Re: I count during Dual N-Back... is this wrong?

[Gore Lando]

Basically, I really, really doubt it.  We can invent lots of theoretical finaglings for why an "intuitive" strategy (quotes because I'd say the counting is more intuitive for at least a very significant portion of users) might be better, but without evidence that directions to not count were made absolutely clear to the people in the dual-n-back study who got the IQ tests improvements, and they followed those directions, I think it's silly to assume that those improvers all did not count.

On Wed, Dec 24, 2008 at 1:38 AM, Navindra Umanee <navi...@cs.mcgill.ca> wrote:

Hi,

I'm trying to figure out whether my "strategy" during the Dual N-Back
task is wrong-headed or not.

Basically, I count to N and start over during the Dual N-Back task.
I
find that this helps me a lot.  For example, during Dual 4-Back, I'd
go 1, 2, 3, 4, 1, 2, 3, 4, 1, 2, 3, 4...   Then I decide whether
number x, matches the previous number x for either audio or position.

Is it wrong to use a strategy like this?  I'm worried that I might be
using a strategy, when I should be training a different kind of
process.

Thanks!
Navin.

[dualnback]

I agree.

On Jan 1, 2:29 am, "Gore Lando" <gorela...@gmail.com> wrote:
> Basically, I really, really doubt it.  We can invent lots of theoretical
> finaglings for why an "intuitive" strategy (quotes because I'd say the
> counting is more intuitive for at least a very significant portion of users)
> might be better, but without evidence that directions to not count were made
> absolutely clear to the people in the dual-n-back study who got the IQ tests
> improvements, and they followed those directions, I think it's silly to
> assume that those improvers all did not count.
>

[Ginkgo]

I think that the main concept is to avoid using contrived strategies
that detract from the mental excersise, in an effort to score high.

If you count or mentally repeat the sequence, that is a pretty natural
reaction to taking the test.

There is nothing natural or implied about staring at a focal point and
reacting to peripheral stimuli while keeping a blank mind. This was
most certainly not how the control groups trained.

if you start using mnemonics, or the clock, or any way you may try to
exploit the test system itself... then you cheating yourself of the
full exercise, as it was meant. The only goal outlined is to
consistently train your brain with the exercise. The score or level is
relatively unimportant in the concept. Obviously as part of your
training regimen it can help to provide progressive resistance and
chart progress.

[Iron]

I think that the reason that this task works despite creating
strategies is due to the fact that you can increase n. You learn a
strategy that enables you to remember more, but no matter how powerful
the strategy there will be a limit to its efficacy. The program
elevates n to a point where you are struggling to master the exercise
and thus your working memory circuitry is being taxed again despite
the strategy. Your brain has to try to make up the difference between
where your strategy brings your ability and that 100% remembered on
whatever n level you're on. Thats why its total time spent doing the
exercise that is correlated with the IQ increase and not max n. The
program always puts you in a position of challenge where no matter
what strategy you are using your working memory circuitry has to kick
in because the program always leaves you at a position of struggle.

[Tim De Lange]

I agree with this wholeheartedly. The mental effort required to keep
up so called "contrived strategies" will eventually become so big,
that the mind will end up using the intuitive approach anyway.

It might even be beneficial to use such strategies, we just don't know.

The fact is though that our personal level of N is a personal score.
It cannot be compared to other n-back users, since our strategies
differ.

As I understand it, the theory is that n-back works by forcing you to
"re-write" bits in your short-term memory, and constantly update
larger and larger portions of your memory.

Using ANY kind of "contrived strategy", You will end up doing just
this, provided youve reached an N high enough.