Retrieval
Bring ideas back from memory.
Research
Built on Evidence.
Learning science matters when it changes what the product does.
Practice, support, review, and mastery checks should reflect how learning actually lasts.
Principle to productEvidence
- Retrieval and spacing decide when important ideas should return.
- Feedback and worked examples shape support before misconceptions harden.
- Fading checks whether the learner can continue without the scaffold.
- Transfer tasks test whether understanding survives when the problem changes form.
Principles
What guides the product.
The research language can be technical, but essential. The product behavior should be simple.
Spacing
Return after time has passed.
Interleaving
Choose methods, not patterns.
Feedback
Make mistakes useful.
Examples
Show good reasoning before solo work.
Fading
Reduce support as independence grows.
Mastery
Move forward with evidence.
Transfer
Practice when the form changes.
Responsible claims
No exaggerated promises.
Polus uses established findings to guide product design. It does not promise higher grades, universal success, test-score gains, income, or specific life outcomes.
Mastery Realm continues to improve through ongoing development and real product use. Measured outcome claims require evidence over time.
Sources
Selected research.
These publications support the general principles described here.
Retrieval Practice
Roediger and Karpicke (2006), test-enhanced learning and long-term retention.
View publicationSpacing
Cepeda and colleagues (2006), a quantitative review of distributed practice.
View publicationInterleaved Math Practice
Rohrer and Taylor (2007), mixed mathematics problems and learning.
View publicationWorked Examples
Atkinson and colleagues (2000), learning from worked examples.
View publicationNext step