differences in pitch be­tween liars and truth tellers are usually very small and therefore only detectable with sophisticated equipment.⁵ Also, sometimes liars' voices sound more tense than truth tellers' voices, another result of arousal. The results concerning speech errors (word and/or sentence repetition, sentence change, sentence incompletions, slips of the tongue, and so on) and speech hesitations (use of speech fillers such as 'ah', 'um', 'er' and so on) show a conflicting pattern. In most studies an increase in such errors (particularly word and phrase repetitions) and hesitations have been found during deception. This increase might have been the result of liars having to think hard about their an swer. Alternatively, the increase might be the result of nervousness. In some studies, however, a decrease in speech errors and speech hesita-

tons occurred. There is some evidence that variations of lie complex­ity are responsible for these conflicting findings (Vrij & Heaven, 1999). Lies that are difficult to tell result in an increase in speech errors and speech hesitations (in line with the content complexity approach), whereas lies that are easy to tell result in a decrease in speech hesita­tons and speech errors (in line with the attempted control approach).

Liars tend to make fewer illustrators (hand and arm movements designed to modify and/or supplement what is being said verbally) and fewer hand and finger movements (non-functional movements of hands and fingers without moving the arms) than truth tellers. The de-crease in these movements might be the result of lie complexity. Cogni­tive load results in a neglect of body language, reducing overall animation (Ekman & Friesen, 1972). Ask people what they ate three days ago, and observe their behavior while they try to remember. Most people will sit still while thinking about the answer. The decrease in movements might also be the result of an attempt to control behavior. Liars may believe that movements wi11 give their lies away, and wi11 therefore move very deliberately and tend to avoid any movements which are not strictly essential. This may result in an unusual degree of rigidity and inhibition (i.e., an overcompensation of behavior), be-cause people normally make movements which are not essential (DePaulo & Kirkendol, 1989). Finally, the decrease in movements might be the result of lack of emotional involvement. In certain situa­tons some degree of emotional involvement is expected which is diffi­cult to pretend if the person actually does not feel that emotion. Thus, a mother who punishes her child for wrongdoing might not look sincere enough if she, in fact, was amused by the child's actions; and a person who falsely claims that he is going on a business trip may not show enough interest or enthusiasm when he discusses the business deal he is going to make.'

The literature reviews revealed that liars do not seem to show clear signs of nervousness, such as gaze aversion and fidgeting. This is per haps surprising because, as we wi11 see in the next section, at least in white Western cultures, there is a strong stereotypical belief amongst observers, including professional lie catchers, that liars look away and make grooming gestures (Akehurst et al., 1996; Vrij & Semin, 1996). However, perhaps the absence of signs of nervousness is noth­ing more than an artifact. Deception research has almost exclusively been conducted in university laboratories where participants (mostly college students) tell the truth or lie for the sake of the experiment. Perhaps in these laboratory studies the stakes (negative conse­quences of getting caught and positive consequences of getting away with the lie) are not high enough for the liar to elicit clear cues to de­ception (Miller & Stiff, 1993).

In order to raise the stakes in laboratory experiments, participants have been offered money if their lies are believed by observers (Vrij, 1995). In other studies, participants are told that theywill be observed

by a peer who wi11 judge their sincerity (DePaulo, Stone, & Lassiter, 1985b). In an attempt to raise the stakes even further, participants in Frank and Ekman's (1997) study were given the opportunity to 'steal' US$50. If they could convince the interviewer that they had not taken the money, they could keep all of it. If they took the money and the in­terviewer judged them as lying, they had to give the US$50 back and also lost their US$10 per hour participation fee. Moreover, some par­ticipants faced an additional punishment if they were found to be lying. They were told that they would have to sit on a cold, metal chair inside a cramped, darkened room labeled ominously XXX, where they would have to endure anything from 10 to 40 randomly sequenced 110-deci­bel starting blasts of white noise over the course of one hour.

A study like this should raise ethical concerns. Yet even despite the ethical issue, one might argue that the stakes in such a study are still not comparable with the stakes in some real life situations, such as during police interviews. Laboratory studies are probably not suitable for examining the responses in high-stake situations as raising the stakes to a comparable extent is not usually possible for ethical rea­sons. Therefore, examining how liars behave in high-stake real life sit­uations is highly desirable (Riggio, 1994). This has proven to be difficult and, as a result, behavioral examinations of real life high-stake situations are virtually non-existent. Researchers face three problems in particular.

First, it is difficult to obtain appropriate video footage. For example, it is only relatively recently that interviews with suspects have been vid­eotaped in a few constabularies in England and Wales, and even when they are videotaped, researchers are rarely given permission to ana­lyze those videotaped interviews.

Second, it is often difficult in real life cases to establish the so-called ground truth, that is, to obtain conclusive evidence that the person is lying or telling the truth. For example, when former U.S. President Bill Clinton testified before the Grand Jury on August 17, 1998, about his alleged sexual affair with Monica Lewinsky, did he lie during that interview? And if so, exactly when during that interview? We simply don't know.

Third, to establish whether someone is lying, a useful method might be to compare the response under investigation with a response the lie detector knows to be truthful. However, the two situations (situation under investigation and responses while telling the truth) should be comparable. Take as an example Clinton during his testimony before the Grand Jury in the Monica Lewinsky case. Betty Currie (who was Clinton's personal secretary) had gone to Monica Lewinsky's home to collect the presents she had received from Clinton. During the testi­mony Clinton was asked several times at different times during the in­terview whether or not he instructed Betty Currie to do this. When addressing this issue, Clinton displayed a remarkable pattern of be­havior that he didn't show in any other part of the interview. For exam-

pie, his posture became rigid, he sat very still and looked straight into the camera (Vrij, 1998, 2002). Why did he do that? Because he was ly­ing? Or did this particular question trigger a specific behavioral response? This latter explanation cannot be ruled out.

Davis and Hadiks (1995) analyzed the behavior of Iraq's President Saddam Hussein when he was interviewed by the journalist Peter Arnett during the Gulf War (1991). The interview was broadcast on CNN. They found that when discussing Israel, Hussein made specific movements with his left forearm. This behavioral pattern only emerged when he was discussing Israel and Zionism. In other words, specific situations sometimes result in specific behaviors. Research­ers (and lie detectors) should be aware of this and should compare sit­uations which are similar to avoid comparing apples with oranges. Unfortunately, during police interviews, apple-orange comparisons are sometimes made (Moston & Engelberg, 1993). In those interviews, suspects' behaviors during smalltalk conversations at the beginning of the interview are compared with their behavior during the actual inter rogation. Although police officers are advised to establish comparable truths in this way (Inbau, Reid, & Buckley, 1986; Inbau, Reid, Buckley, & Jayne, 2001), this is an inaccurate comparison. Smalltalk and the actual investigation are different situations. Not surprisingly, both guilty and innocent people tend to show different behaviors during smalltalk and the actual interview (Vrij, 1995). Researchers make the same mistake too. In a rare example of a real life high stake deception study, Hirsch and Wolf (2001) observed 23 nonverbal and verbal cues displayed by Clinton during his Grand Jury Testimony. They exam­ined a 23-minute segment of the videotape and compared this with 11 minutes of the same testimony when he answered basic questions (his name, the attorney's name, etc.). Significant differences were obtained for 19 cues. They also compared the 23-minute segment with 5 min­utes of a fund-raising speech to a sympathetic crowd. This time, 20 sig nificant differences emerged. Unfortunately, this study tells us nothing about cues to deception. First, the ground truth in the 23-minute seg ment has not been established. Second, the comparisons between this 23-fragment and the other fragments are apple-orange comparisons. The problem with comparing the actual interview with basic questions (smalitalk) has already been discussed above. In addition, it is obvious that people show completely different behaviors when they address a crowd in a fund-raising speech than when they are interviewed about an alleged affair. In this respect, it might be more surprising that significant differences were found for only 19 or 20 cues and not for all 23 cues.

Back in 1953, Reid and Arther published a study regarding the be havior of 486 verified guilty and 323 verified innocent people who were suspected of various criminal offences. Their analysis revealed several indicators of deception. However, no details are given about the ground truth in these cases, neither is information provided about the

actual interview setting where the observations took place (although the title of their article 'Behavior symptoms of lie-detector subjects' gives us some kind of a clue). Horvath (1973) included 100 suspects in his study of which 50 were verified truthful and 50 were verified lying. Suspects' speech content and nonverbal communication were ob­served during a pre-test interview stage of polygraph examinations. Analyses of their nonverbal behavior revealed several diagnostic cues. Again, no information about the ground truth has been provided.

In a recent study (Mann et al., 2002) we properly addressed these is sues. We have examined the behavior displayed by 16 suspects during their police interviews. The suspects were all being interviewed in con nection with serious crimes such as murder, rape, and arson. In this study the ground truth was established in all 16 cases and fair com­parisons between the suspects' truthful and deceptive behavior were made. Regarding the ground truth, clips of video footage were selected where other sources (reliable witness statements and forensic evi­dence) provided evidence that the suspect lied or told the truth. In ad­dition, for each suspect, truths and lies were chosen which were as comparable as possible in nature, for example, a suspect who gave a detailed description about how he had assisted in killing a person (truth), later denied any involvement in the crime (lie). Forensic evi­dence indisputably supported his original version. (See Mann et al., 2002, for further details regarding these ground truth and comparable truth issues.) Table 4.1 shows the results of the study for the total sam­ple (N = 16) and for male suspects only (N = 13).

