For years scientists have studied the biological basis of human speed, and reported that the fastest athletes are short and muscular in build. However, these conclusions were challenged in 2008 when a new athlete, substantially taller than previous world-record holders, was identified as the fastest man in history. Usain Bolt presented the purest expression of human speed on the planet – and raised the possibility that scientists may need to entirely change the way they think about human biometrics.
In the same vein, one might ask whether examinations of the brain at its height of efficiency will present new insights into its workings. Although researchers have historically examined people with a very high IQ (i.e. those with more generalised skills), it has become more and more clear that some individuals only perform extraordinarily well on specific cognitive tasks. Among the most interesting of these is facial identity recognition. In fact, the extraordinary skills of these so-called “super-recognisers” do not seem to correlate with IQ or memory for objects, yet they claim to recognise faces which they have only briefly been seen before, or have undergone substantial changes in appearance. For instance, in a recent scientific report from our laboratory (unpublished), one super-recogniser described bumping into a girl from a children’s’ swimming class he coached as a teenager. He recognised her immediately, despite the fact that he’d not seen her for over ten years and she was now an adult.
So how can these people change the way that scientists think about the human brain? For many years researchers have generally agreed that faces are “special.” That is, we have dedicated brain regions that are used only for face recognition, and we look at faces in a different manner to other objects. This “configural” or “holistic” strategy refers to our ability to see the face as a whole, taking into account the spacing between features as well as the shapes of the features themselves. The idea that faces are special is supported by findings that some people acquire prosopagnosia (face blindness) following damage to the face recognition areas of the brain, and many people with prosopagnosia can no longer process faces in a holistic way. This evidence suggests that holistic processing is essential for successful face recognition, and perhaps super-recognisers rely on this strategy to a greater extent than typical people. However, only modest links have been reported between face recognition ability and holistic processing skills, suggesting other factors may be at play.
Eye-tracking technology has frequently been used by psychologists to identify the regions of the face that are particularly informative in face recognition. Typical people tend to focus on the eyes, suggesting they carry important information about facial identity. In fact, some people with prosopagnosia avoid the eye region, and instead look at the mouth. Could it be that super-recognisers look at the eyes for an even longer period of time than typical people? Work in our laboratory recently examined this issue, comparing the eye movement patterns of super-recognisers, people with prosopagnosia, and typical participants. As predicted, the participants with prosopagnosia spent less time on the eyes, yet it also turned out that super-recognisers spent more time viewing the nose. These findings challenge existing conclusions, suggesting that it is the centre of the face, rather than the eye region, that is optimal for facial identity recognition.
In addition to providing new opportunities to further our understanding of the face recognition system, super-recognisers may also be valuable in policing and security settings. While it is easy to recognise faces of our family and friends, the recognition or matching of unfamiliar faces is a very difficult task. Many errors are made even by people who complete face recognition tasks on an everyday basis, and extensive experience in passport control or CCTV viewing doesn’t seem to help performance. In a bid to improve this situation, London’s Metropolitan Police have created a super-recogniser unit that is used to spot criminals in a crowd or within CCTV footage, and these officers helped identify numerous perpetrators in the 2011 London riots. It’s easy to spot other potential roles for super-recognisers – issues of national security are currently paramount, and they may spot wanted or missing people more readily than typical officers.
The London unit currently identifies super-recognisers by analysing years of on-the-job performance, but it would be much easier if new recruits could be identified by a simple battery of computerised tests. We recently addressed this issue by recreating some real-world face recognition tasks in the laboratory. In the first test, participants viewed a series of trials with a face at the top of the screen and a grid of 10 faces below. Super-recognisers outperformed typical participants by 20% when deciding whether or not the target face was present in the line-up. In a second test, participants viewed 20 images of faces that were extracted from good quality CCTV footage. They were later asked to recognise those faces when displayed under poor lighting conditions. While this was a more difficult task, the super-recognisers were on average 8% more likely to recognise the faces than typical participants. However, while some super-recognisers were very good at matching faces in the first task, the same individuals were not necessarily so good at remembering multiple faces in the second test.
It seems then that while some super-recognisers have superior face memory skills, “super-spotters” may be used in the policing of crowds, and “super-matchers” may be useful for visa verification or passport control – and a variety of laboratory-based tasks are needed that can identify potential candidates who will excel in a range of real-world scenarios. And these findings bring us back to our original question – what can super-recognisers tell us about the human face recognition system? Well, it seems that this “special” system can be further divided into different, independent mechanisms. Further research examining brain functioning in super-recognisers who are apt at different face recognition tasks may shed further light into the precise workings of this very special system within the human brain.
Do you think you may be a super-recogniser? Take our test and find out: http://prosopagnosiaresearch.org/super-recogniser