Solve First or Study Examples? Learning Design

Overview

This series explains when problem solving before instruction can support learning, when worked examples better organize an unfamiliar rule, and when retrieval practice strengthens access. It connects Productive Failure, meaningful variation, feedback, and transfer to a conditional design: diagnose the learner’s representation, match the next cognitive job, and test a changed case. Evidence favors PS-I on average but varies by learners and domain; controlled experiments found retrieval stronger after instruction, while examples or varied retrieval helped when rules had to be inferred. Immediate generalization does not establish durable transfer.

Learning outcomes

PS-I can improve later learning on average, especially when high-fidelity Productive Failure makes the first attempt useful to later instruction.
Worked examples can support rule construction when learners lack a usable representation or independent generation would overload them.
After a rule is taught, retrieval practice with feedback can strengthen access and immediate generalization, though it may take more time.
Meaningful variation can support transfer by enabling comparison, but it can slow practice and does not guarantee generalization.

Keywords

Problem solvingProductive FailureWorked examplesRetrieval practiceTransferVariationFeedbackPrior knowledgeLearning design

Chapters

PS-I vs I-PS: Learning Order

Separate practice performance, retention, and transfer

9 min · Separate practice performance, retention, and transfer

9 min

Productive Failure: When Solve-First Works

Turn an initial attempt into useful instruction

7 min · Turn an initial attempt into useful instruction

7 min

Rules, Schemas, and Retrieval Practice

Match guidance to the learner’s current representation

10 min · Match guidance to the learner’s current representation

10 min

Choose Methods by Learning Job

Match representation, task, and transfer target

9 min · Match representation, task, and transfer target

9 min

Variation, Repetition, and Transfer

Preserve structure while changing surface cues

8 min · Preserve structure while changing surface cues

8 min

Retrieval vs Examples: Rule Availability

Why initial instruction changes the comparison

11 min · Why initial instruction changes the comparison

11 min

Design a Learning Sequence That Adapts

Move from first encounter to independent transfer

11 min · Move from first encounter to independent transfer

11 min

Choose Between Solving and Worked Examples

Adapt the sequence as the learner’s goal changes

11 min · Adapt the sequence as the learner’s goal changes

11 min

Preview every chapter · Unlock full text, audio, and video from chapter 4

Plans

Ch.1-3 FreeCh.4-8 Premium

FAQ

Should learners solve a problem before seeing an explanation?

They should solve first when a meaningful attempt can expose a structure that later instruction will organize; otherwise, guidance or a worked example may be better.

When should learners study worked examples instead of retrieving a rule?

They should study worked examples when the rule or procedure is not yet usable and generating it would overload them; retrieval is more appropriate once the rule is available.

Why use varied practice instead of repeating one item?

Use varied practice when transfer across cases matters, because related instances can reveal a shared structure; expect slower practice and do not assume variation alone guarantees generalization.

How can you tell whether a learner transferred a rule?

Test a new case that preserves the target relationship while changing surface details; add delayed retrieval when durable learning matters.