As can be seen in Table 4.1, results revealed that the suspects in these high-stake situations did not show clear stereotypical nervous behaviors such as gaze aversion, increased speech disturbances, or in creased movements. In fact, they exhibited an increase in pauses and male suspects showed a decrease in hand and arm movements. This is more in agreement with the content complexity and attempted control approaches than with the emotional approach. The strongest evidence that content complexity affected suspects' behavior more than ner­vousness was the finding regarding eye blinks. Suspects made fewer eye blinks when they lied. Research has shown that nervousness re­sults in an increase in eye blinking (Harrigan & O^'Connell, 1996; Tecce, 1992), whereas increased cognitive load results in a decrease in eye blinking (Wallbott & Scherer, 1991).

The apparent predominance of cognitive load processes compared to emotional processes in suspect interviews is perhaps not surpris­ing. Many suspects have had previous regular contact with the police. Therefore, they are probably familiar with police interviews which might decrease their nervousness during that situation. However, sus pects in police interviews are often less intelligent than the average per son (Gudjonsson, 1992). There is evidence that less intelligent people will have particular difficulty in inventing plausible and convincing stories (Ekman & Frank, 1993).

Total sample (N = 16)  Males only (N = 13)

Gaze aversion

Eye blinks

Head movements

Hand/arm movements

Pauses                                             > > Speech disturbances^a

> increase during deception

- no difference

' speech fillers and speech errors combined

How do Police Officers Think Liars Behave?

Which behavioral and auditory cues do people think are associated with deception? In general terms, people find those behaviors that deviate from a normal or expected pattern suspicious (Basket & Freedle, 1974; Bond, Omar, Pitre, Lashley, Skaggs, & Kirk, 1992; Burgoon, Buller, Ebesu, White, & Rockwell, 1996). Thus, eye contact which is either lacking or lasting too long, or pauses which are either too short or too long, etc. all make a suspicious impression. In sev­eral studies people were explicitly asked to indicate how they think li ars behave (see Vrij, 2000a, for a review of such studies). In some of these studies, conducted in various Western countries such as Swe­den, the Netherlands, and the United Kingdom, the same questions have been asked to both laypersons and professional lie catchers such as police officers (Akehurst et al., 1996; Stromwall & Granhag, 2003; Taylor & Vrij, 2000; Vrij & Semin, 1996; Vrij & Taylor, 2003). The answers given by these different groups of respondents were re­markably similar. It appears that there is common belief, at least among Western white people, about how liars behave. Results showed that observers associate deception with a high-pitched voice, many speech hesitations and speech errors, a slow speech rate, a long latency period (period of silence between question and answer), many pauses, gaze aversion, a lot of smiling, and an increase in move­ments. Many of these behaviors are indicators of nervousness. Ap­parently, the stereotypical belief is that liars are nervous and will behave accordingly. The surveys have indicated that people, both laypersons and police officers, particularly associate gaze aversion and fidgeting with deception. For example, around 75% of police offi-

cers believe that liars look away' (Mann, Vrij, & Bull, in press; Vrij & Semin, 1996) and that they make grooming gestures (Vrij & Semin, 1996). As we saw earlier, most of these behaviors are not related to deception (such as gaze aversion) or are related to deception in a dif­ferent way (for example, illustrators and hand/finger movements tend to decrease during deception rather than increase).

One of the reasons why such incorrect views on deception exist is that people, including police officers, are taught these wrong cues (Gordon, Fleisher, & Weinberg, 2002; Hess, 1997; Inbau et al., 1986; Yeschke, 1997; Zulawski & Wicklander, 1993). Inbau et al. (2001) re­cently published a new edition of their manual Criminal interroga­tion and confessions which is an update of their 1986 version (Inbau et al., 1986). This manual is highly influential and, as Inbau et al. (2001) point out in their preface, thousands of investigators have been trained to use the techniques contained within their book. Un­fortunately, the views described in their book about deceptive indica­tors of deception are wrong. They describe in detail how, in their view, liars behave. This includes behaviors such as showing gaze aversion, displaying unnatural posture changes, fidgeting and placing a hand over the mouth or eyes when speaking. They based their view on their extensive experience with interviewing suspects. However, none of these behaviors are found to be reliably related to deception in decep­tion research. Neither do Inbau and his colleagues provide any em­pirical evidence for their claims. Kassin and Fong (1999), however, trained their participants to look at the cues Inbau and colleagues claim to be related to deception, and compared the performance of this group of participants on a subsequent lie detection task with a group of naive observers who received no information at all. The trained participants performed significantly worse compared to the naive observers, which is not surprising given the poor level of train­ing these participants had received. More academics have expressed their concern about police training regarding lie detection (Ekman, 1985; Moston, 1992; Granhag & Stromwall, 1999).

Having incorrect views about cues to deception may have serious consequences for some groups of people, namely those whose natural behavior fits the Western white stereotype of deceptive behavior. Some individuals' nonverbal behavior gives the impression that they are tell­ing the truth (honest demeanor bias), whereas others' natural behav­ior leaves the impression that they are lying (dishonest demeanor bias) (Riggio, Tucker, & Throckmorton, 1988; Riggio, Tucker, & Widaman, 1987; Vrij, 1993; Vrij & Van Wijngaarden, 1994; Vrij & Winkel, 1992b; Zuckerman, DeFrank, Ha11, Larrance, & Rosenthal, 1979). This is re­lated to personality traits. Expressive people, for example, exude credibility, regardless of the truth of their assertions. It is not that they are particularly skilled at lying, but their spontaneity tends to disarm suspicion, which makes it easier for them to get away with their lies (Riggio, 1986). On the other hand, people with a strong sense of public

self-consciousness tend to make a less credible impression on others, regardless of whether they are telling the truth. These are individuals who, while lying, are concerned about being scrutinized by others, which changes their behavior in such a way that it appears dishonest. Introverts and socially anxious people also impress others as being less credible. The social clumsiness of introverts and the impression of tension, nervousness or fear that is natural to socially anxious individuals is interpreted by observers as indicators of deception.

Interestingly, their demeanor seems not to accurately reflect their behavior. For example, introverted people do not lie frequently (Kashy & DePaulo, 1996). Introverts also commit fewer crimes than extraverts (Eysenck, 1984). Furthermore, socially anxious people are less likely to persist in lying as soon as they are challenged (Vrij & Holland, 1998).

People also differ in how they present themselves when discussing emotional experiences. For example, research with rape victims has distinguished two basic styles of self-presentation, an 'expressed' style in which the victim displays distress which is clearly visible to outsid ers, and a more controlled 'numbed' style in which cues of distress are not clearly visible (Burgess, 1985; Burgess & Homstrom, 1974). Al-though the styles represent a personality factor and are not related to deceit (Littman & Szewczyk, 1983), they have a differential impact on the perceived credibility of victims. Emotional victims are more readily believed than victims who report their experience in a more controlled manner (Baldry, Winkel, & Enthoven, 1997; Kaufmann, Drevland, Wessel, Overskeid, & Magnussen, 2003; Vrij & Fisher, 1997; Winkel & Koppelaar, 1991).

Various ethnic minority groups are also in a disadvantageous posi tion due to the behavior they naturally display. Afro-American people display more gaze aversion than white American people (LaFrance & Mayo, 1976), and people from Turkey and Morocco who are living in the Netherlands show more gaze aversion than native Dutch people (Vrij, 2000a; Vrij, Dragt, & Koppelaar, 1992). Such differences are caused by the fact that gaze patterns are influenced by culture, and that looking into the eyes of a conversation partner is regarded as polite in Western cultures but is considered to be rude in several other cultures (Vrij & Winkel, 1991; Vrij, Winkel, & Koppelaar, 1991).

In the Netherlands, we examined the nonverbal behavioral patterns of white native Dutch and black Surinam citizens (citizens originated from Surinam, a former Dutch colony, but now living in the Nether-lands) during simulated police interviews (Vrij & Winkel, 1991). Both a Dutch and a Surinamese interviewer were used, but this had no im­pact on the findings. Surinam people made more speech disturbances (speech fillers such as 'ah', 'um', 'er' and stutters), exhibited more we aversion, smiled more often, and made more self manipulations (scratching the head, wrists, and so on) and illustrators (hand and arm movements designed to modify and/or supplement what is being

said verbally), regardless of whether they were lying or not. These be haviors show an overlap with the behaviors Western white people be­lieve liars display, suggesting that typical 'Surinam' behavior in experiments in Holland correspond with behavior that makes a suspi cious impression on Western white observers. This gives rise to possi­ble cross-cultural nonverbal communication errors during cross-cultural interactions. That is, nonverbal behavioral patterns that are typical for Surinam people in these settings may be interpreted by Western white observers as revealing attempts to hide the truth. We tested this idea in a series of experiments (Vrij & Winkel, 1992a, 1994; Vrij et al., 1991). Videotapes were made of simulated police interviews in which native Dutch and Surinam actors participated. Different ver­sions were made of each interview. The actors demonstrated typical 'Dutch' behavior in one version of the interviews (for example, showed a limited amount of gaze aversion) and typical 'Surinam' nonverbal behavior in another version of the interviews (showed more gaze aver sion). Dutch white police officers were exposed to one version of each interview and were asked to indicate to what extent the actor made a suspicious impression. The actors consistently made a more suspi­cious impression when they demonstrated 'typical Surinam behavior' than when they exhibited 'typical Dutch behavior.' These findings sup-port the assumption that cross-cultural nonverbal communication errors do occur during cross-cultural interactions, and that nonverbal behavioral patterns that are typical for an ethnic group are often interpreted by Western white observers as signs of deception.

