This session will review the basic science of how memory works, discuss its application to the domain of eyewitness identification, and share a firsthand account by an exoneree who was wrongfully convicted based on confident testimony at trial by witnesses who cleared him on the initial test. Research in cognitive science suggests that the most reliable information from a witness's memory comes from the first test (e.g., a photo array) conducted early in a police investigation. Moreover, in DNA exoneration cases, many witnesses who confidently misidentified an innocent defendant at trial provided accurate evidence of innocence on the initial test (e.g., by rejecting the lineup). The initial test not only yields the most reliable information but also unavoidably contaminates the witness's memory. Thus, subsequent test results should be disregarded, as they have been compromised by memory contamination.