New Voices: Detecting lies about intentions

Matti Saari was a 22-year-old from Finland. In 2008 he uploaded several videos to the internet showing himself firing handguns and saying things like 'you will die next' to the camera. On 22 September he was visited by the police, who had been informed about the videos and had issued his licence for owning handguns. Saari convinced them that there was no reason to revoke his gun licence, although they did give Saari an 'unofficial warning' regarding his behaviour. The next day, Saari shot and killed 10 people and committed suicide at a vocational school.

In this case, there was an opportunity to stop the crime from happening. Saari's gun licence could have been revoked and his weapons confiscated. Had this happened, Saari may of course have committed some other crime, but it is unlikely he could have committed such a deadly attack. Had the police officers been able to realise what Saari's intentions were, and the legislation existed to act on this, the tragedy could have been prevented. The idea that such crimes could be prevented has encouraged psychologists with experience in lie-detection research to start investigating methods of identifying concealed or false intentions (Granhag, 2010). Since 2009 I have been one of those researchers: first as a PhD student at the University of Portsmouth and now as a Faculty Fellow at Lancaster University.

Mental time travel

Research into detecting true and false intentions has been drawing on two older fields of psychology: episodic future thought (Szpunar, 2010) and lie detection (Vrij, 2008). Episodic future thought has recently become a growing area of research: it is defined as mentally simulating possible personal future events (Szpunar, 2010). D'Argembeau and Van der Linden (2004, 2006) found that episodic future thoughts tend to be less vivid and detailed, but more positive in emotion than memories of past events. It seems we really do view the future through rose-tinted glasses.

Schacter et al. (2007) have presented evidence suggesting that our ability to imagine the future is based on our ability to remember the past.  Schacter and colleagues (2007) suggest that thinking about the future involves recombining details from past events that are stored in long-term memory. The discovery that remembering the past and imagining the future appear to be conducted using very similar methods has proved a boon for the researchers who are trying to find methods to detect false intentions. Most of the lie-detection methods in use today were developed to detect lies about past events. If mental images of past and future events are created in the same way, it is likely that people will lie about past and future events in a similar way. This in turn suggests that methods developed to detect lies about past events can be adapted to detect lies about intentions.

Detecting lies

People are generally bad at detecting lies: without specialised lie-detection methods their accuracy is about as good as flipping a coin (Vrij, 2008).  There are several methods in use to detect lies about past events at above chance rates, including physiological methods (e.g. polygraphs), and analysing verbal and non-verbal behaviour during the suspected lie (Vrij, 2008). Polygraphs are probably the most well-known 'lie detector', and they have been shown to detect lies at above chance rates. But polygraphs do have disadvantages: they are vulnerable to false positives, labelling truth tellers as liars. Measuring signs of stress (breathing rate, hart rate, blood pressures, skin conductance), polygraphs assume that liars are more stressed than truth tellers. This might be the case for many liars and truth tellers, but truth tellers can also be stressed (e.g. because they are scared they'll fail the test).

People tend to believe that they can spot liars by looking at non-verbal behaviour; many people think that liars fidget more than truth tellers and tend to avoid eye contact. In fact, the research shows that liars fidget less with their hands and illustrate their story with fewer gestures than truth tellers (Vrij, 2008). The research into eye contact shows very mixed results. So, the reason that people aren't very good at lie detection seems to be that they don't focus on the right cues. Verbal cues to deception tend to be more reliable. For example, liars tend to use more negative statements than truth tellers (e.g. 'I am not a crook'), and their stories are rated as less plausible and less immediate (less direct and relevant; more evasive) (Vrij, 2008). But even these cues are by no means perfect.

False intentions

Vrij et al. (2011) interviewed participants both before and after an event they were asked to lie or tell the truth about. When talking about the event after it happened, they found that truth tellers' stories were more detailed and more plausible than liars' stories. When talking about the event before it happened, truth tellers' stories were more plausible but not more detailed than liars'. They also found that participants without lie-detection experience could identify the lies about intentions at above chance rates, but not the lies about past events. This is a promising, but somewhat odd result. The stories about past events contained more cues to deception, but this did not help the participants, another sign that people without lie-detection experience don't look out for the best cues to deception.  

