Abstract: This paper reports on a randomized field experiment to study how teenage girls in Cameroon respond to different school-based HIV education programs. This experiment was specifically designed to test whether the type of HIV information messenger as well as the type of information being provided, affect how much information is retained, how much information is transmitted to peers, and how reported behavior changes among girls and their friends. We first see that lasting impressions of HIV education sessions may not be what matters most when it comes to knowledge and behavior. Second, our results show that there is no silver bullet when it comes to HIV prevention information. We find that a given message or a given delivery type can have different effects in different areas of Cameroon, sometimes of opposite sign. This suggests that local contexts are critical ingredients to be considered when designing prevention campaigns. Finally, our results show the presence of large spillovers on friends, which should encourage even more public intervention in the domain of HIV education. We discussed some potential explanations to the heterogeneity of the impact of the interventions on sexual behavior based on what we observe in our data, but these potential explanations are just correlations and we cannot draw firm conclusions from that. Our attempt to understand the heterogeneity rather provides new hypotheses that should be rigorously tested in future research on HIV education.