In April 2021, Tyrone Allen pleaded guilty to committing four bank and credit union robberies that occurred in Portland, Oregon. But when Allen initially became a suspect, his case posed a unique challenge: how to best construct a lineup to assess Allen’s potential involvement. Surveillance video and descriptions from four bank tellers suggested a culprit with no tattoos—yet Allen has large tattoos on his forehead and right cheek. The police included an altered version of Allen’s picture in a photo lineup, ostensibly reasoning that the culprit’s face was partially disguised and that Allen could have hidden his tattoos with makeup before committing the crimes. The police used Photoshop on Allen’s mug shot to “[paint] over the tattoos . . . Almost like applying electronic makeup.” When the lineup was presented to the four bank tellers, three selected Allen. At trial, the defense argued that the identifications should be suppressed because changing Allen’s mug shot was suggestive. The prosecution countered that the police altered Allen’s mug shot to “prevent [misidentifications of] the suspect,” because the tattoos could have made the suspect stand out in the lineup. Ultimately, the identifications were admitted at trial.
The Allen case raises an important issue: How should police construct lineups when the suspect and/or culprit has a tattoo? Despite the importance of determining how tattoos and other distinctive features affect identification accuracy and the best remedy in such situations, the empirical literature in this area is small. Generally, research shows that when the suspect stands out in a lineup—that is, when their appearance differs noticeably from the other lineup members—they are more likely to be identified than if they do not stand out, regardless of their guilt or innocence. Whatever causes the suspect to stand out (e.g., a smile, their clothing, a tattoo) can lead an eyewitness to infer that this individual is the suspect. The eyewitness may then identify the suspect because they believe the suspect must be the culprit, rather than because the suspect matches their memory of the culprit.
Regarding tattoos and distinctive features, a 2016 study by Melissa Colloff, PhD, and colleagues had some eyewitnesses view “unfair” lineups—where only the suspect had a distinctive feature—while others viewed “fair” lineups. For the fair lineups, researchers replicated the suspect’s distinctive feature on all lineup members or concealed, pixelated, or blocked the area where the feature would have been on all lineup members. Consistent with prior research, innocent and guilty suspect identifications were higher in the unfair lineups. The various fair lineups resulted in similar eyewitness performance to each other; however, other researchers have found that replicating distinctive features is the best way to create fair lineups.
These findings correspond with the U.S. Department of Justice’s rules, which state that if a suspect has a unique feature, police should select filler photographs of people with similar unique features. If police cannot find appropriate fillers, they are instructed to black out the feature and duplicate the blackout on the fillers.
But what about the Allen case, where there was a mismatch between the appearance of the culprit and the suspect? In two recent preprint studies, Jamal Mansour, PhD, and colleagues found that in unfair lineups, innocent suspects with a tattoo similar to the culprit’s were identified as often as innocent suspects with a tattoo identical to the culprit’s. However, very different tattoos protected innocent suspects. They concluded that a mismatch between a suspect’s and a culprit’s tattoo affects identification performance. But how to construct a lineup for a tattooed suspect when the eyewitness did not mention a tattoo remains an open question. This reiterates the need for more research on distinctive features—including tattoos—to ensure these identifications are reliable.
