Facial identification experts often rely on a single feature, like the ear, to make identifications, according to a new study in PLOS One.
Facial recognition experts use single features often
Experts in facial identification often rely on a single feature, such as the ear, to make their decisions, according to a study published in PLOS One on October 21, 2026. The research, conducted by Carina Hahn and colleagues from the National Institute of Standards and Technology (NIST) in Maryland, found that 76 percent of decisions were made after examining just one feature.Study involved specialists from U.S. federal agencies
The study tested 27 specialists hired between October 2023 and July 2024 from U.S. federal agencies. Each participant performed 24 trials, during which they viewed two photos of different people and a third, masked image representing one of the two. The experts were shown fragments of the hidden face, including features like the eyes, nose, or ears.Participants were penalized for errors and extra features
To encourage efficiency, a point penalty system was used, where participants were penalized for each error and for examining more than one fragment of the hidden face. The images used in the study came from an existing identification test known for its difficulty, ensuring the findings were based on challenging scenarios.FBI funded the study on facial identification
The research was funded by an agreement between NIST and the FBI. The findings suggest that experts’ strategies for facial identification may differ from the recommendations of the Facial Identification Scientific Working Group, which advises comparing the entire face feature by feature. The results remain to be tested in real-world conditions.