POLICE OFFICERS' ABILITY TO DETECT LIES

In scientific studies concerning detection of deception, observers are typically given videotaped or audiotaped statements of a number of people who are either lying or telling the truth. After each statement ob­servers (typically college students) are asked to judge whether the statement is truthful or false. In such tasks, guessing whether some one is lying or not gives a 50% chance of being correct. Vrij (2000a) has reviewed 37 lie detection studies in which the observers were college students. The total accuracy rate, the percentage of correct answers, was 56.6%, which is only just about the level of chance. The review fur ther revealed that people are to some extent capable of detecting truths (i.e., correctly judging that someone is telling the truth: 67% accuracy rate) but particularly poor at detecting lies (i.e., correctly judging that someone is lying: 44% accuracy rate). In fact, 44% is below the level that could be obtained by chance alone. In other words, people would be more accurate in detecting lies if they simply guessed. The superior accuracy rate for truthful messages is caused by the truth bias: judges are more likely to consider that messages are truthful rather than de ceptive, and as a result, truthful messages are identified with more ac-curacy than deceptive ones. There are at least four possible

explanations for the truth bias. First, in daily life people are more often confronted with truthful than with deceptive statements, so they are therefore more inclined to assume that the behavior they observe is honest (the so-called availability heuristic; O'Sullivan, Ekman, & Friesen, 1988). Second, social conversation rules prevent people from displaying suspicion. A person will very quickly become irritated if their conversation partner questions everything that they say. Unfortu­nately, it is often necessary to challenge what the person is saying and ask for more information in order to detect deceit (Vrij, 2000a). Third, the truth bias is the result of the incorrect stereotypical views people have about how liars behave. For example, most people expect liars to behave nervously (Akehurst et al., 1996; Stromwall & Granhag, 2003; Taylor & Vrij, 2000; Vrij & Semin, 1996; Vrij & Taylor, 2003). There-fore since many liars do not show nervous behaviors, observers who look for cues of nervousness to detect deceit, yet don't find them, will be inclined to judge many messages as truthful. Fourth, people may be unsure as to whether deception is in fact occurring. Given this uncer­tainty, the safest and most polite strategy may be to believe what is overtly expressed (DePaulo, Jordan, Irvine, & Laser, 1982).

It could be argued that college students are not habitually called upon to detect deception. Perhaps professional lie-catchers, such as police officers or customs officers, would obtain higher accuracy rates than laypersons. It might be that their experience in interviewing peo­ple and catching liars has had a positive influence on their skills in de­tecting deceit. Unfortunately, not many studies have been conducted with professional lie catchers, perhaps due to the fact that professional lie catchers in some countries are reluctant to participate in lie detec­tion studies or do not wish the outcomes to be published. The studies which have been published are reported in Table 4.2.

In a typical experiment (Ekman, O'Sullivan, & Frank, 1999), profes­sional lie catchers watched video clips of twenty people who gave a statement about a number of current controversial issues which either was their true opinion (truth) or an opinion opposite to their true opin­ion (lie). For each statement, the professional lie catchers were asked to indicate whether it was a truth or a lie. Three findings emerged from these studies. First, most accuracy rates were similar to the accuracy rates found in studies with college students as observers (most were in the 45%-60% range), suggesting that professional lie catchers are no better than laypersons in detecting deceit. DePaulo and Pfeifer (1986), Ekman and O'Sullivan (1991), Meissner and Kassin (2002) and Vrij and Graham (1997) directly tested this idea by including both laypersons and professional lie catchers as observers in their experi­ments. DePaulo and Pfeifer (1986), Meissner and Kassin (2002) and Vrij and Graham (1997) found that police officers were as (un)suc­cessful as university students in detecting deception. Second, some groups seem to be better than others. Ekman and O'Sullivan (1991) found that police officers (56% accuracy) and polygraph examiners

Table 4.2
Accuracy Scores of Professional Lie Catchers

Notes : Experienced and inexperienced officers together.

Police officers also took part in Garrido and Masip^'s (2001) lie detection task. However, no accuracy rates were reported.

In Horvath. Jayne and Buckley's (1994) study, (our persons 'trained and experienced in Behavior Analy­sis Interviewing' participated, reaching 78% truth accuracy and 66% lie accuracy rates. However, the professional background of these four lie detectors were not reported.

(56% accuracy) obtained similar accuracy rates to university students (53% accuracy), whereas members of the Secret Service were better at detecting lies than university students (64% accuracy). Ekman et al. (1999) found that U.S. Federal officers (police officers with a special in terest and experience in deception and demeanor) and sheriffs (police officers identified by their department as outstanding interrogators) were considerably better at detecting lies (73% and 67% accuracy re spectively) than 'mixed' law-enforcement officers (officers who had not been chosen because of their reputation as interrogators, 51% accu­racy). Third, the truth bias, consistently found in studies with stu­dents as observers, is much less profound or perhaps even lacking in studies with professional lie catchers. It might be that their jobs and daily experiences make professional lie catchers more cynical and suspicious.

Finally, DePaulo and Pfeifer (1986) and Meissner and Kassin (2002) investigated how confident observers were in the decisions they had made. They found that police officers were more confident than stu­dents, which suggests that being a professional lie catcher may in-crease confidence in the ability to detect deceit, but not accuracy. Allwood and Granhag (1997) pointed out that the tendency to be over-confident is not unique to police officers, but common among many different groups of professional lie catchers.

However, as we pointed out earlier, the poor ability to detect deceit that emerged from detection of deception studies might be the result of an artifact. Deception research has almost exclusively been conducted in university laboratories where participants (mostly college students) tell the truth or lie for the sake of the experiment. Perhaps in these lab-oratory studies the stakes are not high enough for the liar to elicit clear cues to deception, which makes the task for the lie detector difficult (Miller & Stiff, 1993). Indeed, in a series of experiments in which the stakes were manipulated (although the stakes were never really high), it has been found that high-stake lies were easier to detect than low-stake lies (DePaulo, Kirkendol, Tang, & O'Brien, 1988; DePaulo, Lanier, & Davis, 1983; DePaulo, LeMay, & Epstein, 1991; DePaulo et al., 1985b; Lane & DePaulo, 1999; Vrij, 2000b; Vrij, Harden, Terry, Edward, & Bull, 2001).

Given the low stake settings in these deception detection studies, one could are that they do not accurately measure police officers' ability to detect deceit. One could argue that the only valid way to inves­tigate police officers' true ability to detect deceit, is to examine their skills when they detect lies and truths told in real life criminal investi­gation settings. Vrij and Mann (2001a, b) were the first researchers to examine police officers' skills in such situations. Vrij and Mann (2001a) exposed police officers to fragments of a videotaped police in­terview with a man suspected of murder. However, that study had two limitations. First, fragments of only one suspect were shown, and he could have been displaying untypical behavior. That is, he could have been a particularly good or a particularly poor liar. Second, the police officers could not understand the suspect as he spoke in a foreign lan­guage (suspect and police officers were of different nationalities). Vrij and Mann (2001b) exposed judges to videotaped press conferences of people who were asking the general public for help either in finding their missing relatives or the murderers of their relatives. They all lied during these press conferences and they were all subsequently found guilty of having killed the 'missing person' themselves. This study had limitations as well. Most importantly, the judges were only subjected to lies. This is problematic as it does not investigate people's ability to de­tect truths. A good lie detector is good at distinguishing between truths and lies, that is, good at detecting lies and good at detecting truths. For example, a police officer who judges every clip he sees as a lie will have a perfect (1000) accuracy rate for detecting lies but a particularly low

(0%) accuracy rate for detecting truths. Thus, overall this is a poor lie detector. However, in a study where judges are only exposed to lies, such a suspicious lie detector will achieve a very high accuracy score.

In our most recent study (Mann et al., 2003) we overcame these limi­tations. We showed 99 police officers a total of 54 video clips of sus­pects who told truths and lies during their police interviews. These video clips were the clips we have already discussed in Table 4.1 (Mann et al., 2002). After each clip the police officers were requested to indicate whether the suspect was truthful or not. None of the sample of police officers belonged to one of the specific groups which have been identified by Ekman and his colleagues as being superior lie detectors. The study revealed accuracy rates which were higher than generally found in previous studies. The total accuracy rate was 65%, with a 64% truth accuracy rate and a 66% lie accuracy rate. All these accuracy rates were significantly higher than the 50% level of chance, and both the total accuracy rate and the lie accuracy rates were higher than the accuracy rates which have been found in most previous studies (see Vrij, 2000a, for a review).⁸ The accuracy rates were among the highest ever found with ordinary police officers. In other words, this study sug gests that ordinary police officers are better at detecting truths and lies than previous research has suggested.

