Space your practice.
Memory research shows why returning to material across separate sessions matters. Spacing is a core principle in Paivio’s review schedule.
Cepeda et al., 2006 ↗The science behind PaivioBuilt from memory science.
Built from memory science.
Engineered by Paivio.
We began with FSRS, the open-source scheduling algorithm also supported by Anki. On that foundation, we built our own concept-mapping, review-ordering and retrieval systems.
Where we started
The foundations of lasting knowledge.
Retrieve the answer.
Actively recalling information is a central finding in learning research. Paivio puts the question before the answer so recall is part of the routine.
Roediger & Karpicke, 2006 ↗Model memory over time.
FSRS estimates memory states from review history and uses them to set the next review. That is the scheduling foundation we build on.
Anki’s FSRS documentation ↗FSRS is the starting pointTiming is one part
Timing is one part
of the study experience.
When should a card come back?
That is the scheduling question FSRS helps answer. Paivio uses your explicit ratings to advance that schedule, so review becomes a repeatable routine.
We built the rest of the experience around the material itself: its concepts, their relationships and the ways you practice recalling them.
Return to the material over time.
An illustration of spaced review
The Paivio learning systemOur own systems,
Our own systems,
built on that foundation.
When to review.
FSRS turns your Again, Hard, Good and Easy ratings and the time since each review into every card’s next due date. It is the open-source scheduler also supported by Anki.
How ideas connect.
Our concept-mapping system builds a graph of the concepts in your own decks and the prerequisite links between them. When the format your account is assigned uses the graph, on your own decks it orders reviews by prerequisite links within the due cards. It never adds a card or changes a due date. Links that your reviews keep contradicting are pruned.
What form a card takes.
Our retrieval system picks each card’s presentation. Which study mechanisms are delivered depends on the format your account is assigned. Outside the research comparison, on your own decks, a card’s starting presentation follows how strong your memory of that card is, from multiple choice and matching up to typed recall. Multiple choice and matching are held back while an account is being measured in the comparison.
Our research programDeveloped in-house.
Developed in-house.
Studied over time.
Paivio researchThe next layer
The next layer
of the system.
Our proprietary development program extends this foundation with machine-learning methods and integrated learning systems.
Research trials and patent work are underway.
We’re continuing development and completing our research trials while progressing the patent work around our proprietary methods.
We’ll make the research findings and technical details publicly available when that work is complete. The details of our ongoing development remain private for now.
This section describes ongoing research and development. The current study mechanisms are described above.
How the research is structuredA consistent foundation.
A consistent foundation.
A focused comparison.
Shared scheduling.
Review timing, ratings and due dates are identical across the comparison formats. Your explicit ratings remain the basis of the FSRS schedule.
Fixed study formats.
Consenting accounts are assigned once to one of three fixed formats. The delivered study mechanisms vary by format.
What stays with you.
Our research measures recall at seven days or later. Participation details are available in your learning profile, and ordinary study stays available.
A closer look
About the foundation and the system.
How is Paivio connected to FSRS?
FSRS is our open-source scheduling foundation, also supported by Anki. Paivio’s concept-mapping, review-ordering and retrieval systems are separate layers built around that foundation.
What is being developed in-house?
Paivio’s own learning systems bring concept relationships, review ordering and retrieval practice together. Our ongoing research and development includes machine-learning methods.
When will the technical research be available?
We’ll publish the research findings and share technical details after the current research trials and patent work are complete.
How can I start using Paivio?
Create a free beta account, confirm your email and bring your first deck or notes.
Your next chapterBring one deck.
Bring one deck.
Start one session.
Create an account and make your next review a place to begin.
Free during beta · No credit card · For students 18+