'Walking through a train station at 8am can be a nightmare'
Irishman Patrick O'Riordan is part of the London Met's Super Recogniser Unit, a group of officers with the ability to recognise faces. He tells Henrietta McKervey about helping to solve hundreds of crimes and the personal cost of having a real-life superpower
The hunt for missing teenager Alice Gross was the largest police search operation in London since the 7/7 bombings. Small and slight, the last sighting of her was on CCTV as she walked in the rain along a canal towpath in August 2014. It would be five weeks before her body was discovered, hidden under logs in the River Brent. Just days later, Latvian builder Arnis Zalkalns was found dead in nearby Boston Manor Park. Had he not hanged himself, he would have been charged with her murder.
Six hundred officers from eight forces searched 10 square miles of open land, three miles of canals and rivers, and, critically, many hundreds of hours of grainy and often poor-quality CCTV footage. It was CCTV that enabled them to piece together the events and construct the timeline that connected Zalkalns to her death. What was particularly unusual during the five-week search was that an elite team of 10 Metropolitan Police officers were formally testing a unique ability for the first time.
Known as the Super Recogniser Unit, within a couple of days the group had zoned in on Zalkalns and used CCTV images to precisely map out his movements on the day Alice Gross disappeared.
Often referred to as the Metropolitan Police Service's "secret weapon against crime", super-recognisers have a remarkable ability to instantly recognise faces they barely know. Even a fleeting glimpse of a stranger's face, or a couple of features, can be enough for a super-recogniser to make a positive identification.
A poster for Alice Gross, a murder case Det O'Riordan helped solve
Detective Patrick O'Riordan was key to the Alice Gross team. Born in Dublin and brought up both there and in Limerick, he has been living in England since his early twenties. He spent his first years in the Met Police doing plain-clothes police work (he received commendations for bravery then, and again during subsequent recognition investigative cases). However, it quickly became evident that he was unusually good at remembering faces and identifying criminals from CCTV.
"I suppose I always took my ability for granted," he says. "It only became evident when people would approach me to look at their CCTV, as I had gained a reputation for being good with faces and recalling them."
This reputation was given a significant boost in 2011 after the London Riots. Although there had been some media disapproval of the police response to the rioters, there was collective praise for the work of a single unit: the Visual Images, Identifications and Detections Office. The VIIDO is responsible for the collection, distribution and publication of all crime-scene images in London. Over 5,000 images of the riots were published, which led to many rioters being recognised and arrested. The London Riots were the making of the Super Recogniser Unit: suddenly hours of crime footage needed to be analysed quickly. Trawling through the thousands of images, the hastily assembled team identified 609 suspects - some heavily disguised - 65pc of whom ended up in court.
Not long after the London Riots, DCI Mick Neville (now retired) selected O'Riordan and 40 others from 30,000 officers and, in conjunction with Dr Josh Davis of the Department of Psychology, Social Work & Counselling at the University of Greenwich, began to study their ability in earnest.
"He started running various tests on us," O'Riordan says, "and we were also given access to a database called FILM which stores every image in London and can be sorted by crime types and even description. Because we were now doing what we excelled at, our results went through the roof. From then, people were being caught and identified correctly in huge numbers. I was in the top three for making positive identifications, which is what led me to work on the Alice Gross case."
Only unit in world
The as-yet-unknown suspect Zalkalns had been spotted cycling the same route as Alice, 12 minutes behind her. Though the footage was poor quality, and his features hard to make out, the super-recogniser team tracked his journey as he went into an off-licence then cycled to the river. When they recognised him on a different camera later in the same day, he had changed his clothes. That noteworthy detail, O'Riordan says, was what prompted a fresh search of the man's activities - and led to a murderer.
Now based in an unassuming grey stone police building in Lambeth, south London, the Super Recogniser Unit is the only one of its kind in the world. Tests conducted by Professor Mike Burton of the University of York in 2015 confirmed that against the general population, O'Riordan's ability is in the top 1pc. He is one of six officers based in the unit; around 150 others work in various police departments across London.
