Preparatory effects of problem solving versus studying examples prior to instruction

The Productive Failure (PF) approach prompts students to attempt to solve a problem prior to instruction – at which point they typically fail. Yet, research on PF shows that students who are involved in problem solving prior to instruction gain more conceptual knowledge from the subsequent instruction compared to students who receive the instruction first. So far, there is no conclusive evidence, however, that the beneficial effects of PF are explained by the attempt to generate \(\textit {one's own}\) solutions prior to instruction. The literature on example-based learning suggests that observing someone else engaging in problem-solving attempts may be an equally effective means to prepare students for instruction. In an experimental study, we compared a PF condition, in which students were actively engaged in problem solving prior to instruction, to two example conditions, in which students either observed the complete \(\textit {problem-solving-and-failing}\) process of another student engaging in PF or looked at the outcome of this process (i.e., another student's failed solution attempts). Rather than worked examples of the correct solution procedure, the students observed examples of failed solution attempts. We found that students' own problem solving was not superior to the two example conditions. In fact, students who observed the complete PF process even outperformed students who engaged in PF themselves. Additional analyses revealed that the students' prior knowledge moderated this effect: While students who observed the complete PF process were able to take advantage of their prior knowledge to gain more conceptual knowledge from the subsequent instruction, prior knowledge did not affect students’ post-test performance in the PF condition.