28^th FEBRUARY 2011

An improved version of the LTVM scheme

Almost three months passed since I devised the Long Term Vocabulary Management Scheme (see 101130 blog entry). As you can see at the bottom of the left-hand side column, I managed to work through 127 batches of (on average) 10 vocabulary items each. Note that most of these 127 batches were repetitions of already aquired material, though. The material memorised amounts for a total of 3 entire "generations", each comprised of approximately 100 vocabulary entries.

The good news first: There is absolutely no denying that the long-term memorisation of vocabulary works reasonably well. In fact, I estimate the number of vocabulary entries that did not immediately came to mind during repetition sessions at less than 2 percent - the majority of errors that lead to an accuracy rate of only 90% are spelling errors (ommission of diacritics and/or single letters). Even over the long run, memory decay is almost non-existant, meaning that material that was correctly aquired could be retrieved from memory flawlessly and (almost) effortlessly.

There are downsides, though. I assumed that vocabulary entries that were not correctly memorised the first time upon encountering could eventually - due to the upcoming repetitions - memorised correctly. Unfortunately, that does not seem to hold true - well, at the very least not in my case. It turns out that only vocabulary entries that I went to great length to aquire correctly from the beginning would be memorised correctly through the long-term repetition scheme indicating that it is imperative to put in great effort in the first memorisation before actually entering the repetition cycle. And, of course, the scheme puts on a very demanding workload.

Undoubtedly, the biggest challenge is, however, to find the time and proper motivation to follow through with the scheme on every single day.

LTVM v2.0

With this first experience in mind, I devised an improved memorisation scheme that adresses the problems encountered. The basic design for repeting a single batch of vocabulary entries looks as follows.

This new batch design puts more emphasis on correct memorisation from the start, but increases the time stretches between consecutive repetitions.

From this we build the memorisation and repetition scheme for an entire generation, worth 100 vocabulary entries.

Unlike the first version of LTVM, I now abandon the strategy of repeating every batch exclusively in written form. Under the new scheme, only the first of every two consecutive repetitions will be in written form - the second repetition might be done orally, saving me a lot of time while still enforcing memorisation by repetition.

Finally, three generations are interlocked into a new "grand scheme", taking advantage of the vast empty time strechtes between consecutive repetitions.

LTVM v1.0 alotted 47 days for long-term aquisition of 3 generations, meaning 300 vocabulary entries. LTVM v2.0 takes 61 days to aquire 3 generations, meaning that six complete cycles of three generations each could be completed within one year. Having in mind the previous experience with LTVM v1.0, I will consider myself very lucky if I manage to work through 4 complete cycles within the next year, completing long-term memorisation of 1200 vocabulary entries.