Moreover, there were individual differences, and accuracy rates for individual officers varied from a low 30% to a very high 90% (achieved by three officers, Mann, 2001). How could we explain these individual differences? First, accuracy was not related to confidence. In other words, officers who expressed confidence about the accuracy of their judgments were not better than those who were more uncertain. Not obtaining a significant relationship between accuracy and confidence is by no means uncommon in deception research. Also, in previous de ception studies a significant relationship between accuracy and confi­dence was typically not found (see DePaulo, Charlton, Cooper, Lindsay, & Muhlenbruck, 1997, for a meta-analysis). However, neither did the police officers in our study show overconfidence in their ability. On the contrary, they underestimated their own performance. This finding is uncommon as overconfidence was found in several previous studies (see above). This exception might well be to do with how we measured confidence, as it was assessed somewhat differently from how it was investigated in previous studies (see Mann et al., 2003, for further details).

Second, findings showed a positive relationship between having ex perience in interviewing suspects and the ability to detect truths and lies, suggesting that experience does make police officers better able to distinguish between truths and lies. This finding has not been ob­tained in previous deception studies with professionals as observers (DePaulo & Pfeifer, 1986; Ekman & O'Sullivan, 1991; Porter et al., 2000). We believe that this finding was affected by the way we mea­sured experience. Other researchers, for example Ekman and

O'Sullivan (1991), used 'years of job experience' as a measurement for experience. Unfortunately, they did not further define 'job experi ence.' It might well be that this measurement has been too vague as it does not specifically focus on the officers' actual experience in situa­tions where they will attempt to detect deceit such as during interviews with suspects. There is little reason to suggest that a police officer who had worked for many years in a managerial or administrative position within the police force would be a better lie detector than someone with a similar position outside the police force. Therefore perhaps unsurprisingly, we also did not find a significant correlation between general job experience (i.e., 'years of service') and accuracy. In other words, experience might benefit truth and lie detection if only the rele vant experience is taken into account. Perhaps a weakness of the way we measured experience is that it was a self-report measure ('How ex­perienced do you consider yourself to be in interviewing suspects?') and not an objective measure. It would be interesting to see whether an objective measure of experience in interviewing suspects (for example, the number of suspect interviews a police officer has conducted) would correlate with accuracy as well. This would strengthen our argument. Unfortunately, the police do not record objective measures of experience with interviewing suspects.

Third, in our study, police officers were asked which cues they looked at when attempting to detect deceit. Several relationships oc­curred between cues mentioned by the officers as useful to detect de­ceit and their accuracy in truth and lie detection. First, good lie detectors mentioned story cues (vague reply, contradictions in story, etc.) more often than poor lie detectors. Second, the more body cues (gaze aversion, posture, movements, etc.) participants mentioned, the lower their accuracy became. Particularly police officers who men­tioned that liars look away and fidget were poor lie detectors. As men tioned earlier, Inbau and his colleagues (1986/2001) suggested that liars show a variety of body cues, including showing gaze aversion, dis playing unnatural posture changes, exhibiting self manipulations, and placing the hand over the mouth or eyes when speaking. We measured the effectiveness of using Inbau's views in our study by examining how good police officers who endorse these views were in our lie detection task. Interestingly, our findings showed that the more the police offi cers endorsed Inbau's views the worse they became at distinguishing between truths and lies. In other words, looking at Inbau et al.'s (1986/2001) cues is counterproductive.

In other studies the relationship between the cues people claim to pay attention to when attempting to detect deceit and their ability to distinguish between truths and lies were also examined. In one of our own studies (Vrij & Mann, 2001a) we obtained exactly the same out-come as in our recent Mann et al. (2003) study: Those participants who mentioned gaze aversion and fidgeting as cues to deceit achieved the lowest accuracy scores. When other researchers examined such re-

lationships different relationships emerged. For example, Ekman and O'Sullivan (1991) found that those participants who mentioned both speech cues and nonverbal cues obtained the highest accuracy, higher than those who just mentioned speech cues or nonverbal cues, whereas we found that those who just mentioned speech cues were the best lie detectors. Anderson, DePaulo, Mansfield, Tickle, and Greens (1999) and Feeley and Young (2000) found that the more vocal cues (speech errors, speech fillers, pauses, voice) participants mentioned, the higher accuracy they obtained. Porter, Woodworth and Birt (2000) found that the more body cues the participants reported, the better their ability to distinguish between truths and lies became. Frank and Ekman (1997) reported that good lie detectors were better at spotting brief facial expressions of emotions than poor lie detectors. Such micro expressions have not been investigated in any of the other studies discussed here.

In summary, different studies reveal different outcomes, and as a re sult of this, a clear picture of what distinguishes a good from a poor lie detector is yet to emerge. There are at least four explanations for the lack of consistency in the findings of different studies. One explanation is that the relationships between cues mentioned and accuracy are generally weak. Indeed, the significant correlations which are typically reported are low, usually falling into the r= .20 to r = .30 range. An-other explanation is that in different studies participants faced com­pletely different lie detection situations, and therefore, comparisons are difficult to make. For example, in most studies participants were requested to detect truths and lies in low-stake situations, whereas in some studies participants were exposed to high-stake situations. In­terestingly, in both our experiments (Mann et al., 2003; Vrij & Mann, 2001a) participants were requested to detect the truths and lies told during police interviews and both studies obtained similar findings. A third explanation is that perhaps lie detectors simply do not know where they look for deceit, and so lie detection could just be an intu­itive skill. We (Mann et al., 2003) obtained some support for this as­sumption, as good lie detectors reported relying significantly more often on 'gut feeling' than poor lie detectors. Finally, weak relation-ships between cues mentioned and ability to detect deceit and conflict ing findings between different studies may be caused by a flaw in the experimental designs used in lie detection studies. In almost all lie de tection studies published to date, people's skills to detect deceit were only tested once. The fact that they were good or bad at that particular task might have been a matter of luck, and there is certainly no guaran tee that those lie detectors would show the same performance if they were tested a second time. A better way of examining people's ability to detect deceit and the strategies good lie detectors use is to test the same people at various occasions, and to examine which detectors show a consistent performance. Particular attention can then be paid to the cues mentioned by those who are consistently good. To our

knowledge, only Frank and Ekman (1997) exposed lie detectors to multiple lie detection tasks and they reported some kind of consis­tency between the lie detectors' performances on the different tasks.

DIRECTIONS OF FURTHER RESEARCH IN LIE DETECTION

On the basis of the information provided in this article we propose two directions for further research. First, as mentioned in the paragraph above, judges should be exposed to multiple lie detection tasks instead of single lie detection tasks, as multiple tests probably will give more insight into someone's true ability to detect truths and lies than single lie detection tests. Only those judges with a consistently good perfor­mance in such tests should be considered good lie detectors. Then the next step would be to unravel the strategies these good lie detectors use. And perhaps the step after this would be to teach other lie detec­tors these successful strategies. Although this research might sound straightforward, it is, in fact, hazardous. First, it is unknown how many people will be consistently good at such lie detection tests, but probably not many. Therefore, many people need to be tested to obtain a reasonable sample of good lie detectors. Second, unraveling the strategies these good lie detectors use would not be easy either. Al-though researchers could directly ask good lie detectors which strate­gies they use, they probably would find this question difficult to answer, and perhaps do not even know which strategy they use. Rather than asking good lie detectors which strategies they use directly, this could be asked in an indirect way. For example, good lie detectors could be shown videotaped clips of liars and truth tellers and could be asked to indicate which fragments of these clips they consider relevant to their decision making. Researchers could then carefully examine these fragments selected by good lie detectors, particularly those frag­ments which were selected by many good lie detectors. However, it is not certain whether there will be consistency amongst good lie detec­tors in the fragments they select. Neither is it certain that such analyses will provide any meaningful (interpretable) information. Finally, these direct or indirect ways of unraveling lie detection strategies wi11 not work if good lie detection is primarily an intuitive skill (which cannot be ruled out, as argued above). Obviously, if good lie detection is pri­marily based upon intuition, teaching others to become better lie de­tectors would become problematic.

Second, rather than examining the strategies good lie detectors use, the extent to which certain interview techniques might facilitate lie de­tection could be investigated. The issue of how interview styles might benefit lie detection has been virtually ignored by academics to date, but this approach has potential in our view. For example, as we have discussed above, one reason why cues to deceit emerge is because liars experience more cognitive load than truth tellers. We have argued that this cognitive load aspect might be particularly important in police

suspects. Police interviewers could use this by employing interview techniques which increase the cognitive demand in suspects. This should have a greater effect on liars than on truth tellers, thus facilitating discrimination between them.

