Attempt before instruction

Try to solve a new type of problem with your current knowledge before any formal teaching.

Why it works

Attempting a problem activates relevant prior knowledge and surfaces its limitations — the learner discovers exactly what is missing, which creates specific, felt gaps. When subsequent instruction addresses these gaps, it lands on prepared cognitive ground: the learner knows what the instruction is for and why it is needed. Instruction-first learners receive the same information but without the prior activation that makes it stick.

How to do it

1. When facing a new concept or problem type, attempt at least two different solution strategies before looking anything up.
2. Document your attempts in writing — what you tried, what it got you, where it broke down.
3. Only after genuine effort (at least 10 minutes), proceed to the instruction or worked example.
4. As you read the instruction, actively match each element to one of your prior attempts.

Evidence

Kapur (2014) ran randomized experiments showing that students who solved problems before instruction demonstrated better conceptual understanding and transfer than instruction-first students, despite equivalent or lower procedural accuracy. (rct)

Most controlled studies are in mathematics education; transfer to other skill domains is theoretically grounded but less directly tested. The benefit applies specifically to conceptual understanding — procedural accuracy may not exceed instruction-first approaches.

Sources

• Kapur (2014), "Productive failure in learning math," Cognitive Science
• Kapur & Bielaczyc (2012), "Designing for productive failure," Journal of the Learning Sciences

Common mistake

Giving up after the first failed attempt and going straight to the answer, which prevents the full activation of prior knowledge and the felt gap that makes subsequent instruction effective.