A super recognizer is someone with an extraordinary natural ability to remember and identify faces, performing far above average on standardized face recognition tests. The term was coined in a 2009 landmark paper by researchers who identified four individuals scoring roughly two standard deviations above the average population on multiple face recognition measures. Super recognizers sit at the extreme high end of a spectrum that varies continuously across the general population, essentially the opposite of people with face blindness (prosopagnosia), who make up about 2% of the population.
Where Super Recognition Falls on the Spectrum
Face recognition ability isn’t binary. It exists on a smooth continuum, with most people clustered somewhere in the middle. At one end are people with developmental prosopagnosia, who struggle to recognize even close friends and family members by their faces alone. At the other end are super recognizers, who can spot a person they briefly encountered decades ago in a crowd.
Researchers have proposed that both groups are quantitatively different from average rather than qualitatively different. In other words, super recognizers don’t use some secret trick or alternate strategy for processing faces. They appear to use the same mental machinery everyone else does, just with dramatically greater efficiency. Studies comparing how super recognizers process facial features like shape and skin tone found no unique strategy separating them from typical viewers. The underlying representations they use to encode faces seem to be the same; they’re simply better at it.
How It’s Measured
There’s no single test that confirms someone is a super recognizer. Researchers typically require high performance across multiple assessments, each testing a different aspect of face processing: memory for faces learned briefly, matching unfamiliar faces shown side by side, and remembering faces over longer periods.
The most widely used measure is the Cambridge Face Memory Test Long Form (CFMT+), which asks participants to memorize a set of faces and then identify them under increasingly difficult conditions, such as different lighting, angles, or visual noise. To qualify as a super recognizer in research studies, participants generally need to score above 90 out of 102 on this test. But scoring high on one test alone isn’t enough. The standard approach requires performance at least 1.96 standard deviations above the typical population on at least two out of three validated tests.
Because widely used tests like the CFMT+ and the Glasgow Face Matching Test are increasingly shared online, which can compromise their reliability through practice effects and exposure, researchers have developed newer screening tools. The UNSW Face Test, for example, was built specifically as an online screening tool that can reliably identify super recognizers in large-scale testing without undermining the diagnostic power of the established lab-based measures.
What’s Different in the Brain
Neuroimaging research has identified measurable brain differences between super recognizers and people with face blindness. Super recognizers show larger face-selective areas in the part of the brain responsible for processing faces, a region in the lower back of the brain that responds more strongly to faces than to other objects. They also display higher face selectivity in this region, meaning it fires with greater intensity and specificity when looking at a face compared to other visual stimuli. Super recognizers additionally show stronger responses to faces in a region of the front temporal lobe on both sides of the brain, an area involved in linking faces to identity and personal knowledge.
Genetics Play a Major Role
Face recognition ability is one of the most heritable specific cognitive traits ever measured. A twin study found that identical twins had a face recognition score correlation of 0.70, more than double the 0.29 correlation seen in fraternal twins. That gap is strong evidence for a large genetic contribution. Notably, face recognition scores showed low correlations with both general visual recognition and verbal memory, suggesting that the genes involved are largely specific to face processing rather than reflecting broader intelligence or memory capacity.
This means super recognition is overwhelmingly something you’re born with, not something you can train. While experience with faces matters (people are generally better at recognizing faces from their own racial or age group), the ceiling of your ability appears to be set by your genetics.
Is It Specific to Faces?
Whether super recognizers are also unusually good at recognizing non-face objects has been a contested question. Some research suggests face recognition is largely independent from recognizing other categories of objects, pointing to a specialized, face-specific system. But newer work complicates that picture. Researchers have found that face and object recognition may reflect a shared underlying ability to discriminate between visually similar items, and that this ability becomes apparent only when someone has enough experience with a given category.
In practical terms, someone with high face recognition ability who also has deep experience with, say, birds or cars might prove unusually skilled at telling those apart too. The implication is that super recognition could reflect a domain-general visual discrimination talent that happens to show up most obviously with faces, since nearly everyone has extensive experience looking at faces from birth. This same ability could theoretically support performance in fields like radiology, fingerprint analysis, or airport security screening.
How Police Use Super Recognizers
London’s Metropolitan Police Force was among the first law enforcement agencies to formally recruit super recognizers, establishing a dedicated team based at New Scotland Yard’s Central Forensic Image Unit. Team members are identified from within the police force based on demonstrated interest in face recognition and performance on internal facial memory testing. Their tasks include identifying suspects from security camera footage and spotting wanted individuals in live crowd settings.
In formal testing, Metropolitan Police super recognizers on the team significantly outperformed their colleagues on standardized face recognition measures. Their ability to match a grainy CCTV still to a suspect photograph, or to recognize someone seen briefly at a crime scene weeks earlier, gives investigators a tool that facial recognition software still struggles to replicate reliably, particularly when image quality is poor or faces are partially obscured.
What Daily Life Feels Like
Before they learn there’s a name for their ability, many super recognizers assume everyone experiences faces the same way. One super recognizer described recognizing a woman she hadn’t seen in 36 years from a split-second glimpse, then approaching her enthusiastically only to be met with a completely blank stare. That pattern, recognizing people who have no memory of you, is a common experience. It can feel socially awkward, even unsettling, before you understand what’s happening.
Some super recognizers report initially worrying they were being followed when they kept noticing the same stranger in a shopping center, only to realize they were simply re-recognizing an ordinary person going about their day. Once they understand the ability, most describe it as a relief rather than a burden. The main annoyance, as one super recognizer put it, tends to be pestering friends and family with observations like identifying a background extra from a 20-year-old movie.