There are several ways in which cognitive demand could be in-creased. First, by asking suspects to elaborate on issues they have pre viously mentioned. This might be more difficult for liars than for truth tellers, especially if they didn't expect to be asked for such elaboration and, subsequently, haven't prepared responses to such questions prior to the interview. In that case, liars need to invent answers which sound plausible and convincing, and which do not contradict evi­dence. They also need to invent those spontaneous answers rather quickly; otherwise an usually long delay before answering the ques­tions might give the lie away. They also need to remember their inven tions in case the interviewer asks them to repeat what they have just said. Second, one valid speech content related cue frequently found in deception research is that truth tellers tend to tell their stories in a more unstructured way than liars (Vrij, 2000). That is, liars tend to tell their stories in a more fixed chronological order (this happened first, and then this, and then that, and so on) than truth tellers. It has been suggested that it is very difficult for liars to tell a fabrication in a non-chronological order (Steller, 1989; Kohnken, 1999; Zaparniuk, Yuille, & Taylor, 1995). Lie detectors could exploit this difficulty by asking interviewees to tell their stories in a non-chronological order, for example in reverse order. Asking people to tell their stories in re-verse order is a technique currently employed in police interviews as part of the Cognitive Interview (Fisher, Brennan, & McCauley, 2002; Fisher & Geiselman, 1992; Fisher, Geiselman, & Amador, 1989; Milne & Bull, 1999), however, it is used as a memory enhancing technique while interviewing witnesses which results in interviewees recalling more information about an event they have witnessed, rather than used as a technique to detect lies in suspects. It is worthwhile to inves tigate whether lie detection is facilitated by telling stories in reverse or-der. We expect this to be the case, as it will increase the cognitive demand more for liars than for truth tellers. One factor that might be relevant is rehearsal.

From our informal conversations with the police, we understand that police detectives have the impression that liars, more than truth tellers, rehearse their stories prior to police interviews. Liars may be lieve that good preparation of the made-up event is essential for them in order to make a credible impression during the police interviews, whereas truth tellers do not feel the same need to be prepared as they can simply explain what they have witnessed. Rehearsal could affect the liars' performance in two ways. It might actually be even more diffi cult for liars to tell a story in reverse order when they have prepared their false story in advance than when they are unprepared for fabri cating. If liars prepare themselves for an interview about an event they

probably will rehearse the event in forward order, because that is the natural way for events to unfold in time. If they are in fact then asked to describe the event in reverse order, they are forced to recall a story in a manner that goes against their own preparation. No rehearsal implies less commitment to a forward-order rendition, which should make it easier to reverse the ordering. Alternatively, lying in reverse order after rehearsal could be easier than being unprepared. This means that when the story is rehearsed a liar has easier access to the prepared details of the made-up event.

EPILOGUE

Police officers use nonverbal cues when they attempt to detect deceit. This is a hazardous exercise. Clear cut cues to deceit, the equivalent of Pinocchio's growing nose, do not exist so there is nothing lie detectors can really rely upon. Also, police officers have some incorrect beliefs about how liars behave, partly because they are misguided by police manuals which teach them the wrong cues. Relying on incorrect cues to deception has negative consequences for those whose natural be­havior embodies these incorrect beliefs, such as introverted and so­cially anxious people, and members of several ethnic minority groups. Despite this, some people appear to be better at distinguishing be­tween truths and lies than others, though a clear picture of what distin guishes a good from a poor lie detector has yet to emerge. Given the fact that detecting lies is an important aspect of several professions, for ex-ample police work, more insight into the strategies and techniques that could improve lie detection are welcome. We have given two sug­gestions for research: attempting to unravel the strategies used by good lie detectors, and employing interview techniques which could fa­cilitate distinguishing between liars and truth tellers. We consider the second line of research particularly fruitful.

REFERENCES

Akehurst, L., Kohnken, G., Vrij, A., & Bull, R. (1996). Lay persons' and police officers' beliefs regarding deceptive behaviour. Applied Cognitive Psychol­ogy, 10, 461-473.

Allwood, C. M., & Granhag, P. A. (1997). Feelings of confidence and the realism of confidence judgments in everyday life. In P. Juslin & H. Montgomery (Eds.), Judgment and decision making: Neo-Brunswikian and process-tracing approaches (pp. 123-146). Mahwah NJ: Lawrence Erlbaum Associates.

Anderson, D. E., DePaulo, B. M., Mansfield, M. E., Tickle, J. J., & Green, E. (1999). Beliefs about cues to deception: Mindless stereotypes or untapped wisdom? Journal of Nonverbal Behavior, 23, 67-89.

Baldry, A. C., Winkel, F W, & Enthoven, D. S. (1997). Paralinguistic and non verbal triggers of biased credibility assessments of rape victims in Dutch police officers: An experimental study of 'Nonevidentiary' bias. In S.

Redondo, V Garrido, J. Perze, & R. Barbaret (Eds.), Advances in psychol­ogy and law (pp. 163-174). Berlin: Walter de Gruyter.

Baskett, G. D., & FYeedle, R. O. (1974). Aspects of language pragmatics and the social perception of lying. Journal of Psycholinguistic Research, 3, 117-131.

Bond, C. E, Omar, A., Pitre, U., Lashley, B. R., Skaggs, L. M., & Kirk, C. T. (1992). Fishy-looking liars: Deception judgment from expectancy violation. Journal of Personality and Social Psychology, 63, 969-977.

Brown, R. (1995). Prejudice: Its social psychology. Cambridge, MA: Blackwell Publishers.

Bull, R., & Rumsey, N. (1988). The social psychology of facial appearance. New York: Springer Verlag.

Buller, D. B., & Burgoon, J. K. (1996). Interpersonal deception theory. Com­munication Theory, 6, 203-242.

Buller, D. B., Strzyzewski, K. D., & Hunsaker, F. G. (1991). Interpersonal de­ception II: The inferiority of conversational participants as deception detec­tors. Communication Monographs, 58, 25-40.

Burgess, A. W. (1985). Rape and sexual assault: A research book. London: Garland.

Burgess, A. W., & Homstrom, L. L. (1974). Rape: Victims of crisis. Bowie: Brady. Burgoon, J. K., Buller, D. B., Ebesu, A. S., White, C. H., & Rockwell, P. A. (1996).

Testing interpersonal deception theory: Effects of suspicion on communica‑

tion behaviors and perception. Communication Theory, 6, 243-267. Davis, M., & Hadiks, D. (1995). Demeanor and credibility. Semiotica,106, 5-54. DePaulo, B. M., Charlton, K., Cooper, H., Lindsay, J. L., & Muhlenbruck, L.

(1997). Personality and Social Psychology Review, 1, 346-357. DePaulo, B. M., & Friedman, H. S. (1998). Nonverbal communication. In D. T.

Gilbert, S. T. Fiske, & G. Lindzey (Eds.), The handbook of social psychol‑

ogy (pp. 3-40). Boston, MA: McGraw-Hill.

DePaulo, B. M., Jordan, A., Irvine, A., & Laser, P. S. (1982). Age changes in the detection of deception. Child Development, 53, 701-709.

DePaulo, B. M., & Kirkendol, S. E. (1989). The motivational impairment effect in the communication of deception. In J. C. Yuille (Ed.), Credibility assess­ment (pp. 51-70). Dordrecht, the Netherlands: Kluwer.

DePaulo, B. M., Kirkendol, S. E., Tang, J., & O'Brien, T. P. (1988). The motiva­tional impairment effect in the communication of deception: Replications and extensions. Journal of Nonverbal Behavior, 12, 177-201.

DePaulo, B. M., Lanier, K., & Davis, T. (1983). Detecting the deceit of the moti‑
vated liar. Journal of Personality and Social Psychology, 45, 1096-1103.

DePaulo, B. M., LeMay, C. S., & Epstein, J. A. (1991). Effects of importance of success and expectations for success on effectiveness at deceiving. Person­ality and Social Psychology Bulletin, 17, 14-24.

DePaulo, B. M., Lindsay, J. L., Malone, B. E., Muhlenbruck, L., Charlton, K., & Cooper, H. (2003). Cues to deception. Psychological Bulletin, 129, 74-118.

DePaulo, B. M., & Pfeifer, R. L. (1986). On-the-job experience and skill at de‑
tecting deception. Journal of Applied Social Psychology, 16, 249-267.

DePaulo, B. M., Stone, J. L., & Lassiter, G. D. (1985a). Deceiving and detecting deceit. In B. R. Schenkler (Ed.), The self and social life (pp. 323-370). New York: McGraw-Hill.

DePaulo, B. M., Stone, J. I., & Lassiter, G. D. (1985b). Telling ingratiating lies: Effects of target sex and target attractiveness on verbal and nonverbal decep tive success. Journal of Personality and Social Psychology, 48, 1191-1203.

Ekman, P (1985). Telling lies: Clues to deceit in the marketplace, politics and marriage. New York: W. W. Norton & Company. (Reprinted in 1992 and 2001).

Ekman, P, & Frank, M. G. (1993). Lies that fail. In M. Lewis & C. Saarni (Eds. ), Lying and deception in everyday life (pp. 184-201). New York: Guilford Press.

Ekman, P, & Friesen, W. V. (1969). Nonverbal leakage and clues to deception. Psychiatry, 32, 88-106.

Ekman, P, & Friesen, W. V. (1972). Hand movements. Journal of Communica­tion, 22, 353-374.

