How does the face memory test work?
8 faces appear one at a time, 2.5 seconds each. Then you see 16 faces, one at a time: the 8 you studied mixed with 8 new ones, in a random order. For each, answer Seen or New, with the buttons or the S and N keys. Every face is drawn by this site from features such as head shape, hair, eyes, nose, mouth, glasses and beard. None is a photo, and none is a real person.
How is your result calculated?
Your result is how many of the 16 faces you sorted correctly. You also see how many of the 8 studied faces you recognised, and how many new faces you mistook for ones you had seen. Within a session every two faces differ in at least 3 features, so no two are near-twins.
Is this the Cambridge Face Memory Test?
No. The Cambridge Face Memory Test, published by Bradley Duchaine and Ken Nakayama in 2006, uses photographs of real faces with the hair hidden, and researchers use it to study face blindness, also called prosopagnosia. This test uses drawn faces whose features you can name, like curly hair or round glasses, so it measures how well you remember those features more than how you recognise real faces. It can't tell you whether you have face blindness.
How to remember faces better
- Pick two features per face. "Glasses and a beard" is quicker to check later than a vague impression.
- Give each face a name. A made-up name tied to a feature ("Curly Carl") gives the memory a handle.
- Watch for near-misses. New faces often share some features with studied ones. Check the whole face before answering Seen.
This is a recreational test, not an assessment. If you often fail to recognise people you know, talk to a doctor. For memory for words, try the Verbal Memory Test; for memory for places, the Object Location Memory Test.