Can you really sort out the liars from the truth tellers?
It’s broadly accepted that people are very poor at distinguishing between good liars and actual truth tellers. But researchers keep trying to figure out how we can do better. And for this, we are grateful.
The latest entry in this research looks at the effect of increasing cognitive load and truth telling. Cognitive loading while talking in this case involves having someone do a task that involves thought while at the same time asking them to tell you about a series of events. The task increases their cognitive load and thus results in liars slowing down (because constructing a lie is also a demand on cognition) while truth tellers do not slow as much since they are not using energy to both lie and complete the task. It’s a common research strategy and one we have written about before.
This time though the setup for the experiment is intriguing. Participants in the research were divided into groups who would receive instructions to lie and instructions to tell the truth. (This division was done randomly and not through some strong intuition or advanced perceptive power on the part of the researchers.) Those assigned to be ‘truth tellers’ sat in a room and completed five separate tasks. Those assigned to be liars observed the truth tellers through an observation window and were then asked to craft a story (aka a lie) about how they themselves completed the same tasks. Then all of the subjects from both groups completed an interview on their experience doing those tasks. The liars were given a directive to be as convincing as possible during the interview portion of the task. They were to attempt to convince the interviewer they had actually completed the task rather than merely observing it.
During the interview, participants were asked to sort objects by touch alone and only while they were responding to the interviewers questions. While the interviewer was speaking, the participant was to sit quietly, listen, and not sort. They were told there was a bucket with wooden objects in it which were either shaped like hearts, round balls or other things (i.e., blocks, cylinders and pie-shaped wedges). There were also three trays on a table. They were to place hearts on one tray, round balls on another, and any other object that was not either a heart or a ball on a third tray. Once the interview began, a curtain blocked their view of the bucket with the objects and the trays themselves–thus insuring they were unable to see what they were doing and the sorting was done by touch alone.
The sorting task was designed to increase cognitive load. That is, to preoccupy the research participant and thus result in a slower response time to interviewer questions. The researchers believed that those assigned to the lying condition would take longer to respond. Part of the interview process was also designed to increase cognitive load by asking the research participants to describe what they had done in the task room and then to describe what they had done in reverse chronological order, and then to identify the location of the objects they worked with during the tasks in relation to two other objects int the room, and then to describe the objects in the room as though they were sitting in a chair at the table in the center of the room. (I don’t know about you, but that set of questions makes my head hurt even without the sorting task they are being asked to do at the same time.)
The researchers comment that liars tend to prepare for their lies and they would likely expect to be asked to recount what they had done. However, it would be unexpected to be asked to tell the story backwards and likely they would also not anticipate the need to respond to where objects in the room were as compared to two other objects in the room. The researchers were asking both expected and unexpected questions to see if it would throw the liars off their game. So. Did they find that liars took longer to respond?
Of course they did. Although truth tellers took a while to respond as well since they were rearranging what they had done in their heads while verbalizing it to the interviewer, considering where objects were in the room in relation to each other and sorting invisible shapes into three groups all at the same time. It would take anyone a while to respond in that sort of cognitively demanding environment. But liars took even longer. And, naturally, it wasn’t just how long it took the participants to speak or the details present in what they said when they did speak that the researchers were measuring! It was also how accurately and how quickly they were able to sort the unseen objects behind the curtain as they listened to the interviewer and responded verbally.
The researchers found that when liars had anticipated the questions they would be asked, there was no difference in the level of detail in the responses from truth tellers and liars. In the case of the unanticipated question, however, liars were not as detail-oriented in their responses as were the truth tellers. Additionally, liars completed fewer accurate sorts (in the secondary task of sorting unseen objects into three separate trays) than did truth tellers. Liars appeared more focused on responding to the interviewer questions persuasively than on sorting the disparate yet unseen objects into like groups.
The researchers say their results are preliminary but they offer a suggestion for interviewers attempting to sort liars from truth tellers.
Pair expected questions with follow-up questions that are unexpected.
For example, you would ask the interviewee to “tell me everything that you can remember about the event”. This question would be expected to elicit the same level of detail from truth tellers and liars. Second, you ask an unexpected question which will be much more difficult for the liar to answer. The unexpected question could be asking for the story in reverse chronological order.
If there is a large decline in detail in the answer–you could see that as indicative of deception.
If there is not a large decline in detail, you could see that as indicative of a truthful initial response.
Obviously, this is a pretty complex study and it couldn’t be used in any sort of sworn testimony. In fact, as a strategy it isn’t sensible for internal investigations in a company or in any normal setting we can imagine (you have to run comparable subjects and compare their pattern of reactions to the suspect group, and that doesn’t really work out on the job.) What the researchers end up saying is they believe they have discovered a strategy to begin to more accurately sort liars and truth tellers, which is true if everyone works in a productivity lab. For now, perhaps it is best for you to know two things:
Researchers will continue their efforts to identify useful ways to differentiate between truth tellers and liars.
We will continue to read and blog about their efforts.
If they find a way to distinguish liars and truth tellers that you can use in court–you will read about it here! In great detail. And that’s the truth.
Lancaster, G., Vrij, A., Hope, L., & Waller, B. (2013). Sorting the Liars from the Truth Tellers: The Benefits of Asking Unanticipated Questions on Lie Detection. Applied Cognitive Psychology, 27 (1), 107-114 DOI: 10.1002/acp.2879