Ekman. P, Friesen, W. V.. & Scherer, K. R. (1976). Body movement and voice pitch in deceptive interaction. Semiotica, 16, 23-27.

Ekman. P. & O'Sullivan. M. (1991). Who can catch a liar? American Psycholo­gist, 46, 913-920.

Ekman, P, O'Sullivan, M., & Frank, M. G. (1999). A few can catch a liar. Psy­chological Science, 10, 263-266.

Eysenck, H. J. (1984). Crime and personality. In D. J. Muller, D. E. Blackman. & A. J. Chapman (Eds.), Psychology and Law (pp. 85-100). New York: John Wiley & Sons.

Feeley. T. H.. & deTurck. M. A. (1997). Perceptions of communication as seen by the actor and as seen by the observer: The case of lie detection. per presented at the International Communication Association Annual Confer­ence. Montreal, Canada.

Feeley, T. H., & Young, M. J. (2000). The effects of cognitive capacity on beliefs about deceptive communication. Communication Quarterly, 48, 101-119.

Fisher, R. P, Brennan, K. H., & McCauley, M. R. (2002). The cognitive interview method to enhance eyewitness recall. In M. L. Eisen, J. A. Quas, & G. S. Goodman (Eds.). Memory and suggestibility in the forensic interview (pp. 265-286). Mahwah, NJ: Lawrence Erlbaum Associates.

Fisher, R. P. & Geiselman, R. E. (1992). Memory-enhancing techniques for in­vestigative interviewing: The cognitive interview. Springfield, IL: Charles Thomas.

Fisher, R. P, Geiselman, R. E., & Amador. M. (1989). Field test of the cognitive interview: Enhancing the recollection of actual victims and witnesses of crime. Journal of Applied Psychology, 74, 722-727.

Frank, M. G. & Ekman, P (1997). The ability to detect deceit generalizes across different types of high-stake lies. Journal of Personality and Social Psy­chology, 72, 1429-1439.

Frank, M. G., & Gilovich, T. (1988). The dark side of self- and social percep­tion: Black uniforms and aggression in professional sports. Journal of Per­sonality and Social Psychology, 54, 74-85.

Garrido, E., & Masip, J. (2001). Previous exposure to the sender's behavior and accuracy at judging credibility. In R. Roesch, R. R. Corrado, & R. Dempster (Eds.), Psychology in the courts: International advances in knowledge (pp. 271-287). London: Routledge.

Giles, H., & Johnson, P (1986). Perceived threat, ethnic commitment and interethnic language behavior. In Y. Y. Kim (Ed.), Interethnic communica­tion: Current research. Beverly Hills, CA: Sage.

Giles, H., & Powesland, P F. (1975). Speech style and social evaluation. Lon-don: Academic Press.

Gordon, N. J., Fleisher, W. L., & Weinberg, C. D. (2002). Effective interviewing and interrogation techniques. San Diego, CA: Academic Press.

Granhag, P A., & Stromwall, L. A. (1999). Repeated interrogations: Stretching the deception detection paradigm. Expert Evidence: The International Journal of Behavioural Sciences in Legal Contexts, 7, 163-174.

Granhag, P A., & Stromwall, L. A. (2000). Effects of preconceptions on decep­tion detection and new answers to why lie catchers often fail. Psychology, Crime, & Law, 6, 197-218.

Granhag, P. A., & Stromwall, L. A. (2001a). Deception detection: Examining the consistency heuristic. In C. M. Breur, M. M. Kommer, J. F. Nijboer, & J. M. Reijntjes (Eds.), New trends in criminal investigation and evidence (Vol. 2, pp. 309-321). Antwerpen: Intresentia.

Granhag, P A., & Stromwall, L. A. (2001b). Deception detection: Interrogators' and observers' decoding of consecutive statements. The Journal of Psy­chology, 135, 603-620.

Granhag, P A., Stromwall, L. A., & Reiman, A. C. (2002). Partners in crime: How liars in collusion betray themselves. Manuscript submitted for publi­cation.

Greuel, L. (1992). Police officers' beliefs about cues associated with deception in rape cases. In F. Losel, D. Bender, & T. Bliesener (Eds.), Psychology and Law: International perspectives (pp. 234-239). Berlin, Germany: Walter de Gruyter.

Gudjonsson, G. H. (1992). The psychology of interrogations, confessions and testimony. Chichester, England: Wiley & Sons.

Hall, J. A., & Carter, J. D. (1999). Gender-stereotype accuracy as an individual difference. Journal of Personality and Social Psychology, 77, 350-359.

Hargie, O., & Tourish, D. (1999). The psychology of interpersonal skill. In A. Memon & R. Bull (Eds.), Handbook of the psychology of interviewing (pp. 71-88). Chichester: Wiley.

Harrigan, J. A., & O'Connell, D. M. (1996). Facial movements during anxiety

states Personality and Individual Dierences, 21, 205-212.

Hess, J. E. (1997). Interviewing and interrogation for law enforcement.

Reading, United Kingdom: Anderson Publishing Co.

Hirsch, A. R., & Wolf, C. J. (2001). Practical methods for detecting mendacity: A case study. The Journal of the American Academy of Psychiatry and the Law, 29, 438-444.

Horvath, F. (1973). Verbal and nonverbal cues to truth and deception during polygraph examinations. Journal of Police Science and Administration, I , 138-152.

Horvath, F., Jayne, B., & Buckley, J. (1994). Differentiation of truthful and de­ceptive criminal suspects in behavioral analysis interviews. Journal of Fo­rensic Sciences, 39, 793-807.

Hummert, M. L. (1999). A social cognitive perspective on age stereotypes. In T. M. Hess & F. Blanchard-Fields (Eds.), Social cognition and aging (pp. 175-196). San Diego, CA: Academic Press.

Inbau, F. E., Reid, J. E., & Buckley, J. P. (1986). Criminal interrogation and confessions, third edition. Baltimore, MD: Williams & Wilkins.

Inbau, F. E., Reid, J. E., Buckley, J. B, & Jayne, B. C. (2001). Criminal interro­gation and confessions, fourth edition. Gaithersburg, Maryland: Aspen Publishers.

Kassin, S. M., & Fong, C. T. (1999). 'I'm innocent!': Effects of training on judg­ments of truth and deception in the interrogation room. Law and Human Behavior, 23, 499-516.

Kaufmann, G., Drevland, G. C., Wessel, E., Overskeid, G., & Magnussen, S. (in press). The importance of being earnest: Displayed emotions and witness credibility. Applied Cognitive Psychology.

Kohnken, G. (1987). Training police officers to detect deceptive eyewitness statements. Does it work? Social Behaviour, 2, 1-17.

Kohnken, G. (1999, July). Statement Validity Assessment. per presented at the pre-conference program of applied courses Assessing credibility' orga­nized by the European Association of Psychology and Law, Dublin, Ireland.

Krauss, R. M., & Chiu, C. Y. (1998). Language and behavior. In D. T. Gilbert, S. T. Fiske, & G. Lindzey (Eds.), The handbook of social psychology (pp. 41-88). Boston, MA: McGraw-Hill.

Lance, M., & Mayo, C. (1976). Racial differences in gaze behavior during conversations: Two systematic observational studies. Journal of Personal­ity and Social Psychology, 33, 547-552.

Lane, J. D., & DePaulo, B. M. (1999). Completing Coyne's cycle: Dysphorics'

ability to detect deception. Journal of Research in Personality, 33, 311-329. Mann, S. (2001). Suspects, lies and videotape: An investigation into telling

and detecting lies in police/suspect interviews. Unpublished PhD thesis.

University of Portsmouth, Department of Psychology.

Mann, S., Vrij, A., & Bull, R. (2002). Suspects, lies and videotape: An analysis

of authentic high-stakes liars. Law and Human Behavior, 26, 365-376 Mann, S., Vrij, A., & Bull, R. (in press). Detecting true lies: Police officers' abil‑

ity to detect suspects' lies. Journal of Applied Psychology.

Meissner, C. A., & Kassin, S. M. (2002). 'He's guilty!': Investigator bias in judg‑

ments of truth and deception. Law and Human Behavior, 26, 469-480. Mehrabian, A. (1972). Nonverbal communication. Chicago: Aldine-Atherton. Mehrabian, A., & Ferris, S. R. (1967). Inference of attitudes from nonverbal

communication in two channels. Journal of Consulting and Clinical Psy‑

chology, 31, 248-252.

Mehrabian, A., & Wiener, M. (1967). Decoding of inconsistent communication. Journal of Personality and Social Psychology, 6, 109-114.

Millar, M. G., & Millar, K. U. (1997). The effects of cognitive capacity and suspi­cion on truth bias. Communication Research, 24, 556-570.

Millar, M. G., & Millar, K. U. (1998). The effects of suspicion on the recall of cues

used to make veracity judgments. Communication Reports, 11, 57-64.
Miller, G. R., & Stiff, J. B. (1993). Deceptive Communication. Newbury Park

CA: Sage.

Milne, R., & Bull, R. (1999). Investigative interviewing: Psychology and prac­tice. Chichester: John Wiley & Sons.

