Subjects participated in two immediately consecutive experiments In the first, they either experienced a deception and debriefing, learned about deception in the abstract, did not learn about deception In the second, they either did or did not hear a reference to the possibility of a deception in that experiment A measure of incidental learning of the message in the second experiment showed that experiencing deception and learning about it in the abstract were not functionally equivalent, that only experiencing deception tended to produce absolute bias, that this bias was probably caused by “vigilance” rather than by “negativism,” that the reference to deception before the second experiment did not itself cause bias, and that the reference eliminated the difference in performance due to experiencing a deception as opposed to only learning about it in the abstract Furthermore, suspiciousness and the reported legitimacy of deception were positively related to each other at the group mean level Separating out their contribution to experimental performance showed that they complexly interacted to determine performance. These results were discussed with reference to the fear that subjects who have heard about deception from friends might perform differently from subjects who have not, with reference to the necessity for deception in some attitude change research, with reference to one alternative to deception, and with reference to explanations of why some experiments have shown no relationship between suspiciousness and experimental performance while others have shown a negative relationship