The U.S. Supreme Court enumerated five conditions to be considered in evaluating eyewitness identification evidence: the witness' opportunity to view the criminal at the time of the crime, the length of time between the crime and the identification, the level of certainty demonstrated by the witness, the accuracy of the witness' prior description, and the witness' degree of attention during the crime. Four psychological studies tested these factors. The findings demonstrate the complexity of the certainty-accuracy relationship. Certainty and accuracy are sometimes moderately related and sometimes not at all. No relationship was found between description accuracy and identification accuracy; witnesses who accurately describe a target person are no better able to identify that target person than witnesses who cannot describe the target accurately. People who focused on the target persons' faces were better able later to identify the target persons than were people who focused on what the target persons were holding in their hands. Footnotes, tables, and a list of 102 references are provided.