Moston, S. (1992). Truth or lies: Are police officers able to distinguish truthful from deceptive statements? Policing, November 8.

Moston, S., & Engelberg, T. (1993). Police questioning techniques in tape re-corded interviews with criminal suspects. Policing and Society, 3, 223-237.

Ofshe, R. (1989). Coerced confessions: The logic of seemingly irrational ac­tion. Cultic Studies Journal, 6, 1-15.

O'Sullivan, M., Ekman, P., & Friesen, W. V. (1988). The effect of comparisons

on detecting deceit. Journal of Nonverbal Behavior, 12, 203-216.

Porter, S., Woodworth, M., & Birt, A. R. (2000). Truth, lies, and videotape: An

investigation of the ability of federal parole officers to detect deception. Law

and Human Behavior, 24, 643-658.

Porter, S., & Yuille, J. C. (1995). Credibility assessment of criminal suspects through statement analysis. Psychology, Crime, & Law, 1, 319-331.

Porter, S., & Yuille, J. C. (1996). The language of deceit: An investigation of the verbal clues to deception in the interrogation context. Law and Human Be­havior, 20, 443-459.

Porter, S., Yuille, J. C., & Birt, A. R. (2001). The discrimination of deceptive, mistaken, and truthful witness testimony. In R. Roesch, R. R. Corrado, & R. Dempster (Eds.), Psychology in the courts: International advances in knowledge. London: Routledge.

Porter, S., Yuille, J. C., & Lehman, D. R. (1999). The nature of real, implanted and fabricated memories for emotional childhood events: Implications for the recovered memory debate. Law and Human Behavior, 23, 517-537.

Reid, J. E., & Arther, R. O. (1953). Behavior symptoms of lie-detector subjects.
Journal of Criminal Law, Criminology and Police Science, 44, 104-108.

Rio R. E. (1986). Assessment of basic social skills. Journal of Personality and Social Psychology, 51, 649-660.

Riggio, R. E. (1994). Epilogue: Where are we going, and how do we get there? Journal of Language and Social Psychology, 13, 514-518.

Riggio, R. E., Thcker, J., & Throckmorton, B. (1988). Social skills and decep‑
tion ability. Personality and Social Psychology Bulletin, 13, 568-577.

Riggio, R. E., Thcker, J., & Widaman, K. F. (1987). Verbal and nonverbal cues as mediators of deception ability. Journal of Nonverbal Behavior, 11, 126-145.

Rozelle, R. M., & Baxter, J. C. (1975). Impression formation and danger recognition in experienced police officers. The Journal of Social Psychology, 96, 53-63.

Ruby, C. L., & Brigham, J. C. (1996). A criminal schema: The role of chronicity, race, and socioeconomic status in law enforcement officials' perceptions of others. Journal of Applied Social Psychology, 26, 95-112.

Stangor, C., Lynch, L., Changining, D., & Glass, B. (1992). Categorization of in­dividuals on the basis of multiple social features. Journal of Personality and Social Psychology, 72, 207-218.

Steller, M. (1989). Recent developments in statement analysis. In J. C. Yuille (1989), Credibility Assessment (pp. 135-154). Deventer, the Netherlands: Kluwer.

Steller, M., & Kohnken, G. (1989). Criteria-Based Content Analysis. In D. C. Raskin (Ed.), Psychological methods in criminal investigation and evi­dence (pp. 217-245). New York: Springer-Verlag.

Street, R. L., & Hopper, R. (1982). A model of speech style evaluation. In E. B. Ryan & H. Giles (Eds.), Attitudes toward language variation: Social and applied contexts (pp. 175-188). London: Edward Arnold.

Stromwall, L. A., & Granhag, P. A. (2003). Hoe to detect deception? Arresting the beliefs of police officers, prosecutors and judges. Psychology, Crime, & Law, 9, 19-36.

Stromwall, L. A., Granhag, P. A., & Jonsson, A. C. (2003). Deception among pairs: 'Let's say we had lunch together and hope they will swallow it'. Psy­chology, Crime, & Law, 9, 109-124.

Taylor, R., & Vrij, A. (2000). The effects of varying stake and cognitive complex­ity on beliefs about the cues to deception. International Journal of Police Science and Management, 3, 111-124.

Tecce, J. J. (1992). Psychology, physiology and experimental. In McGraw-Hill yearbook of science and technology (pp. 375-377). New York: McGraw-Hill.

Vrij, A. (1993). Credibility judgments of detectives: The impact of nonverbal behaviour, social skills and physical characteristics on impression forma­tion. Journal of Social Psychology, 133, 601-611.

Vrij, A. (1995). Behavioral correlates of deception in a simulated police inter-view. Journal of Psychology, 129, 15-28.

Vrij, A. (1997). Wearing black clothes: The impact of offenders' and suspects' clothing on impression formation.Applied Cognitive Psychology, 11, 47-53. Vrij, A. (1998). To lie or not to lie. Psychologie, 17, 12, 22-25.

Vrij, A. (2000a). Detecting lies and deceit: The psychology of lying and its im‑

plications for professional practice. Chichester: John Wiley and Sons.
Vrij, A. (2000b). Telling and detecting lies as a function of raising the stakes. In

C. M. Breur, M. M. Kommer, J. F. Nijboer, & J. M. Reintjes (Eds.), New

trends in criminal investigation and evidence II (pp. 699-709).

Antwerpen, Belgium: Intersentia.

Vrij, A. (2002). Telling and detecting lies. In N. Brace & H. L. Westcott (Eds.),

Applying Psychology (pp. 179-241). Milton Keynes, Open University. Vrij, A. (2003). Criteria-Based Content Analysis: A qualitative review of the

first 37 studies. Manuscript submitted for publication.

Vrij, A., Akehurst, L., Soukara, R., & Bull, R. (in press). Detecting deceit via analyses of verbal and nonverbal behavior in adults and children. Human Communication Research.

Vrij, A., Dragt, A. W, & Koppelaar, L. (1992). Interviews with ethnic interview­ees: Nonverbal communication errors in impression formation. Journal of Community and Applied Social Psychology, 2, 199-209.

Vrij, A., Edward, K., & Bull, R. (2001b). Stereotypical verbal and nonverbal re­sponses while deceiving others. Personality and Social Psychology Bulle­tin, 27, 899-909.

Vrij, A., Edward, K., Roberts, K. P. & Bull, R. (2000). Detecting deceit via anal­ysis of verbal and nonverbal behavior. Journal of Nonverbal Behavior, 24, 239-263.

Vrij, A., & Fisher, A. (1997). The role of displays of emotions and ethnicity in judgements of rape victims. International Review of Victimology, 4, 255-265.

Vrij, A., Foppes, J. H., Volger, D. M., & Winkel, F. W. (1992). Moeilijk to bepalen wie de waarheid spreekt: Non-verbaal gedrag belangrijkste indicator. Algemeen Politie Blad, 141, 13-15.

Vrij, A., & Graham, S. (1997). Individual differences between liars and the ability to detect lies. Expert evidence: The international digest of human behaviour science and law, 5, 144-148.

Vrij, A., Harden, F., Terry, J., Edward, K., & Bull, R. (2001d). The influence of personal characteristics, stakes and lie complexity on the accuracy and confidence to detect deceit. In R. Roesch, R. R. Corrado, & R. J. Dempster (Eds.), Psychology in the courts: International advances in knowledge (pp. 289-304). London: Routledge.

Vrij, A., & Heaven, S. (1999). Vocal and verbal indicators of deception as a

function of lie complexity. Psychology, Crime, & Law, 4, 401-413.

Vrij, A., & Holland, M. (1998). Individual differences in persistence in lying

and experiences while deceiving. Communication Research Reports, 15,

299-308.

Vrij, A., & Lochun, S. Neuro-linguistic programming and the police: Worth-while or not? Journal of Police and Criminal Psychology, 12, 25-31.

Vrij, A., & Mann, S. (2001a). Telling and detecting lies in a high-stake situa­tion: The case of a convicted murderer. Applied Cognitive Psychology, 15, 187-203.

Vrij, A., & Mann, S. (2001b). Who killed my relative? Police officers' ability to
detect real-life high-stake lies. Psychology, Crime, & Law, 7, 119-132.

Vrij, A., & Mann, S. (in press). Detecting deception: The benefit of looking at a combination of behavioral, auditory and speech content related cues in a systematic manner. Group Discussion and Negotiation.

Vrij, A., & Semin, G. R. (1996). Lie experts' beliefs about nonverbal indicators of deception. Journal of Nonverbal Behavior, 20, 65-80.

Vrij, A., Semin, G. R., & Bull, R. (1996). Insight in behaviour displayed during deception. Human Communication Research, 22, 544-562.

Vrij, A., & Taylor, R. (2003). Police officers' and students' beliefs about telling and detecting little and serious lies. International Journal of Police Science and Management, 5, 1-9.

Vrij, A., & Van Wijngaarden, J. J. (1994). Will truth come out? Two studies about the detection of false statements expressed by children. Expert evi­dence: The international digest of human behaviour, science and law, 3, 78-84.