My colleagues and I (Warmelink et al., 2012) replicated the finding that liars are no less detailed than truth tellers when simply describing their intention. Only when they are asked unexpected questions about the context of the intentions (rather than the core of the intention) do truth tellers give more details than liars. When asked expected questions, liars can even be more detailed than truth tellers, presumably because the liars have mentally prepared detailed answers to these questions before they were asked. Granhag and Knieps (2011) found that liars' reports of their images of the future are less detailed than those of truth tellers. These studies suggest that it is important to ask the right questions if you want to detect lies. If you ask expected questions, the liar could have prepared their answers, making it impossible to detect the differences in detail that exist between in truth tellers' and liars' intentions.  

Another way to detect lies about intentions is to use reaction times (Ask et al., 2013). This technique relies on findings that future images have a more positive feel (D'Argembeau & Van der Linden, 2004). Participants were instructed to complete a reaction time task in which they had to classify adjectives as positive or negative as fast as possible. Before each adjective they were primed with a word related to an intention they had received instructions about. Half of the participants thought they were going to complete the intention and had been instructed to tell the truth about the task. Half of the participants thought that they were not going to complete the intention, but that they would have to lie and pretend they were going to complete the task. It was found that those participants who thought they were going to perform the intention were slower to classify a negative adjective than a positive one. Presumably this is because the negative adjective was incongruent with the intention-related prime, which had a positive feel for them. Participants who were lying about performing the task responded equally fast to positive and negative adjectives: for them the primes have no emotional connection.     

Physiological measures have also been used to detect lies about intentions (Warmelink et al., 2011). My colleagues and I used thermal-imaging cameras to detect lies. We found that liars' facial temperature increases more during the interview than that of truth tellers. Presumably liars are either more nervous than truth tellers or they are finding the interview more strenuous, making them warmer. However, liars are not warmer than truth tellers before the questions are asked, which means that you can't use the thermal imaging cameras to screen large groups of people for those with possible criminal intentions.  

The future 

Research in detecting lies about intention has a very clear application: to help prevent crime. However, this goal is very distant at the moment. We have found some cues that help pick out some lies, sometimes. They can give us a clue that something might be wrong with a statement of intent, but do not give us enough confidence in our judgement to act based on them. Even if we could detect lies about intentions in a reliable way, we would have to wonder how this would be dealt with legally. Can people be brought to court, arrested or charged for things they intend to do? We do not live in a Minority Report world, where people go to jail for murders they are yet to commit. In this area, we have many theoretical, applied and philosophical challenges ahead of us.

• Lara Warmelink is a Faculty Fellow at Lancaster University [email protected]

References 

Ask, K., Granhag, P.A., Juhlin, F. & Vrij, A. (2013). Intending or pretending? Automatic evaluations of goal cues discriminate true and false intentions. Applied Cognitive Psychology, 27(2), 173–177.
D'Argembeau, A. & Van der Linden, M. (2004). Phenomenal characteristics associated with projecting oneself back into the past and forward into the future: Influence of valence and temporal distance. Consciousness and Cognition, 13, 844–858.  
D'Argembeau, A. & Van der Linden, M. (2006). Individual differences in the phenomenology of mental time travel. Consciousness and Cognition, 15, 342–350.  
Granhag, P.A. (2010). On the psycho-legal study of true and false intentions: Dangerous waters and some stepping stones. Open Criminology Journal, 3, 37–43.
Granhag, P.A. & Knieps, M. (2011). Episodic future thought: Illuminating the trademarks of true and false intent. Applied Cognitive Psychology, 25, 274–280.
Schacter, D.L., Addis, D.R. & Buckner, R.L. (2007). Remembering the past to imagine the future: The prospective brain. Nature Reviews: Neuroscience, 8, 657–661.
Szpunar, K.K. (2010). Episodic future thought: An emerging concept. Perspectives on Psychological Science, 5(2), 142–162.
Vrij, A. (2008). Detecting lies and deceit: Pitfalls and opportunities (2nd edn). Chichester: Wiley.
Vrij, A., Leal, S., Mann, S. & Granhag, P. A. (2011). A comparison between lying about intentions and past activities: Verbal cues and detection accuracy. Applied Cognitive Psychology, 25(2), 212–218
Warmelink, L., Vrij, A., Mann, S. et al. (2011). Thermal imaging as a lie detection tool at airports. Law and Human Behaviour, 35, 1, 40–48.Warmelink, L., Vrij, A., Mann, S. et al. (2012). The effect of question expectedness and experience on lying about intentions. Acta Psychologica, 141, 2.