Losing Face

Episode Summary

Executive Summary: The episode explores the psychology and consequences of facial recognition, contrasting super recognizers with faceblind people. Through stories from a London police officer and a woman who struggles to remember faces, it shows that most people are overconfident but mediocre at identifying unfamiliar faces, with real stakes for policing, airport security, and everyday social life.

Main Topics: Super recognizers in policing (Priority: 5/5): London officer Alison Young describes being selected for a police unit that uses exceptional face recognition to identify suspects from CCTV and link crimes across cases. Limits of ordinary face recognition (Priority: 5/5): Psychologist Mike Burton explains that most people, including professionals such as passport officers, are surprisingly poor at matching unfamiliar faces. Familiarity and recognition bias (Priority: 4/5): People are much better at recognizing faces they know locally, and they wrongly assume others will find the same faces easy to identify. Faceblindness and social consequences (Priority: 4/5): Julie Dorschlag recounts repeated awkward encounters caused by her difficulty recognizing people, including misidentifications that have affected her social and professional life. Race, in-group familiarity, and recognition (Priority: 4/5): The episode notes research showing that people often struggle more to recognize faces from unfamiliar racial groups than faces from their own group. Technology, human judgment, and security (Priority: 3/5): The piece suggests that while algorithms may improve, human face recognition remains crucial in some investigations where humans can still outperform or complement machines.

Key Arguments: Humans are highly skilled at recognizing familiar faces but often poor at identifying unfamiliar ones, even in professional settings. Overconfidence makes people believe they are better at face matching than they really are, which can distort expectations in security and everyday interactions. Specialized individuals, such as super recognizers, can add substantial value in police work by identifying suspects from poor-quality or partial images. Face recognition ability varies widely across individuals, with some near the top of the spectrum and others far below average. Face recognition difficulties have serious consequences in criminal justice, where mistaken or missed identifications can affect investigations and sentencing. People tend to generalize their own familiarity to others, wrongly assuming other observers will find the same faces easy to identify. Social friction often follows faceblindness, as others may interpret failure to recognize them as rudeness, arrogance, or prejudice. Our modern, high-contact society demands a cognitive skill that evolved for much smaller communities, creating persistent mismatch between brains and environments.

Data Points: Super recognizer test ranking: 2nd out of all officers - Alison Young’s score on face-recognition testing led to her placement in the Super Recognizer Unit. Criminals linked per suspect: around 25 to 30 different crimes - Alison explains how snapping can connect one suspect to multiple offenses. Age of complainants: 21, 15, and 16 - Transit police case involving inappropriate touching on buses. Time since witness recognition: 15 years before - Marty Dorschlag recognized a restaurant server from a prior encounter in Columbus, Ohio. Age when Julie’s mother met school acquaintances: 7 or 8 - Alison describes her mother remembering people from primary school decades later. Distance from familiarity example: 90 minutes - Narrator stared at an actor during a play before realizing he was an NPR colleague.

Pivotal Quotes: "I'm just used to being embarrassed." — Julie Dorschlag: Julie describes the emotional toll of repeatedly failing to recognize people. "That is exactly what I'm telling you." — Mike Burton: He explains that people are generally bad at matching unfamiliar faces, even for security tasks. "I don't think you could teach someone to be able to just do it at all." — Alison Young: Alison speculates that her face-recognition ability may be innate rather than learned.

Implications: Face recognition is uneven, overtrusted, and operationally important. Listeners should be cautious about assumptions in security, eyewitness ID, and social interactions, and institutions may need better training, screening, and technology to reduce errors.