After reading an article this morning in the BBC about this practice I was very intrigued and immediately realized that if it worked I could use it for studying and memorization in foreign language. So I created a few experiments comparing rote memorization versus locuses already in my long-term memory and discovered the following:
Prior to trying this with actual words I did a few tests with numbers. The conclusion of these tests was that I was able to juggle 3 lists of 20 random numbers between 1 and 999 without error in their original order. That is not to say I was limited to repeating them only in order. If I did not want to include a number I simply "walked around" an item so to speak. BTW if you do not know what the hell I am talking about then I will explain in the paragraph after next.
First I took five verbs and five nouns and listed their conjugations (or declensions) and proper stresses. I then performed the Rote Test. It took me about 5m14s to feel comfortable enough to test my memory. I made 2 to 3 errors per word when I attempted to copy the list from memory. I should note that for all of these tests I purposely did not use the knowledge of conjugating and declining that I already knew. I just wrote down whether or not I honestly remembered what I had read from the list without thinking of the grammatical rules for the gender and stem endings and so on.
Second I took another five verbs and five nouns as before. This time I used the Method of Loci, attributing each word to a place I was very familiar with and its conjugation/declension to an item or thing in this place. I started small, by only trying it with one word. After 1m2s I felt comfortable with the word and tried to write it down (including stresses). I succeeded with no errors and I barely had to think about it at all. I just imagined walking through this area (one of the first apartments I lived in) and immediately recalled all of the conjugations. I was curious if I could manage two lists, and tried a noun. This time, already understanding how the process works, I was comfortable with my memory map after only 34s. I tried to write down both lists and it took only 2 or 3 seconds to make each. I proceeded to memorize the rest, and although it took me considerably longer to feel comfortable with the list (compare 10m28s to 5m14s for rote memorization), I was able to easily recall all of the tables.
Third, I went and had my twentieth cup of coffee for the day and came back when the timer on my watch was around 4m58s (eh close enough). I tested myself again and still was able to write down the full listings for all 10 words without any pausing or mistakes. I could not, however, remember more than one or two of even the infinitive/nominative forms of the words I chose for the first experiment. Of course, one cannot make any judgments about this as most of this information had been pushed out of my brain by the second experiment.
My conclusion: if you have a vocabulary test or something I highly recommend trying this approach to memorizing the words (it is probably even more efficient for memorizing just the lists of words rather than full tables of conjugations; I am sure with this method I could memorize several dozen within 10-15 minutes). As far as I have so far been able to tell it is an astonishingly efficient replacement for rote learning of boring data.
PS: It has now been 12 hours since I did this test and I still remember all of them!