I was curious what you wrote but a search turned up no comments/messages from you on Patreon (likely their bug!)—do you happen to remember what post you commented on?
Conversation
Must have been around end of August 2020. Patreon don't let me go so far. Once you unsupport all archive is gone.
I'm not sure if my feedback went as a comment on a post, a tweet or elsewhere.
1
1
my feedback was essentially asking why u are so focused on spaced repetition & didn't explore other ways like retrieval practice or generation.
Was short.
A few months later you explained (a post or talk) that you will engage more w/ your sponsors but I was on my way out anyway
1
1
Thanks for looking—I can't find it on either Twitter or Patreon.
One curiosity: spaced repetition (at least as I've been working with it) necessarily involves retrieval practice; in fact, that's the mechanism at play. Is there some other sense of that phrase you mean?
1
2
I don't agree. SR is a special case of space practice but retrieval practice is a lot more general and don't imply repetition. You could very well vary retrieval practice. One time check terms/definition, another time retrieval and again later generation and finally an MQ.
1
SR works very on terms,unequivocal Q/A like extending foreign language vocabulary, acquiring jargon or exact definitions.
I used flashcards a lot for all those purposes.
but two/three main drawbacks ...
1
1/ Knowledge is rarely forgotten completely, what is forgotten is the index, the key. The way you learn (ex: list of formulas) is rarely the same you will use to retrieve it at the time of apply. Multiplying the pathways that leads to K is essential. SR don't encourage that ...
1
2/ Preparing cards takes time and is limited to the understanding at that time. Time better used at enlarging your knowledge to understand deeper ( like reading examples of use, or how it became so).
1
3/ SR encourage narrow views. One problem, one answer in isolation of the context, the history, prior and future knowledge or gaps. Generation and retrieval overcome that.
2
Yep, these are all good criticisms of SR as ordinarily practiced; and it's true that SR generally implies a specific subset of all possible retrieval practice.
I'm quite interested in practice mechanisms which invoke generation, connection, etc.
1
1
That said, I suspect you may underrate the value of SR for non-rote knowledge. Lots of literature on SR for conceptual learning; see e.g.:
researchgate.net/profile/Andrew
mrbartonmaths.com/resourcesnew/8
Replying to
there are use cases for sure. But a lot less than people in tft advocate.
I think the last objection is that it's easy to code. So for software vendors it's an easy answer, box checked. No need to look further. work serves as a cover, a validation, sad.
1
1
Yes, I think this objection is quite true.
1
Show replies
Replying to
First one: It's in a school setup, directed learning. It's not the same as SDL where the learner is both learning and organizing his learning. The conclusion I read ..
1
I don't discuss it "repeated practice at retrieving information from
memory produced better transfer to several different types of
questions than repeatedly studying the same information"
But the transfer back to applying is not covered, long term duration of memory either,
Replying to
Second is a lot more interesting for me as it compares with some kind of generation using concepts maps.
I'll need some help for making an opinion. did you read this one?
2
1
I love the conclusion and I refer back to the very root of the convo with : "pave the way for the design of new educational activities based on consideration of retrieval processes"
That's what knowledge graph based TFT coders should aim to. possibly in collaboration