"One of our key tactics is what we call 'snapping'," he explains.
"We link offences together. So, if I go into the database and put in a description of say, a white male with a shaved head and sideburns, I will go through all the images that correspond and can begin to link that same person to more offences and make idents."
That he may not have seen the person for years or that they may have changed their hair or aged significantly doesn't make any difference.
"It's as though my brain is hardwired to remember them".
Within a year, the six-person unit had made over 2,000 identifications, or as O'Riordan succinctly puts it, "80pc of the idents in London last year were done by six people, whereas the average officer on the beat might get one a year".
Despite there being an estimated half a million CCTV cameras in London, footage had tended to be used ineffectively in prosecutions and court proceedings.
In his article for the British Psychological Society, 'I Never Forget A Face', Professor Josh Davis noted: "It is impossible to know the 'ground truth' as to suspect guilt, despite subsequent legal decisions; nevertheless, when confronted with CCTV images, over 75pc confess in police interview."
Clearly, the Super Recogniser Unit has had a huge impact in how CCTV images are used to link what were once considered unrelated crimes.
O'Riordan agrees. "I had a man on suspicion of dozens of thefts in the station a couple of years ago, and I spread print-outs of all the CCTV images out in front of him - there must have been 40 at least, all from different cameras in completely different parts of London - and I told him to show me which pictures weren't him… He pointed at one. His solicitor just put her head in her hands."
O'Riordan has always had this ability, but admits, "as I was born this way, I didn't necessarily understand what it meant".
He alone in his family has super-recognition, though he jokingly says his mother could recall every number plate she has ever seen. His skills have been honed over the last number of years, which has led to a significant downside.
"As with the others in the unit, I'm hyper-aware of my surroundings when out on the streets," he says. "Walking through Victoria Train Station at eight in the morning can be a nightmare because I automatically scan everyone I pass. It's a strange almost hyper-awareness, to the extent that everything around me seems to slow down and develop a greater clarity."
He always keeps handcuffs in his pocket, he says, because, in common with other super-recognisers, he tends to spot wanted criminals in the street.
Faces are critical to our understanding of the world because they provide social clues; even infants appear able to distinguish their mother from other women within two days of birth.
However, there are large individual differences in face-recognition ability. Most facial recognition is processed in the brain's fusiform gyrus, a long, thin area in both the temporal and occipital lobes (the same areas which process colour). Abnormalities in this region are associated with facial hallucinations and prosopagnosia, the inability to recognise faces that is also known as 'face blindness'. Though scientists are still trying to understand what's going on in the brains of super-recognisers, studies have found that they aren't any better than the average person at recognising things that aren't faces.
Curiously, the super-recogniser team all have different approaches to making identifications. O'Riordan's approach is based on scanning.
"Myself and maybe two others I know would be very much that way. We use the same skill pro-actively to catch people, which is why we are used a lot to spot people in very crowded environments, such as Notting Hill Carnival or at big concerts. People's faces change very little when you follow a few simple rules and have a natural ability, so what we're doing is cherry-picking people out of crowds but pretty much doing it subconsciously."
Others have described their recognition process as that of a developing image, similar to a pinhole camera effect. They are a competitive unit.
"The office is quiet every morning when the new footage comes in because we're all staring at our screens until there's a first shout of 'got one!'"
As with the others in the Super Recogniser Unit, O'Riordan's caseload is always heavy. Most days involve sitting in front of two large screens, on which he reviews thousands of images per hour. He is currently working on Operation Northleigh, the police investigation into the Grenfell Tower fire.
"As with Alice Gross, the very nature of watching victims while they are still alive, before it all happens, but knowing that they are now deceased… it has a profound impact," he says. "How could it not?"
Henrietta McKervey is the author of Violet Hill, a crime thriller about a super-recogniser working for the Metropolitan Police. It is out now and is published by Hachette Ireland