Sure. (:

Well, I guess there are a couple of potential questions.

There's maybe a more difficult question, which I think is what you're
getting at, which is maybe trying to infer the structure that is in a
body of knowledge. Like, we are handed a bunch of flash cards and
learning history, as from Mnemosyne data, and we try to infer which
facts are related to which. Clues would be along the lines of figuring
out when presentation of one card made recall for a subsequent card
easier than the program expected. It would take tons of data, as you
suggest.

But, i was thinking about trying to understand dynamics of learning
when we already know a structure. An instructor might, for example
know which recall tasks compose other recall tasks, which recall tasks
are composed of which other recall tasks, which recall tasks are
equivalent to one another, and which recall tasks are unrelated to
each other.

Taking that structure as a given, if you had a large enough pool of
data, you could start attributing "prediction error" in mnemosyne to
the presentation of cards that are related: the algorithms behind
mnemosyne could be used to predict an expected self-reported
difficulty for each card presentation. Sometimes this will be in
error. Enough data, and we can infer the presentation of which cards
are responsible for some of this error.

--Joe

On Nov 23, 4:52 pm, George Wade <[email protected]> wrote:
> Several comments, Joseph,
>
> I can't imagine the scientific method being stretched, easily, to cover
> your interest.  As soon as we add layers to the system it becomes too
> complex to attach numbers to;  or the number of combinations becomes so
> large that it would take millions of years to complete the experiments:
> in which time we would have become extinct.  Genetics is an area with
> thousands of variables.  The problem was managed by subtracting one gene
> at a time from complete cells to compare properties.
>
> Can you see a possibility for experimenting along those lines ?  Or of
> making a software model that can be run many times.
>
> The ancient Chinese used repetitive contemplation for ordering their
> lives.  That was and is a system.
>
> Mind Mapping follows structure.  You could ask people in the field.  You
> could just try Googling the question to see if anybody understands it.
> A good question.
>
> On 21/11/2011 17:46, Joseph wrote:
>
> > I'm curious, is there any research on spaced repetition within systems
> > of knowledge? I know the classical research on memory for isolated
> > facts/associations, showing the exponential decay of likelihood of
> > remembering, becoming shallower with each recall. But a lot of our
> > knowledge has some structure, like hierarchical sub-steps of steps of
> > larger processes, complementary information.
>
> > Is there any research on how recalling one item in these structures
> > affect likelihood of recall for others?

-- 
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 this group at 
http://groups.google.com/group/mnemosyne-proj-users?hl=en.

Reply via email to