On Dec 6, 3:21 am, Oisín <[email protected]> wrote:
> I'd still expect spaced tests to produce significantly better results over
> a 6 week period, and of course would allow a much greater amount of
> information to be learned since it's far more time-efficient than cramming.

If you mean longer-term, durable memory, I agree. But, if you mean a
larger vocabulary in 6 weeks, I don't think I'd agree with that. Not
if we're defining SRS by its ideal of once per day, 20 minute session?

>From my own experience, I can acquire much more (shorter-term)
information through short repetitive practices. Once per day doesn't
work for me, for that purpose. But, it does work very well (and
efficient) for longer-term retention.

I also look at "time-efficient" a bit differently. You touched on it
when you said that, if someone crams, at least they're studying. To
me, SRS is efficient in the sense of "all things being equal." If we
lived in a perfect world where it was a zero-sum choice between
studying or using my time for something equally rewarding, then
cramming would be wasteful.

But, in the real world, we have changing priorities. We might have an
urgent requirement to study for an exam, and the alternative to not
studying is watching tv. We don't have an equally valuable activity to
spend the time on. It's essentially a perishable commodity. If spent
studying, at least it didn't go to as much waste as watching tv.

I know it's more complicated than that. Our brain has a gestation
period. Relaxing, watching tv is part of the process of letting the
brain do what it does. But, I'm thinking of goals that require an
accomplishment faster than we gestate memories.

To me, that gets to the constant friction between SRS ideology and the
real world. There's no denying the brain's delivery of a memory can't
be rushed, just like a real pregnancy can't be. But, things like
school classes expect us to perform faster than our brain. We only
have 4-8 weeks to master a subject which, to truly master, may take a
year.

For me (with Anki 1.2) cramming worked very well when it employed SRS-
style ordering and spacing. I might review 20 times per day, with
cards coming due in 10 minutes to 10 hours. My performance fed back
into the scheduling, influencing relative order and spacing between
cards.

That's definitely not SRS (corresponding to the forgetting curve,
which is measured in days). But, it was a million times more effective
than typical "cramming" the same material over and over again. It
leveraged the SRS's ability to focus me time on what I needed to see.
My time is finite, and I could spend as much as possible reviewing
only the cards I was having the most trouble acquiring and retaining
(short term).

It seems like it's already accepted that this is a valid tactic. For
example, Anki 2 now uses a more isolated "learning mode" with sub-day
"steps" which transitions (abruptly, IMO) to traditional SRS day-sized
intervals.

It seems like this topic is often complicated by definitions.
Efficient based upon circumstances (alternatives to spending time),
and timeframe (short-term performance or long-term mastery).

It's an interesting topic because I think everyone's describing much
the same thing, just from different perspectives.

-- 
You received this message because you are subscribed to the Google Groups 
"mnemosyne-proj-users" group.
To post to this group, send email to [email protected].
To unsubscribe from this group, send email to 
[email protected].
For more options, visit https://groups.google.com/groups/opt_out.


Reply via email to