Witness identification may be the leading cause of wrongful convictions in the United States, accounting for nearly 75% of convictions overturned through DNA testing.
Some parts of the error rate we can’t control; others, we can. For example, we can use “double-blind” police lineups, where neither the witness nor the officer knows which person is the suspect. And we can present the possible suspects to the witness, or show their photos, serially – that is, one at a time – rather than all at once.
Not only can we do these things, moreover, but we must, as decades of social science research have shown. If we don’t, we may continue to live with a problem that the Supreme Court once said “probably accounts for more miscarriages of justice than any other single factor — perhaps it is responsible for more such errors than all other factors combined.”