source: https://old.reddit.com/r/Anki/comments/i8mt06/when_you_find_anki_and_rsealise_how_inefficient/
Over the years, a few essays and blog posts have improved public understanding of SRS (Spaced Repetition Software) and generally ‘leveled up’ the community. In my opinion, some of these essays are so critical that you need to read them to have a current understanding of Anki. Unfortunately, they risk getting buried by newer posts because of the recency bias of search engines.
If the first resource for understanding Anki should be the manual, the second stop should be the practical, encyclopedic essay by Gwern:
Spaced Repetition for Efficient Learning, by Gwern Branwen
My key takeaways:
- the spacing effect exists
- forgetting curves exist
- the spacing effect is counterintuitive, which probably contributes to the niche status of SRS
- cramming, predictably, yields weak memories
- studies show that active recall, testing, and spacing (vs. massing) improve retention for most domains
- the quality and format of questions impacts retention
- SRS does not help with comprehension, which should come first
- SRS is good for accumulating knowledge
- Gwern uses Mnemosyne (I personally recommend Anki instead)
- use SRS for useful knowledge only
- the 5-minute rule is an example heuristic for choosing what to Ankify: “if, over your lifetime, you will spend more than 5 minutes looking something up or will lose more than 5 minutes as a result of not knowing something, then it’s worthwhile to memorize it with spaced repetition”
- SRS daily reviews decay in size if you don’t add new cards
- SRS daily reviews can be quite short for even large decks
- before bed may be the best time to do SRS reviews
- cards can be generated algorithmically
- an SRS-powered knowledge repository can be thought of as an anology to rainbow tables; by making certain data available instantly to your brain you free up resources for more important computation
- SRS is still a hidden gem
- Gwern maintains an outstanding list of external links
Augmenting Long-term Memory, by Micheal Nielsen
This essay is famous because of Micheal Nielsen’s effort to articulate his theories about memory and his useful observations.
My key takeaways:
- digital memory augmentation is culturally intertwined with computing history
- memory is fundamental to our thinking
- dismissal of ‘rote memory’ may be misguided
- “Anki makes memory a choice, rather than a haphazard event, to be left to chance.”
- Anki can be used to learn complex and deep material
- Anki’s SRS algorithm improves study time by a factor of 20 over conventional flashcards
- Nielsen’s heuristic is a 10-minute rule, plus facts he finds “striking”
- Nielsen uses Anki in all parts of his life
- Anki is good for new subjects, and it’s gratifying to know you can reliably retain what you’re learning
- don’t use premade decks
- there should be some purpose behind the cards you’re making
- use Anki for reading papers
- “What you Ankify is not a trivial choice: Ankify things that serve your long-term goals.”
- Anki notes should be written attentively so you don’t mislead yourself
- make most Anki questions and answers as atomic as possible
- “Anki use is best thought of as a virtuoso skill, to be developed”
- avoid orphan questions, keep your knowledge connected
- review all of your cards together
- memory palaces and similar mnemonic techniques aren’t efficient for long-term learning the way SRS is
- don’t distract yourself with all of Anki’s features, options, and plugins; your time is better spent reviewing
- it may be a challenge to encode knowledge in such a way that you think of it when you need it (rather than just knowing it when prompted by Anki)
- an efficient memory system will “intrinsically be somewhat difficult to use” because you should always be testing facts that are almost forgotten
- people sometimes think they’re having an issue with the complexity of a subject when really they just need to memorize more of the requisite notation, terminology and background
- Nielsen proposes a model of memory involving “chunks” where increasing the size, amount, and quality of the chunks is the key to reasoning about tough problems, while your working memory limits the number of chunks you can be actively reasoning with:
“In particular, someone with a lower IQ but able to call on more complex chunks would be able to reason about more complex situations than someone with a higher IQ but less complex internalized chunks.
In other words, having more chunks memorized in some domain is somewhat like an effective boost to a person’s IQ in that domain.”
- scientific studies can’t catch up with the practical application of SRS because of complex psychological topics (e.g. motivation)
Other Essays
I think Effective learning: Twenty rules of formulating knowledge, by Dr Piotr Woźniak, is a must-read. The key points are already summarized and numbered in bold. Piotr Woźniak helped jump start the SRS revolution as early as the 1980s (he’s the guy with the Wired article about him).
Link Bomb
As a recap, I recommend reading the following documents.
- Read the Anki manual
- Read Spaced Repetition for Efficient Learning, by Gwern Branwen
- With less urgency, read Augmenting Long-term Memory, by Micheal Nielsen
- Read Effective learning: Twenty rules of formulating knowledge, by Dr Piotr Woźniak
- (optional) read other documents from Gwern’s list of links, Piotr Woźniak articles (scroll down), or the Anki subreddit