This research examined the impact of formative quizzes on e-learning designed to teach volunteers how to tutor struggling readers. Three research questions were addressed: (1) Do embedded quizzes facilitate learning of e-content? (2) Does the announcement of upcoming quizzes affect learning? (3) Does prior knowledge interact with quizzing and learning? Participants (n=178) were randomly assigned to condition: (1) no warning, no quiz (control), (2) no warning, quiz, and (3) warning, quiz. Users demonstrated significant learning gains. A predicted main effect for quiz warning was not supported. However, the effect of quizzes on learning was significantly moderated by the amount of prior experience. Low experience tutors learned significantly more when quizzes were embedded compared with expert tutors--a classic expertise reversal effect. A new neurocognitive interpretation for this effect is presented, as well as a call for more "stealthy" assessments that take into account novelty and the learner's zone of proximal development. (Contains 4 figures, 4 tables and 4 notes.)