Self-assured witnesses are more reliable

Police data refute assumption of an already bad self-assessment

How safe is the witness in identifying the offender? © Rich Legg / thinkstock
Read out

Eye-Witnesses Put to the Test: An Houston police investigation exposes two assumptions on testimonies as false. So the self-assessment of a witness is indeed an important indicator of whether his statements are reliable or not. And a juxtaposition in which all persons are seen at the same time, is more effective than showing the perpetrator photos. Here the police and justice would have to improve, according to the researchers in the journal "Proceedings of the National Academy of Sciences".

In the investigation of an accident or crime, it is often on eyewitnesses. But whether a testimony is reliable seems more of a matter of luck. For example, studies show that witnesses who are over-nightly are more likely to remember wrongly and that a false testimony can occur even if the witness is completely sure. Among other things, it has been assumed that the self-assurance of a witness is not a suitable indicator of the reliability of his testimony.

One culprit, five innocents

John Wixted of the University of California at San Diego and his colleagues have now re-examined this by analyzing the results of 348 Houston police comparisons in 2013. Police officers in the robbery department showed 717 eyewitnesses each six photos in the context of their educational work, either in succession or at once. One photo showed the real culprit, the others were filling photos.

All witnesses were asked not only who they recognized, but also how confident they were of identification. In addition, the police did not know which photo was the alleged perpetrator - this should ensure that they do not influence the eyewitnesses by involuntary signals.

At the same time is better than succession

The result unmasked two assumptions as false: Firstly, the comparison using simultaneously displayed photos proved to be much more reliable than presenting the photos one after the other, as the researchers report. This is exactly the opposite of what is currently recommended in the US legal system. Because there are the police departments should, if possible, on the successive presentations change. display

The second finding: the certainty expressed by the witnesses themselves with regard to their verdict is of far greater importance than previously assumed. In the case of counter-reports, self-assured eyewitnesses' data proved to be significantly more reliable and more accurate than those of their own choosing, as the evaluation revealed. "Most false identifications were made by witnesses who clearly articulated their insecurity, " says Wixted.

How sure is the witness?

Therefore, it is important to take the self-assurance of the witness into account. "By the time the witnesses make their first identification, they can very well give us reliable information about their own accuracy, " Wixted said. "If you ignore the doubts of the witnesses, you make a serious mistake." After all, false testimonies are considered one of the biggest causes of wrongdoing in the US.

Conversely, according to the researchers, it is just as wrong to dismiss witness testimony as well as the self-assessment of the witnesses on the whole as unreliable. In some US courts, the jurors are even instructed not to take into account when an eyewitness seineu ert his doubts or his confidence in the identification. "That's a big mistake that our legal system is making, " says Wixted. "Because the self-assessment of a witness is indeed an indicator of its accuracy." (Proceedings of the National Academy of Sciences, 2015; doi: 10.1073 / pnas.1516814112)

(University of California San Diego, 22.12.2015 - NPO)