People identify faces of their own racial group more accurately than faces of other groups, with substantially higher false identification rates for the latter.
Meissner and Brigham's 2001 meta-analysis of some thirty years of studies found the effect robust, with other-race faces roughly one and a half times more likely to be falsely identified. The best-supported explanation is differential experience and how faces are encoded rather than any perceptual deficit, and it has direct consequences for the reliability of eyewitness identification.
In practice
- A line-up identification across racial groups carries a materially higher error rate, and the witness is no less confident.
- A face-matching system trained mostly on one population reproduces the same asymmetry.
- Further reading
- Wikipedia
- Subject
- Psychology & Behavior
- Essay
- A century and a half of measuring the mind