Researchers display confirmation bias when they persevere by revising procedures until obtaining a theory-predicted result. This strategy produces findings that are overgeneralized in avoidable ways, and this in turn binders successful applications. (The 40-year history of an attitude-change phenomenon.