Trying to find a tell-tale sign of deceit (a “Pinocchio’s nose”) in human nonverbal behavior has been the subject of much effort, and many suggestions have been put forward. In lay people’s thinking and in police interrogation manuals alike, one can find numerous ideas about detecting deceit from nonverbal behaviors such as eye contact or gestures. The scientific research shows, however, that overall only a few nonverbal behaviors are associated with deception. Under certain conditions, such as time to prepare the lie, special motivation to convincingly tell a lie, and when the lie is about concealing a transgression, there seem to be some nonverbal behaviors that may distinguish liars from truth tellers.
Research on beliefs about deception shows that presumed experts (e.g., police officers) and lay people (e.g., college students) have very similar beliefs. They mostly indicate nonverbal signs of deception, especially a decrease in eye contact, when lying. Furthermore, presumed experts and lay people alike believe that an association exists between deception and an increase in body movements.
Scientific Study of Nonverbal Behaviors
To find out about potential nonverbal correlates of deception, psychologists and other researchers conduct experiments. They instruct some people to lie and/or tell the truth (the lies are most often “constructed” for the sake of the experiment) and videotape the telling of truths and lies in interviews or mock interrogations. (If the focus is on the speech-related variables, audiotapes are of course sufficient.) Then, these videotapes are closely analyzed, and the frequency and/or duration of a list of nonverbal behaviors are scored. The scored behaviors are then summarized for truths and lies separately, and if statistical comparisons show significant differences, researchers conclude that there are systematic nonverbal signs of deceit and truthfulness. A great number of such studies have been published. In this research paper, findings from several meta-analyses and research overviews are summarized.
Included in the concept nonverbal behavior are body movements (e.g., gestures and leg movement), facial indicators (e.g., eye contact, smiling), and speech behaviors (sometimes called paraverbal behaviors; e.g., response latency and pitch of voice).
Why would the nonverbal behavior of a liar give him or her away? Scientists usually suggest three different processes (or approaches) that might answer that question. According to the emotional approach (sometimes called the affective approach), the three most common types of emotion associated with deception are guilt, fear, and excitement. A liar might feel guilty because he or she is lying, might be afraid of getting caught, or might be excited about the possibility of fooling someone (“duping delight”).
According to the cognitive complexity approach (sometimes called cognitive load or working memory model), the lie should be possible to detect from the liars’ nonverbal behavior because it is more difficult to lie than to tell the truth. The liars have to come up with believable answers, avoid contradicting themselves, and tell a lie consistent with what the interviewer knows or might find out. Additionally, they have to remember what they have said, in order to declare the same things again if asked to repeat their statement.
The attempted control approach emphasizes that liars may be concerned that their lies will be detected by, for example, nonverbal behaviors and will therefore try to suppress such behaviors. In other words, they will try to make a convincing impression by, for example, suppressing their nervousness and masking evidence of thinking hard. However, when controlling their body language, liars may overcontrol their behavior, therefore exhibiting body language that will appear planned, rehearsed, and lacking in spontaneity. For example, liars may believe that bodily movements will give their lies away and will consequently avoid any movements not strictly essential, resulting in rigidity.
All three processes may occur simultaneously. That is, liars could—at the same time—be nervous, have to think hard, and try to control themselves. Which of these processes is most prevalent depends on the situation. In high-stake lies, nervous responses are more likely to occur. In complicated lies, indicators of increased cognitive load are more likely to occur. Attempts to control behavior, voice, and speech may especially occur in motivated liars.
Before turning to the outcomes of reviews about nonverbal behavior, it should be emphasized that the approaches only suggest that the presence of signs of emotions, content complexity, and impression management may be indicative of deceit. None of these approaches claim that the occurrence of these signs necessarily indicates deception. Truth tellers might experience exactly the same processes. For example, innocent (truthful) suspects might very well be anxious if they worry about not being believed in a police interview. Because of that fear, they may show the same nervous reactions as liars who are afraid of being caught. The lie catcher is then put in a difficult position: Should the nonverbal behaviors be interpreted as signs of guilt or of innocence? The behavior does not provide the answer. The false accusation of a truth teller on the basis of the emotional reactions displayed has been labeled the Othello error, after Shakespeare’s play.
Nonverbal Behavior and Deception in General
The most notable result of research to date is that nonverbal behaviors generally do not correlate strongly with either deception or truthfulness; very few reliable nonverbal cues to deception have been found.
There is evidence that liars tend to speak in a higher-pitched voice, which might be the result of experienced arousal. However, differences in pitch between liars and truth tellers are usually small and detectable only with technical equipment. Furthermore, sometimes liars’ voices sound tenser than truth tellers’ voices, another result of arousal. Speech errors (e.g., word and/or sentence repetition, sentence incompletion, slips of the tongue) occur more often during deception, and response latency is longer before giving deceptive answers. There is also evidence of message duration being shorter for liars, who also tend to make fewer illustrators (hand and arm movements modifying what is said verbally). The decrease in movements might be the result of lie complexity or overcontrol of behavior. Moreover, compared with truth tellers, liars tend to sound vocally less expressive, more passive, and more uncertain. This might all be the result of overcontrol of behavior. Liars also sound less involved and come across as being less cooperative and tending to make more negative statements. This might be caused by a negative emotion felt by the liar.
