How does the type of learning material impact what is learned? The current research investigates the nature of students’ learning of math concepts when using manipulatives (Uttal, Scudder, & DeLoache, 1997). We examined how the type of manipulative (concrete, abstract, none) and problem-solving prompt (metacognitive or problem-focused) affect student learning, engagement, and knowledge transfer. Students who were given concrete manipulatives with metacognitive prompts showed better transfer of a procedural skill than students given abstract manipulatives or those given concrete manipulatives with problem-focused prompts. Overall, students who reported low levels of engagement showed better learning and transfer when getting metacognitive prompts, whereas students who reported high levels of engagement showed better learning and transfer when getting the problem-focused prompts. The results are discussed in regards to their implications for education and instruction.