Mistaken eyewitness-identification testimony is at the heart of a large share of the convictions of people whose innocence was later proven using forensic DNA testing. A considerable amount is now known about how to lower the rate of mistaken identifications through the use of better procedures for conducting identification. Several procedural reforms are described, such as double-blind lineups and pristine assessments of eyewitness-identification confidence. Although numerous jurisdictions have made improvements to their identification procedures in recent years, a large share of jurisdictions have still not made significant reforms. Although some courts have been making better use of the scientific findings on eyewitness identification, most courts are still using an approach that is largely unsupported by scientific findings.