Anyone have a “magic”number per lesson sesh?

Have any of y’all played around with the standard 5 lessons per session and found any optimal bunches? Is there some science behind WK’s standard of 5 per round? Hit me with your knowledge or opinions!

I like to do 10 per batch because it’s like a 1 minute SRS level that makes it more likely that you actually remember the things after you finish the lesson.


this is adjustable??

Yes, on the settings page -> Lesson batch sizing

1 Like

I use Lesson filter script to get the number I need for Radicals and Kanji but when Im only doing vocab I go with 8.

1 Like

I still have my individual lesson size sent to 5 per lesson, but I usually do 2 lessons. I like keeping it at 5 because I feel like its a good number to keep steadily moving forward but not get bogged down. Also if I typically do WK in my downtime at work, and 5 does not take very long. However, if I have the time and/or I’m not getting overwhelmed (which is typical) I’ll do 10 or sometimes even 15.


That’s pretty much what I do too, yeah

1 Like

Yeah so I concur with the 10 per sesh however you want to break it up. I do, most days, my “WANKAN20” as a morning 10 and an evening 10, except on days when I work late, then sometimes it’s only a morning 10 :/. Nice to hear from you all. Although I do like the tidiness of a script that keeps boundaries between radicals, Kanji, and Vocab…

Depends for me what I’m learning. If it’s kanji then I try to limit it to ten but will do two sets of that in the day. Vocab I find far easier so I tend to do 20 - 25 depending on how many I have. As far radicals, I don’t find them hard to remember so tend to do a session of 15 items (all radicals + some kanji).

I don’t know if Koichi meant it or not but the number 5 was actually a good choice. Their are two commonly stated numbers for how many items short term memory can hold, I might get these names wrong as it has been some time since my degree but Cowen states 7±2 items, which is the number usually stated but a more recent study by Miller(?) stated the number to be more likely 4±1. Of course things aren’t so basic that a number can just be assigned, especially when we don’t have a great grasp of the mechanisms behind memory.


I’m likin’ the science, here

This topic was automatically closed 365 days after the last reply. New replies are no longer allowed.