Eyewitness identification plays a key role in the justice system, but there is always the potential for error – sometimes with profound consequences.
Researchers at the University of Birmingham expert are helping to reduce the potential for such errors with a novel interactive line-up procedure which allows witnesses to rotate and dynamically view line-up faces from different angles.
The technique significantly improves witness discrimination accuracy compared to the widely used procedure using static frontal-pose photos.
Professor Heather Flowe is one of the experts who developed the advanced technique and she demonstrated its use to an audience of young people and their parents at the recent Royal Institution Christmas Lectures, which were televised on BBC Four.
Interactive line-ups enhance eyewitness accuracy compared to static photo lineups – illustrating how basic science can help direct the key applied policy issue of how best to conduct a police lineup and reduce eyewitness errors.
Professor Heather Flowe - University of Birmingham
Following a ‘traditional’ ID parade, which is seldom used by police, Professor Flowe guided two young volunteers through using the interactive system as part of a ‘courtroom’ demonstration of how forensic techniques could help to snare a jewel thief.
Forensic psychologist Professor Flowe commented: “Interactive line-ups enhance eyewitness accuracy compared to static photo lineups – illustrating how basic science can help direct the key applied policy issue of how best to conduct a police lineup and reduce eyewitness errors.
“The Royal Institution Christmas Lecture gave us a fantastic opportunity to present our research to a wider audience. We demonstrated how it can be used in a judicial setting to help increase the identification of guilty suspects and minimise identification of innocent suspects - potentially preventing miscarriages of justice.”
The researchers developed the interactive line-up procedure using basic science evidence that face recognition is aided by active exploration of 3-D facial features and motion cues. They confirmed that enabling witnesses to actively explore faces and view them from multiple increased identification accuracy.
They found that interactive line-ups improved witness accuracy compared to the sequential static photo line-ups used in around 5,000 law enforcement agencies in the US and in other countries worldwide.
“Eyewitness misidentifications are common and reliance on unreliable identifications can result in innocent people being incarcerated while guilty people remain at large to commit further crimes,” added Professor Flowe. “We must, therefore, build science-based procedures to support eyewitness identification accuracy.”