3 ms·
Warning: comment strongly biased towards a specific type of learning technology :) At a molecular level, all memory (dendritic spines) declines at roughly the
by hereme888 7y ago
Warning: comment strongly biased towards a specific type of learning technology :)
At a molecular level, all memory (dendritic spines) declines at roughly the same rate. Some memories last much longer because the dendritic spines formed were much stronger (such as traumatic event, a well-formulated memory, or a memory strongly linked to other things such as a smell). But they all decline at the same rate nonetheless.
I've been a fan of spaced-repetition learning for a few years. Programs like Anki use a variation of an old algorithm (SM-2) based on decay-rates for memory extrapolated from data gathered through tens of thousands of people over many years.
I use a similar program for rote memorization first, and then to remember concepts.
After I grasp the basics of a subject I can eventually delete that flashcard and create a concept-based flashcard that allows me to understand the forest for the trees. Example: "very frothy urine is a possible clinical symptom of what disease? Proteinuria/Nephrotic syndrome" → "what macro-nutrient, when mixed vigorously in liquid, makes it frothy? Protein". So now I understand the key concept for both nephrotic syndrome and why I should only add whey protein to my shakes at the end of the mixing to avoid creating bubbles, etc.
If I had started using these programs when young, I would have saved years of life not having to re-learn the same thing year after year, because it takes much longer to re-learn than to remember. And now imagine I hadn't been forced to rote-memorize useless things just to pass a boring exam. Years of life...
I wonder if spaced-repetition can also be used to remember skills that aren't practiced often.
The only limitation I know with spaced-repetition algorithms is that the data shows that it's not very useful for children until they're about 7 yo. This is because the algorithms are tuned to adult-level memorization, and children under 7 usually aren't nearly as good as adults at retention (for many reasons).
- harperlee 7y agoDo you have a pointer to where do you get that children under seven have bad retention? Or is it a personal observation? (this is not a "citation needed", i'm genuinely interested as I look into spaced repetition from time to time).
- hereme888 7y agoSure. It's from the personal wikipedia maintained by Dr. Piotr Wozniak, the guy who invented spaced repetition and put it into software back in 1986 (SM-2 algorithm) when computers became available to him. The relevant part of the article starts here: https://supermemo.guru/wiki/Childhood_amnesia#Plasticity-vs-stability_trade_off https://supermemo.guru/wiki/Childhood_amnesia#Plasticity-vs-... The summary of the article can be read here: https://supermemo.guru/wiki/Childhood_amnesia#Summary:_Childhood_amnesia https://supermemo.guru/wiki/Childhood_amnesia#Summary:_Child...