Abstract: Sometimes people should have known certain things they didn’t know. This can happen even in cases in which a person has no good evidence for the thing they should have known. For instance, cardiologists should know about recent major developments in research that bear on their practice; if a cardiologist didn’t know such things, they should have. Can what a person should have known matter to what they’re justified in believing? If the cardiologist believes that T1 is the best treatment for a patient’s heart condition but should have known that T2 is better than T1, could this make a difference to whether the cardiologist’s belief is unjustified, even if the cardiologist’s belief fits their evidence very well? There is some temptation to think that, yes, it can make a difference. I try to make sense of the source of this temptation: what is the intuitive reasoning that suggests that the cardiologist isn’t justified in believing T1 is the best treatment? And is this sort of justification epistemic?