Perhaps the most remarkable outcome of the literature reviews is the finding that several signs of nervousness, such as gaze aversion (avoidance of eye contact) and fidgeting, are generally unrelated to deception. One reason why nervous behaviors do not seem to be systematically related to deception is that truth tellers could be nervous as well. Another reason could be that in most deception studies, people are requested to lie or tell the truth for the sake of the experiment, and in such studies, liars might not be aroused enough to show cues of nervousness.
Summarizing the literature, there seem to be a greater number of reliable verbal cues to deception than nonverbal cues. This contradicts most police interrogation manuals, which typically emphasize nonverbal cues to deception, and contradicts presumed experts’ and lay people’s beliefs about what gives a liar away as well.
The results presented so far are at the most general level—across all available studies without taking into account differences in the experimental designs. There are, however, a few presumably moderating factors that have been studied often enough to allow for interesting conclusions; three of these are discussed below.
An important factor, and most relevant to the forensic context, is the distinction between lies that are about transgressions and those that are not. Lies about transgressions are told to hide and/or deny acts such as cheating, stealing, and committing other crimes, small and large. In other studies, participants, for example, pretended to experience another emotion that they did not in fact experience or lied about their opinions. The question is whether differences between liars’ and truth tellers’ nonverbal behavior emerge when they are interviewed about transgressions they have or have not committed.
The literature describing the lies that were not about transgressions shows only one behavior that separates the liars from the truth-tellers, and that is fidgeting. When participants were talking about their likes or dislikes, their opinions and emotions, or anything else that did not involve a bad behavior, they fidgeted more when lying than when telling the truth. The cues to lies about transgressions are more important in legal contexts. People lying about transgressions look more nervous than do truth tellers; they also blink more and have a faster speech rate. Additionally, they are more inhibited than truth tellers in the sense that they move their feet and legs less often.
In many studies, the liars did not have any special motivation to tell a convincing lie. Many simply participated as part of a study, with no special rewards for succeeding or punishments for failing. It is of importance to separate those studies in which participants had some special motivation to do well and those in which they did not. The question is this: If people are motivated to get away with their lies, will that show up in the form of fewer cues to deception because they are trying harder to tell a good lie or will their lies become more obvious as the stakes are raised?
Research has shown that when participants had no special incentives, there were no obvious nonverbal cues to deception. When people do not have that much invested in their lies, others will have a very hard time knowing when they are lying. However, when liars do care about getting away with their lies, then several behaviors may betray them. It is only when partici-pants are motivated to do well that they speak in a higher pitch when lying than when telling the truth. Although liars also seem tenser than truth tellers regardless of motivation, the difference is pronounced for those who are highly motivated to get away with their lies. In the previous section, in which results were summarized for all studies, there were no differences whatsoever in how often liars looked at the other person and how often truth tellers did. But when participants are motivated to do well, then one stereotype about liars becomes a reality: They make less eye contact than truth tellers do. There was also some evidence, under high motivational conditions, that liars made fewer foot and leg movements than truth tellers.
Sometimes suspects know beforehand that they are going to be interviewed, which gives them a chance to prepare their answers. Presumably, liars should manage to appear more like truth tellers when they can plan their answers in advance than when they cannot. The available research indicates that when having time to plan, liars have shorter response latency than truth tellers. When there is no time to prepare, the opposite pattern is found. There is also some evidence that liars show shorter message duration than truth tellers when they have time to prepare their responses.
Limitations and Conclusions
Although researchers have in some studies tried to raise the motivation of and the stakes for the lying participants, the question still remains how the results from laboratory-based studies reflect what may happen in real-life high-stakes situations such as police interviews.
In a few studies, the behavior of real-life suspects, interviewed about serious crimes such as murder, rape, and arson and facing long prison sentences if found guilty, has been examined. Results revealed that these suspects did not show the nervous behaviors typically believed to be associated with lying, such as gaze aversion and fidgeting. In fact, they exhibited an increase in pauses; a decrease in eye blinks; and (for male suspects) a decrease in finger, hand, and arm movements. This is more in line with the content complexity and attempted control approaches than with the emotional approach.
In summary, under certain conditions, there seem to be a few differences between truth tellers and liars in their nonverbal behavior. However, it is of great importance to realize that these differences, albeit significant in meta-analyses, are not large. Since the observed effect sizes are small, the practical value may be quite low. None of the behaviors discussed here can be used as a fail-safe decision rule. The available research thus indicates that there is no nonverbal indicator of deception that always works—there is no “Pinocchio’s nose.”
- DePaulo, B. M., Lindsay, J. J., Malone, B. E., Muhlenbruck, L., Charlton, K., & Cooper, H. (2003). Cues to deception. Psychological Bulletin, 129, 74—118.
- Granhag, P. A., & Vrij, A. (2005). Deception detection. In N. Brewer & K. D. Williams (Eds.),Psychology and law. An empirical perspective (pp. 43-92). New York: Guilford Press.
- Sporer, S. L., & Schwandt, B. (2006). Paraverbal indicators of deception: A meta-analytic synthesis. Applied Cognitive Psychology, 20, 421-446.
- Vrij, A. (2000). Detecting lies and deceit: The psychology of lying and the implications for professional practice. Chichester, UK: Wiley.
- Zuckerman, M., DePaulo, B. M., & Rosenthal, R. (1981). Verbal and nonverbal communication of deception. In L. Berkowitz (Ed.), Advances in experimental social psychology (Vol. 14, pp. 1-60). New York: Academic Press.
- Detection of Deception: Reality Monitoring
- Detection of Deception in Adults
- Statement Validity Assessment (SVA)