This article empirically supports the thesis that there is no clear and unequivocal argument in favor of simulations and experiential learning. Instead the effectiveness of simulation-based learning methods depends strongly on the target group's characteristics. Two methods of supporting experiential learning are compared in two different complex simulations with students and apprentices as the target groups, and with knowledge acquisition and control performance as the learning criteria. As a main result, students outperform apprentices in knowledge acquisition and transfer performance in general. And whereas students feel most self-efficient in the guided exploration learning setting, for apprentices it is just the opposite: apprentices feel less self-efficient after the guided exploration learning phase. Also in respect to knowledge acquisition, students benefit from both support methods equally well in the low complex simulation, whereas for apprentices the difference is great. Transfer performance and processing time is also affected by the target group, although the simulation complexity plays a greater role here. Finally, after the experiment both target groups differed largely in their motivation to use simulation-based learning environments again. Psychologically relevant differences in target groups are discussed. (Contains 2 tables and 7 figures.)