Vrij, A., & Winkel, F. W (1991). Cultural patterns in Dutch and Surinam non-verbal behaviour: An analysis of simulated police/citizen encounters. Jour­nal of Nonverbal Behavior, 15, 169-184.

Vrij, A., & Winkel, F. W. (1992a). Cross-cultural police-citizen interactions: The influence of race, beliefs and nonverbal communication on impression formation. Journal of Applied Social Psychology, 22, 1546-1559.

Vrij, A., & Winkel, F. W. (1992b). Social skills, distorted perception and being suspect: Studies in impression formation and the ability to deceive. Jour­nal of Police and Criminal Psychology, 8, 2-6.

Vrij, A., & Winkel, F. W (1994). Perceptual distortions in cross-cultural inter-rogations: The impact of skin color, accent, speech style and spoken fluency on impression formation. Journal of Cross-Cultural Psychology, 25, 284-296.

Vrij, A., Winkel, F. W, & Koppelaar, L. (1991). Interactive tussen politie­functionarissen en allochtone burgers: Twee studies naar de frequentie en het effect van aan-en wegkijken op de impressieformatie. Nederlands Ti~dschrift voor de Psychologie, 46, 8-20.

Walkley, J. (1985). Reading the suspect. Police Review, 15 February.

Wallbott, H. G., & Scherer, K. R. (1991). Stress specifics: Differential effects of coping style, gender, and type of stressor on automatic arousal, facial ex­pression, and subjective feeling. Journal of Personality and Social Psy­chology, 61, 147-156.

Waltman, J. L. (1983). Nonverbal communication in interrogation: Some ap­plications. Journal of Police and Science Administration, 11, 166-169.

Winkel, F. W, & Koppelaar, L. (1991). Rape victims' style of self-presentation and secondary victimization by the environment. Journal of Interpersonal Violence, 6, 29-40.

Yeschke, C. L. (1997). The art of investigative interviewing: A human ap­proach to testimonial evidence.

Zaparniuk, J., Yuille, J. C., & Taylor, S. (1995). Assessing the credibility of true and false statements. International Journal of Law and Psychiatry, 18, 343-352.

Zuckerman, M., DeFrank, R. S., Hall, J. A., Larrance, D. T., & Rosenthal, R. (1979). Facial and vocal cues of deception and honesty. Journal of Experi­mental Social Pychology, 5, 378-396.

Zuckerman, M., DePaulo, B. M., & Rosenthal, R. (1981). Verbal and nonverbal communication of deception. In L. Berkowitz (Ed.), Advances in experi­mental social psychology, volume 14 (1-57). New York: Academic Press.

Zulawski, D. E., & Wicklander, D. E. (1993). Practical aspects of interview and interrogation.

ENDNOTES

1. Whether the lie detector is an actual interviewer or a passive observer might be another variable that influences whether people pay more attention to speech content or to nonverbal communication, although the findings are inconsistent. Buller, Strzyzewski and Hunsaker (1991) and Feeley and deTurck (1997) found that active interviewers pay most attention to nonver­bal communication, whereas observers mostly look at speech content. Granhag and Stromwall (2001b), however, found the opposite effect. In addition, being suspicious may also have an effect on paying attention to speech content or nonverbal communication while attempting to detect de­ceit (Millar & Millar, 1997, 1998). Again, the findings are not clear.

2. Somewhat different from speech content is speech style. Speech style ad-dresses the issue whether certain words or phrases indicate deception. Re-searchers have examined several speech style aspects, including verbal immediacy (the present tense is more immediate than the past tense, and phrases such as 'Here's Johnny' is more immediate than 'There's Johnny'), generalizing terms (everyone, no one, all, every, etc.), self-refer­ences (I, me, mine, myself), group references (we, us, ours), other refer­ences (he, she, they, them), tentative constructions (may, might, could, etc.), ritualized speech (you know, well, really), and negative statements (includ ing complaints). DePaulo, Lindsay, Malone, Muhlenebruck, Chariton, and Cooper (2003) reviewed studies examining these aspects and found that li ars speak in less immediate terms and include more negative statements into their accounts than truth tellers.

3. Zuckerman, DePaulo, & Rosenthal (1981), who introduced these three fac­tors, also included a fourth factor in their theoretical model, labelled 'arousal'. We left this factor out as, in our view, it shows an overlap with the emotion factor. Zuckerman et al. (1981, p. 9) themselves already suggested this by finishing their arousal factor paragraph with the following state­ment: 'It is possible, however, that the general autonomic responsivity to deception reflects specific emotions. If so, cues to deception may be ac-counted for by the particular affects that are involved rather than by general arousal.'

4. The three processes are hypothetical and are typically introduced post hoc to explain nonverbal and verbal differences between liars and truth tellers. Apart from this 'three factor model', other theoretical models for explaining nonverbal cues to deception are given as well in the deception literature (Buller & Burgoon, 1996; DePaulo et al., 2003; Ekman, 1985; Ekman & Friesen, 1969). See DePaulo et al. (2003) for a description of each of these theoretical models. By discussing only the 'three factor model' we do not suggest that this theoretical model is superior to the other theoretical mod-

els. However, there is evidence that liars actually experience the three pro­cesses described in the three factor model when they lie, whereas similar studies have not been carried out regarding the other theoretical models. In one of our own experiments (Vrij, Semin, & Bull, 1996), participants were asked either to lie or to tell the truth. Afterwards they were asked to what ex-tent they had experienced the three processes. Results showed that liars ex­perienced all three processes significantly more than truth tellers. In another experiment (Vrij, Edward, & Bull, 2001) we found individual differ­ences in experiencing these processes. For example. a negative correlation was found between being good at acting and having to think hard while ly­ing. Although these studies were correlational studies, the relationship be­tween the three processes and lying is more likely to be causal: They are the consequence of being engaged in lying.

5. Commercial companies have exploited this idea and brought several voice analyzers on the market which, they say, can be used to detect deceit. How-ever, these analyzers are not as accurate as many companies claim them to be. See Vrij (2000a) for problems lie detectors face when using equipment which measures physiological responses.

6. Ekman's work (1985) has revealed that observing emotional micro-expres­sions in the face might reveal valuable information about deception. Strongly felt emotions almost automatically activate muscle actions in the face. Anger. for example, results in a narrowing of the lips and lowering of the eyebrows, and eyebrows which are raised and pulled together and a raised upper eyelid and tensed lower eyelid typically denote fear. If a person denies an emotional state which is actually being felt, this person will have to suppress these facial expressions. Thus, if a scared person claims not to be afraid, that person has to suppress the facial micro-expressions which typically indicate fear. This is difficult, especially because these emotions can arise unexpectedly. For instance, people do not usually deliberately choose to become frightened, this happens automatically as a result of a particular event that took place, or as the result of a particular thought. The moment fright occurs, a fearful facial expression may be shown which may give the lie away. People are usually able to suppress these expressions within 0.5 of a second after they begin to appear (Ekman, 1985).

The opposite can occur as well. Someone can pretend to experience a par­ticular emotion, whereas in fact this emotion is not felt. Someone can pre-tend to be angry, whereas in reality the person is not angry at all. In order to be convincing, the liar should produce an angry facial expression, that is, the liar should try to narrow the lips. However, this muscle action is very dif­ficult for most people to make voluntarily (Ekman, 1985).

It is also difficult to fake an emotion other than the one which is actually felt.

For example, a potential hijacker may become scared during a conversa­tion with security personnel when he realizes that they might find out what his plans are, but can decide to mask this emotional state by pretending to be angry with the security personnel because they are checking on him so thoroughly and apparently do not trust him. In order to be convincing, he therefore has to suppress his fearful facial expression and replace it with an angry facial expression. This is difficult, because he has to lower his eye-brows (sign of anger) whereas_. his eyebrows tend to raise (sign of fear) (Ekman, 1985). Ekman's observations could well be of value. For example, in one of our studies (Vrij & Mann, 2001a) we included a person who held a

televised press conference to ask for information about his missing girl-friend. Later it turned out that he himself had killed his girlfriend. A de-tailed analysis of the videoclip revealed that he showed a micro-expression of a (suppressed) smile during that press conference. His smile was in the given context interesting. Why did the man smile? And why did he attempt to suppress that smile? Although his smiling at a press conference cannot be interpreted as a definite indication of deceit, at least, it made the man suspicious. Unfortunately, no empirical test of the frequency of occurrence of emotional micro-expressions during lying and telling the truth yet ap­pears to have been published in peer reviewed journals, which is also the reason why these micro-expressions did not emerge as cues to deception in recent literature reviews.

7. Sometimes professional lie catchers tell us that they believe that eye move­ments are associated with deception. They then typically refer to the neurolinguistic programming (NLP) model. However, not a single scientific study has demonstrated that eye movements are related to deception in the way described in the NLP model (Vrij & Lochun, 1997). NLP teachers who claim the opposite therefore are engaged in deceiving their pupils.

8. Although the total and lie accuracy rates were significantly higher than the total and lie accuracy scores obtained by laypersons (mostly college stu­dents) in previous research, it cannot be concluded that police officers are actually better lie detectors than laypersons. Laypersons were not included in Mann et al.'s (2003) study, and perhaps they would have achieved similar accuracy rates as police officers if they had participated.