The usual suspects?

Eyewitnesses of crimes are often required to make an identification decision as to whether the police suspect is, or is not, the real culprit of the crime. In an identification parade or lineup, an image of the suspect is presented alongside images of similar-looking people (who are known to be innocent) and the witness has to identify the culprit if he or she is there, or state that the culprit is not present.

In our study, we wanted to find out how the police should accommodate suspects with distinctive facial features (e.g. tattoos, scars, piercings, bruising) in lineups. If the police suspect stands out in a lineup because they have a distinctive feature, this is not a good test of the witness’s memory. The witness might pick the suspect simply because it is obvious that they are the focus of the police investigation. Alternatively, the witness might pick the suspect just because they are the best match – but not necessarily an exact match – to their memory of the culprit, compared to the other lineup members. Basically, if a distinctive suspect stands out, a witness is likely to pick the distinctive suspect, even if they are not the real culprit. Estimates suggest that over one-third of all police suspects have distinctive facial features. But police guidelines on how to accommodate distinctive suspects in lineups are not currently guided by research – in fact, there are thousands of studies on lineups, but only a handful that explore lineups for distinctive suspects.

In our study, we examined three techniques currently used by the police to prevent distinctive suspects from standing out and compared these techniques to doing nothing to prevent the distinctive suspect from standing out. Let’s say the suspect has a black eye. One thing the police might do is digitally add a black eye to all of the other faces in the lineup (replication). Alternatively, they might cover up the suspect’s black eye and cover up a similar area on the faces of the other lineup members. In practice, the police can cover up the feature by either overlaying the area of the feature with a black block (block), or by pixelating the area of the feature (pixelation). In our study, we compared replication, pixilation, and block lineup techniques against lineups in which nothing was done to prevent a distinctive suspect from standing out – that is, lineups in which the suspect was the only person with a black eye.

Replication

Pixelation

Block

Only the suspect has the distinctive feature

Key findings

When the police suspect was the only person with the distinctive feature, people were likely to pick the suspect regardless of whether that person was innocent or guilty. But what’s worse, we found that when the suspect was the only person with the distinctive feature, this actually made people more likely to confuse who was guilty and who was innocent. That’s because people weren’t really using their memory of the culprit’s face; they were simply picking the only plausible option – the only person in the lineup with the black eye that they remembered from the crime video – and this made it difficult for people to tell the difference between the real culprit and an innocent suspect who had a similar feature. Worryingly, people also made high-confidence errors: they selected an innocent suspect yet were confident that they had got it right. This finding warrants concern because we know from other research, and real-life cases, that highly confident witnesses can be very influential in the courtroom.

Eyewitness identification decisions can influence how a case progresses through the criminal justice system. An incorrect identification can result in a guilty person going free, or an innocent person being charged with a crime they did not commit. Our results show that it’s crucial to prevent distinctive suspects from standing out in lineups. Interestingly, we found that all three techniques currently used by the police were equally effective. This means that there are multiple ways in which the police can accommodate distinctive suspects in lineups. It doesn’t seem to matter if you add the suspect’s feature to the other faces, or cover up the suspect’s feature using a block or a pixelated area – all three methods are much better than merely leaving the distinctive suspect to stand out.

Legal argument

Legal professionals could use our research findings to ensure that identification procedures for suspects with distinctive features are fair and effective. Our study suggests that eyewitness identification decisions are substantially less accurate when the suspect is the only person in the lineup with the distinctive feature. What’s more, when the suspect is the only person in the lineup with the distinctive feature, erroneous identification decisions are even made with high levels of confidence. Therefore, legal professionals could check the identification parade images used by the police to ensure that the suspect does not stand out.

Melissa Colloff is a PhD student and Kimberley Wade an associate professor in the psychology department of the University of Warwick, and Deryn Strange is an associate professor of psychology at the John Jay College of Criminal Justice, Central University of New York

www2.warwick.ac.uk

www.jjay.cuny.edu

@warwickuni

@JohnJayCollege