Are some people better liars, or are some lies more convincing?

While most people are useless at spotting lies, research has suggested that some of us make for more convincing liars than others. However, studies that have reached this conclusion have failed to consider that perhaps some lies are just easier to see through, argue Sarah Volz at the University and Kassel and colleagues in Perspectives on Psychological Science. Indeed, in their new study, in which they disentangle the effects of lie and liar, they find only very limited support for the idea that some people are better liars.

Volz and her colleagues re-analysed data collected for an earlier study. For this earlier research, the team had created videos of 80 different students and staff members at Miami University making four statements each. In the first, they talked positively about a person that they liked (a positive truth). In the second, they talked negatively about a person that they liked (negative lie). In the third, they talked negatively about someone they disliked (negative truth). And in the fourth, they talked positively about a person they disliked (a positive lie).

In the new study, the team divided these videos into sets of 16 statements to show participants. Every set contained statements by four White males, four White females, four Black males, and four Black females, and each demographic category included all four types of statement (positive truth, negative lie, and so on).

A total of almost 3,000 participants each viewed one of these sets of 16 statements. For every message, they indicated not just whether it was true or a lie, but also how confident they were in that judgement.

The re-analysis of the data showed that overall, the participants were right only about 51% of the time (i.e. they essentially did no better than chance). Participants showed a bias towards thinking a message was truthful: when a statement was true, they got this right about 71% of the time, but they also incorrectly believed deceptive messages were true about 69% of the time.

In fact, this bias meant that the single biggest factor determining whether participants were correct in their judgements was whether the statement was true or false: they were correct far more often for true statements than false ones. Some other idiosyncratic characteristics of the messages, beyond whether they were true or not, also had an influence on judgement accuracy, albeit small.

However, there was little evidence for the idea that some people are brilliant liars. Some of the speakers were generally more likely to be believed, whether or not their message was truthful, and the team thinks this might be because they came across as being more sincere. But this effect was very small.

The team also found that some judges were more inclined than others to believe whatever they were being told. Similarly, some participants were generally more confident in their answers. However, these people were generally no more often right than the others, "suggesting that at least some judges were not aware of their lack of lie-detection ability."

The team acknowledges that the work does have some limitations. For example, the people in the videos were all young adults, and many were psychology students. Other research has found that attractiveness and social and emotional skills can affect a person's credibility. But these factors may have varied only a little within this particular group of speakers, restricting the potential impact of the speaker on the judges' ability to spot truths and lies.

Also, in the real world, we, the general public — not to mention police officers and other interrogators — generally have a lot more to go on than a very brief video when judging whether someone is telling the truth or not.

However, a brief video is typical of research on lies, the team notes. They argue that their research shows that the message itself has been underestimated as a factor in typical lie-detection studies, which have generally focused on the speakers and/or the judges. "This work has substantial implications for the design and direction of future research," they write.