Wednesday, September 29, 2021

 So this blog is and has been long since abandoned a number of things have shifted and changed in my life but my core idea that I want to learn everything hasn't really changed so I am going to continue posting here. 

I don't really see that many blogs anymore so perhaps this can be my little platform shouting into the wind. I do want to try and track some of what I am thinking and doing in hopes that it will be useful to others. 


I have been reading quite a bit a material on memory and how to more effectively use memory. I have been impressed fora number of years with the effectiveness of anki and skritter for learning all sorts of things(chinese specifically for skritter) and it always has impressed me that more effective and efficient learning doesn't have to be and often isn't harder. 


Many years ago as a young missionary in the MTC I wanted to learn bo-po-mo-fo(zhuyin) because we had some resources and vocabularly manuals that had zhuyin but not pinyin. (For those unware zhuyin and pinyin are ways of representating pronouciation of chinese characters, zhuyin primarily used in Taiwan and pinyin used elsewhere). At the time I believed I was very smart(since learned better) and I had always scoffed at the idea of using mnemonic devices in order to help  learn something. I thought I would just do it with raw brainpower and determination. I spent a few days forcing/beating zhuyin into my head and it worked pretty good. 

Several of the other elders were interested and decided to learn zhuyin as well and one elder in particular sat down one afternoon and came up with various hilarious and memorable mnemonics to learn the zhuyin symbols. It took that elder probably 30 minutes to learn the whole set and actually me being there helped solidify my own grasp of zhuyin. It was a stark lesson to me that I was being dumb and doing things the dumb way didn't prove I was smarter. Finding an efficient way to learn (especially if quirky or silly) is the smart way because you get to learn more things for far less effort. Elder Zappe not sure if you remember that day the same way I do but it sure taught me a lesson. 

Since then I have used anki and other tools but I do find I often scoff at mnemonics, which is a habit I still try and break. 

I ahve been reading alot over at https://mullenmemory.com/memory-palace-step-up mullen memory about memory palaces. I have been familiar with the concept for many years(remember reading on the old mentat wiki that I think doesn't exist anymore) but I haven't used the technice(oops like liek an old copy is mirrored here https://www.ludism.org/mentat/HomePage). 

There are a number of similiar systems but the core concept is doing work ahead of time to build a process so that you can easily and quickly store new things in memory in a way that aids recall. Major systems, various palace approach, peg systems they are all about doing some work ahead of time/leveraging existing memory to store new memories. 

It is interesting it sounds very much like proficiency in most tasks. As a user gains proficiency they start thinking in large chunks because they have the basic items so well understood they use them as building blocks to udnerstand and perform more complicated things. In go for example you can often seen many moves ahead because you reconize certain existing patterns that you are already quite familiar with. Trivial example is a ladder where a new player needs only a bit of experience to start to be able to see and understand basic ladders. 

To that end I want to 
1. Build a system based on the chinese radicals to help me improve my writing of chinese characters. Ideally this is used to help me transition the material into a long term faster and deeper representation. Studies show method of loci doimprove recall but they also show it is done at a cost of slower recall. If I want to read at 300 cpm I need several things to happen. First I need to be able to recognize every character I read, and second the vast majority of them need to be recognized very very fast. Fortunatly those two goals work together exceptionally well. 

I link out to an interesting page that talks about chinese reading proficiency. https://www.chinesethehardway.com/article/hsk-6-gets-you-halfway/ The core issue is that uncommon characters are very common. What I mean by that is illustrated in that post but I have a quick summary of the idea here. Last harry potter book

《哈利波特与死亡圣器》

CharactersTotalHSK 6%HSK 6
Unique3,2212,24175.16%
All307,817296,07996.19%

If you notice that on a given page hsk 6(probably roughly where I am in reading ability maybe a little lower) would allow to read 96% of the characters on an average page but only 61% of the total unique characters in the book. Those last trailing characters are super expensive to learn because they are rare but there are so many of them that every few sentences I would hit one. 

If I can get the normal characters well entrenched in memory and use a radical based memory system(that might be slower on recall) for the 5% on a page that are rare I think I can potentially really get to the point where I can read without continually missing characters. 




1 comment:

  1. And even better did some more research and of course as I can do a fun filter in anki that allows me to grab cards that have lapsed more than a specified number of times(or at all) and then I could then give those the special mnemonic memory radical system treatment.

    ReplyDelete