Look-alike and sound-alike drug names are a concerning source of confusion and medication errors. Although drug names currently undergo tests to assess their potential for confusion prior to approval, these tests have not reliably predicted real-world error rates. This study describes the development and validation of four drug name memory and perception laboratory tests. Eighty participants completed the tests and their results were analyzed against actual errors in two large outpatient pharmacy chains. The laboratory tests performed very well, demonstrating a strong association between drug name confusion errors seen during testing and those observed in real-world experience. The authors suggest that regulators and drug companies consider using these tests prior to approval of new drug names.