Instruction Then Problem-Solving, or Vice Versa?

by Kristen Mosley

The need and desire to actively involve students in learning experiences is evergreen, yet the timing or ways in which an instructor can do so vary widely. While some instructors prefer instruction-first approaches, others take an inverted route with problem-solving first approaches. Yet the question remains: which design is more impactful for student learning?

Let’s first break down the two approaches:

1. Instruction-first approach: this instructional design is sequenced as it is named. First, students receive instruction on a topic. Next, students engage in opportunities to apply or problem-solve with the topic at hand. The thinking behind this traditional approach is that if students lack prior knowledge on the topic, then receiving direct instruction prior to engaging in problem-solving should cue them to the key features of the problems to be solved. Equipping students with this knowledge prior to asking them to engage in novel problem solving should therefore lessen the burden on their working memory capacity.¹

2. Problem-solving first instruction: this instructional design is sequenced in the opposite fashion of its predecessor. First, students engage in a novel problem involving a yet-to-be-learned topic. Then, students receive instruction on the topic that they just explored. Proponents of this instructional design believe this approach better prepares students for future learning and applying, as they will inevitably find themselves in novel situations and with inadequate prior knowledge.² In essence, this approach emphasizes learner agency over attempts to control working memory capacity, as this hurdle is inherent to most problem-solving situations.

Of additional note, one more term is needed to understand these two different approaches and how they have been found to differentially impact student outcomes. Productive failure is a unique component of problem-solving first instruction in which students are engaged in problem-solving that is intentionally designed to result in failure.³ However, this instructional aspect is not inherent to problem-solving first approaches. That is, while productive failure necessitates the use of problem-solving first instruction, the use of problem-solving first instruction does not require the use of productive failure. Instead, productive failure is a sub feature that can, and perhaps should,⁴ be used within problem-solving first instructional designs.

A recent meta-analysis examined 53 studies that compared the impact of instruction-first and problem-solving first approaches on student learning. Roughly one-third of the comparisons in this study (36.7%) examined research with undergraduates, and roughly half of all comparisons in the study examined differences in student outcomes that resulted from quasi-experimental designs within real classroom settings.⁴ The findings therefore hold critical implications for undergraduate teaching and learning. 

Amongst the many findings of this study, a significant effect size in favor of problem-solving first approaches (Hedge’s g = .36) demonstrated that, when comparing students’ conceptual knowledge and transfer outcomes in the two different instructional settings, a problem-solving first approach was more effective than an instruction-first approach.⁴ Having found this effect for conceptual knowledge and transfer outcomes, though, it is important to also note where effect sizes were insignificant.

The benefits of problem-solving first instruction did not prove to be significantly greater than instruction-first approaches for procedural learning nor did they prove to be as strong when conducted without maintaining fidelity to productive failure instruction. That is, results suggest problem-solving first instruction that also includes productive failure design supersedes the use of problem-solving first instruction alone.⁴

In sum, evidence suggests problem-solving first instruction, as compared to instruction-first approaches, is a stronger facilitator of students’ long-term, conceptual learning as well as their ability to later transfer this learning to other domains. The benefits of problem-solving first instruction were also strongest when conducted in tandem with productive failure…which is the topic of our next post! Come back next week to learn more about productive failure design and how this approach can be incorporated into your instruction. 

1. Sweller, J. (2020). Cognitive load theory and educational technology. Educational Technology Research and Development, 68(1), 1-16.

2. Loibl, K., Roll, I., & Rummel, N. (2017). Towards a theory of when and how problem solving followed by instruction supports learning. Educational Psychology Review, 29(4), 693–715. https://doi.org/10.1007/s10648-016-9379-x

3. Kapur, M. (2008). Productive failure. Cognition and Instruction, 26(3), 379–424. https://doi.org/10.1080/07370000802212669

4. Sinha, T., & Kapur, M. (2021). When Problem Solving Followed by Instruction Works: Evidence for Productive Failure. Review of Educational Research, 91(5), 761–798. https://doi.org/10.3102/00346543211019105