How Review Works
Review uses FSRS-6 to schedule your tasks. Each successful review increases the interval until the next one—here's an example of how that looks over time.
Initial Learning
Establishing a baseline for your first review.
Growing Strength
Success extends the time between reviews.
Harder Recall
High difficulty brakes the interval growth.
Adaptive Growth
The schedule adapts for difficult material.
Understanding the Numbers
When you view a task's details, you' ll see three key metrics. Here's what they mean.
Recall probability
Your probability of recalling this memory right now. It starts at 100% after learning and decays naturally over time.
Memory strength
How long your memory will last, measured in days. The stronger your memory, the longer the intervals between reviews become.
Difficulty
A dynamic score from 1 to 10 reflecting task complexity. It automatically adjusts based on your review performance.
Settings
The defaults work for most people, but you can adjust these to fit your learning style.
Target Retention
This controls the acceptable "risk" of forgetting. When a task's recall probability falls to this level, it joins your daily review queue.
The recommended setting. A good balance between memory retention and study time.
Maximum Interval
The longest Review will wait before showing a task again. Default is 100 years—effectively unlimited.
If you know something perfectly, why review it? However, you can cap this to ensure a refresher once a year (365 days) if preferred.
Algorithm Deep Dive
Interested in the mathematics powering Review? Explore the full technical specification of the FSRS algorithm, including the formulas for Recall probability, Memory strength, and Difficulty.
View Technical Spec